87 951 VEMS Install
 

As requested my VEMS install thread.

Background: at a WSIR track day before last thanksgiving going up the hill in turn 3 my car started feeling like it was out of gas, loss of power etc. Limped back to pit and put some 91 VR gas in. Problem seemed to go away, but same thing next session. Thought it was bad gas maybe, limped to chevron and filled up. Didn't make it home, could get power sometimes, and other times would fall flat and miss. Didn't think I could make it over the mtns that way, so had some drinks at the Palmdale outback waiting for the tow truck.

Going on the bad gas idea I changed everything in the fuel system, eventually flushing tank, all filters, lines, new pump already, FPR and Damper. I had recently replaced every hose and vacuum line, and installed ported K27/6. Replaced S and R sensors (that was fun), tried known good DME, nothing cured the problem, checked all sensors for resistance and function.

Took it to a dyno to see if we could replicate problem on dyno, watched fuel pressure during pulls, it was good, first run 340whp, second run went lean and backed out, 3rd run 353whp. This confused me more, but also assured me it wasn't mechanical.

Also checked compression and TDC cam alignment. At wits end.

A I met some fellow rennlisters to try out the DME, and VEMS came up. Heard good things. We decided to do a group buy.

I figured the DME wasn't telling what was going on, and I was tired of new chips every time I added something. Knowing VEMS had trigger logs, regular logs and a way to view everything that's going on it was a good restart on the problem.

I had been talking with Raceboy about it for a while and set it up, and waited for my box.

I merrily tore out my AFM and distributor, and fabricated a turbo intake tube with AOS, vapor purge and BOV ports, and welded in a bung for temp sensor right uphill of the throttle body, and had them powder coated.

My box came, but the 80# Injectors were accidentally included in another kit so I installed everything else while I waited for them.
https://cimg6.ibsrv.net/gimg/rennlis...448e3eb373.jpg


https://cimg8.ibsrv.net/gimg/rennlis...fc19e00abe.jpg

https://cimg9.ibsrv.net/gimg/rennlis...9c94ff6c09.jpg

https://cimg0.ibsrv.net/gimg/rennlis...215227cea7.jpg

https://cimg1.ibsrv.net/gimg/rennlis...a967761c3d.jpg

My disto is back on to cover the hole until I machine out some clearance on the prototype block off plate from 3D printed Porsche parts.

So when the injectors arrived I was ready to go. Put them in, turned it over without DME relay to build oil pressure since it'd been months, and got ready for the glorious first start.

And then ran my battery down cranking. Never figured it out with Raceboys help that weekend, could get it to catch but never run. Next weekend tried it again, and with some config tweaks got it to fire up. I idled ok and I drove it around, but it felt like it was missing power. Also, it took starting spray to help it to life. Still not right.

With various checks, tweaks and changes suggested by Peep, today is got it running right. Starts up quick, pulls strong, auto tune works a charm for getting it right. It took checking my s and r sensors many times, reversing polarity on the ref sensor and changing it back, tracing pins back to vems to double and triple check. Eventually we tried changing something in the trigger ref table, and that started it up, but it ran like a tractor and wouldn't rev. Used timing strobe and ignition lock to get to TDC with reversed ref polarity (could be cheap facet sensors, Bosch ones are in mail), and then it ran but still had trigger errors. Switched polarity back to normal, reset TDC after trigger with light, and no more trigger errors, clean triggers and an engine that fires right up.

Drove around South Bay auto tuning, and with my MBC in place still (haven't implemented EBC yet), I get 17psi by 3200, and it pulls well to my 6599 redline, although the boost does start to drop off by then.

https://cimg9.ibsrv.net/gimg/rennlis...9e1d3c8419.jpg

https://cimg1.ibsrv.net/gimg/rennlis...f4ddcceaa4.jpg

I still have more tweaking and tuning and learning to do, and setting up EBC with Peep, but even though in my case there were some weird things to figure out, it has solved my initial problem, and looks to be the start of having my car be function fast and fun again. Thanks Raceboy, who is very gracious with how many questions I ask him via FB messenger a day.

Any questions don't hesitate to ask, with the age of DMEs and AFMs I think this is a great option to keep the character but step up management into a more modern realm.

Sam

Hell yeah. Standalone is the answer for our cars, for both performance as well as reliability and diagnostics. My NA has been nothing but flawless since installing MicroSquirt and my 951 is getting something even better.

Glad we got it sorted, there were issues that I have not enountered ever before but it did not even take a year to sort out :P :D

Now there's just a boost solenoid to tune when I come over there soon!

*totally unrelated to VEMS*
I also had trouble with the cheaper speed and reference sensors that I ordered from Pelican Parts. I ONLY replaced the sensors in the first place because I had to smash the original ones when I did a clutch install. Car presented with no start a couple days after the clutch install. I assumed it was the S/R wiring harness, so i replaced it with lindsey racing. no change. Eventually I got it running by switching the speed and reference sensor position on the bell housing.... just the physical sensor location, the wiring had been correct. This made no sense so I ordered the Bosch sensors. Continued using the FAE ones while I waited, and even though it would run for the most part, there were some random stalling issues. Recently put the Bosch ones in, and no trouble since.

I run a single sensor on the front of the crank with an aftermarket trigger wheel. The sensor is $30 new and provides a digital signal... no intermediate analog/digital conversion required, which causes the most common issues with aftermarket ECU setup. Obviously the VEMS unit works fine with the stock ones out of the box. But I'll take a single, cheap, and easy to access sensor any day.

Very interested to see how much power is made going to VEMS. What was your engine management previously?Mods? Have similar mods myself. Also, is it possible to use aftermarket crank and trigger wheel setup on VEMS?

Of course it is possible but why bother if stock trigger arrangement works well? BTW there's one 944 on Hawaii that runs with VEMS PnP :)

Vems is in the short list of addons for my 951. It appears that the install is much more complicated than I thought. Is there an installer in the Northeast who I can contact?

The install is not hard, if I can do it - anyone can do it. Mine fired up first try and the sensor calibration is simple.

Ill ask a general question as I know Peep prob gets too many emails - I drove mine around and "auto-tuned" and then couldn't figure out how to burn config to ECU haha, how do you do that? Mine actually seemed to drive pretty damn well out of the box so there must be some kind of 951 based tune installed. I also need to setup the round gauge - gotta do some more reading on that.

I blew an intercooler hose off during my first drive while logging/auto-tuning so no big deal I couldn't save it - Ill just do it again, going to throw new plugs in it before I do that as mine seems to have a small miss at idle / kind of a tiny backfire or pop...once rolling seems to clear out and pulls well.

What's the current delivery time for the Vems? I could ask directly from Peep, but i'm sure he is busy enough making them and and answering more important questions.

I just started my car for the first time with VEMS after the winter storage. It was a nice surprise that then car fired up on the first crank after sitting for 5 months. The pre-crank prime cycle for the fuel pump makes a big difference in starting the car. Previous years would involve cranking and cranking and cranking.

The install is not difficult, best purchase I've made for my car. I bought the solenoid from Peep to integrate boost control into VEMS but I also have a Turbosmart EBC sitting on the shelf, I'm not sure which would be easier to make work.

Sam, you can easily email me if you have any questions! :) I always find time to answer.
But you burn the config by just going to any menu or dialog where BURN button is present (it turns red with autotune active and making changes).

Congrats!!
Interested in this soon once I get a few other projects out of the way.

Standalone is def the way to go. It's not hard to install if you just do the straight plug and play, wasted spark makes it a little harder but not much, you just have to mount a new coil and figure out distributor block off.

You can just get an old MAF and hollow it out for pre turbo intake, but I wanted to make something. Or hollow out AFM for stock look, which is on my list of things to do around November of this year.

Power wise it dynoed previously at 353WHP on a mustang dyno, so I can't imagine it won't go up a little. I'm not running higher boost, but it does breath a lot better, it hits 17psi at 3000rpm in a 4th low rpm pull.
https://cimg4.ibsrv.net/gimg/rennlis...ec0bb4b745.jpg

And yes SamGrant951, Peep is very good with questions, even the hundreds of dumb ones I've asked him!

Thanks Peep. Ive been in contact with him and we're on the same side of the Island also! Just in case old sensors were worn, it would be good to have an option to run an aftermarket setup. This will definitely be my next mod. Thanks for posting this thread.

Oh and SamGrant951 it's 'ALT B'

Awesome, thanks!

re: making a pre-turbo intake / using an old MAF, I had Tim @ SFR make this one for me.

https://c1.staticflickr.com/3/2851/3...82b56008_c.jpg

Blow me away

Sam that's a lot prettier than mine!

November = smog test? Has anyone gotten one of these through Ca dyno-smog yet? In principle it could be better, but Ca can be difficult. It may help that most folks have no idea what these look like under the hood, but the coil pack and K&N could trip them up.

what are the costs for these setups?

Gary

Contact Raceboy - I believe he has a stealth setup as well, where you can hollow out the stock AFM / Barn door.

MAP sensor location?
 

Sam and Sam
Where did you put the MAP sensor? I think I see a longer banjo bolt on Sam-LA's manifold. Is the MAP hanging from a short bit of hose from that? Is there a max length of hose to avoid delays in response?

thanks
Gary

MAP sensor is within the VEMS ECU, just run the KLR line already in the cabin/footwell that went to the KLR to the VEMS box. IAT sensor goes in the intercooler pipe (pictured above) which has a provided plug in harness that goes to the old AFM plug.

I used the 4 port banjo from LR when I redid my vacuum system, to give the cleanest signal to as much as possible direct from manifold without many Ts. Line to VEMS is direct from that. I also eliminated the T from the brake vacuum line and ran that from one of the ports.

I don't remember how the 951 is plumbed but I put my MAP sensor line on a Y right at the throttle body, the same port that feeds the FPR/FPD.

thanks
 

Sam and Sam ... thanks, I hadn't realized it was on-board. Sounds like is trivial to move from KLR to VEMS input.

thanks
Gary

Hey Sam, quick question, what hole did you use to run the plug connector for the wideband O2? Is there one big enough on the firewall?

I ran ran mine through the shifter foam, as the lead wasn't long enough to go through the firewall. It's still a bit short so I may lengthen it. There's only so much tail on the Bosch sensor side, and I wanted to get that connection inside, but on the VEMS side I might make it a little longer to accommodate my interior routing under the radio.

Ah! There's an idea! I did buy a cable lengthen the O2 sensor connector about 5 feet so this may work perfectly! Also how far after the turbo did you install the sensor? Is a foot after too close?

I just went off of the LR pics, right where the main exhaust bends toward the WG dump installed bung there 45 deg upwards of level.

I ran mine through the factory wiring harness rubber plug on the firewall. I didnt have much luck de-pinning the WB02 plug (its quite large to try and get through the firewall) so I just cut it and re-pinned with a new connector I bought after I pulled the wire through. I had plenty wire to route the cable.

Okay, so at least 2 good options here, thanks guys. I do have a bung about a foot after the turbo for an O2 sensor and I already have a wideband (aem kit) on the main pipe bend under the car. I guess if the sensor doesn't last long closer to the turbo it'll get moved. I just need to get this car up and running, it's so rich right now with the maf kit.

So what are you guys using to run the VEMs? I did get the Bluetooth for smart phone connectivity but thinking I may want to get a small notepad or laptop dedicated to the car.

I bought an Lenovo thinkpad X220 12.5", came with windows 7 pro installed, only cost about $150 on eBay, has bluetooth. Works perfect for this application, Peep recommended it. I also picked up an 8" Asus zenpad to monitor everything without laptop.

i just bought a bluetooth USB dongle for my small laptop that i use anyways.

Ok, got mine up and running today, it idles but surges pretty bad. Just up and down, up and down. It does staet way better so far. Where on the VEMs site is the app you download? I see they have nightly releases and stable releases...What's the difference?

Figured it out now I just need the pin. Nevermind, had an assist...

Thanks Sam for the help!:cheers:

Nice, standalone is great, i just got my 951 started yesterday.

But what ignition tables do you run ?

I just run Raceboy's spark tables he sent with it, haven't tweaked them yet, but will when he is in LA tuning.

I used the latest stable version, haven't tried the nightly versions.

Re: idle surge, do you have an IAC hooked up and you're sure there's no air leaks? Could be a few things but working through all the idle options you might discover something. Any trigger errors?

Yes, still have the iac hooked up, was running fine on m-tune, no changes other than some plug and play stuff with the VEMs. It starts out smooth then about 30 seconds into startup is when the idles starts to surge.

Can you make changes with the Bluetooth or do you have to actually connect into it with the cable?

What Does your charge table And idle controll table Looks like :)

You can do everything with the Bluetooth laptop connection.

Ok cool. Talked with Peep, gotta reverse the polarity in VEMs for the IAC valve, should fix the idle surge.

Have you checked your IAC for vac leaks. Just curious.
I posted a thread on here about doing the seal on it.

I just got my new seal in the mail.

Huge difference with the new seal 3mmx33mm

Leaks with old one and no leak with new seal.

what seal did you replace? All the hoses are a few years old, even the mounts are new(er)

This is the seal for the body of the valve?

The internal seal. If you plug one end and blow on the other end it will give you an indication of how good the seal is.

Massive difference with the new seal in on mine compared to the old one.

My idle on my car was bouncing all over the place with the M-tune due to I am sure my IAC. My engine is out of the car at the moment but I am sure it will run much better when I get it back together.

https://rennlist.com/forums/944-turb...tabilizer.html

ok, thanks for the heads up!

I don't have AC and I got tired of leaking IACs so I'm running without one now. As long as I give it a little throttle on first start of the day for a minute or so it does fine. Stupid leaky IACs.

It is very easy to fix them. I was tired of the same sort of issues. Then I decided F@#$ it and took it apart. Wished I would have done that ages ago. If one has any simple mechanical skills it can easily be done.

Yeah I fixed one, new o ring and epoxied it shut, then it blew open after, and I grew tired of removing my manifold to mess with it so I just left it out. If I lived in a colder place I might worry about it more.

If I ever decide to implement some anti lag I'll need it back though. But that will probably be right before I plan on replacing the turbo, crossover and header.

Injector size
 

Sam, you are using the 80 lb injectors as well so then under engine setup it should be an 841cc injector size right? Mine days 380cc. That may be part of my problem with the idle because changing polarity of my iac didn't fix the problem.

markl951 posted this pic for the wiring for reference.


:thumbup:

in VEMS the "injector size" box is just for your own reference - changing it doesn't actually change anything about the tune.

you need to tweak the "reqfuel" depending on injector size.

Thanks rocket.
 

Clears things up some. Looks like I'm sitting right @ 5.0ms, sounds about right with e10(93). I just can't wrap my head around why my idle is ****. I have logs if anyone would like to see them. Totally stumped...

small changes make a big difference, and you need to adjust the VE table as well.
fwiw my setting for NA injectors is 12ms.
turbo injectors are 50-60% larger than NA so you might need to bump up to 7 or 8ms.

dont bother thinking about fuel ethanol content with reqfuel - get it running smooth and let the VE auto tune figure that out.

Really, that much? I did the calculation to get the number and it's actually 4.8. you​ running 12 on an n/a is very interesting. I will try bumping mine up and see what happens.

Req fuel is good to have set correctly but ultimately your VE table and enrichment curves are what determines your fueling. They're all scaling your req fuel, so either way it'll be ok.

On my NA, best idle is around 13-14:1 AFR and it likes about 15-20* of ignition advance. Can you check with a timing light to see if you're actually getting your base timing? All you need is a basic one, the rest is setting up fixed timing temporarily in the tuning software.

My timing keeps jumping up and down from 8 to 14 degrees with the idle, cant get it any closer than that. The only thing I have seen that makes it kinda smooth out is by running it pig-rich (.65-.70) and I know that's not right...

Making the req fuel higher changed the frequency on the surge, made it faster.

What's your set up? I can look at my config. Where are you located?

Setup is the remnants of my rogue kit, 80 lb injectors and air filter. I have a t04e turbo, about a 60 trim, stage 5 hot side. Exhaust is 3 in downpipe and mid pipe then goes to 4. Stock internals. Til 38mm wastegate with MBC.

I am in the bottom of PA and top of MD, Cumberland area.

i did the calculation (that equation somewhere) too and the number just didn't work for me.
i ended up at 12 based on advice from Peep and then my own tuning experiments.

try this..
go to the starting/idle tab and go to Idle Control, Spark cut based. disable that.
then go to the starting/idle tab and go to Idle Control, Ign. advance, enable it, and set the max advance and max retard to 4 degrees.

you might also try simple adjusting your target idle rpm up from the stock 850 or whatever it is to 1000 (idle control general)

mine also likes to idle at 20 degrees target and i have it at about .97-.99 lambda.

Will do, I'll give it a go. Hope for something because at this point anything would be better. I will try tonight if I can but if not probably tomorrow then.

Hopefully my email from Peep will come back with some more advice...stay tuned.

Your timing is jumping around at idle because the load and RPM is fluctuating due to the surge. Try this... in your ignition map, find the column that's below your target idle. If you don't have a column that low, that's a contributing issue, create the column and spread out the others accordingly. Now, make that column BELOW your target something like 20* advance. Go to the next column, which should be around your target idle RPM or slightly higher, and make it slightly lower...say, 15-19*.

What happens is that when your revs drop low, the ECU adds timing and it comes back up. Think of it like balancing a marble on an upside down bowl. It tends to roll off the sides. But when you have the marble inside the bowl, it naturally rolls toward the center (target). Not a great analogy but you get the idea - you can create a "valley" in your timing table where the engine will settle into a target area. If it goes too low it will increase cylinder pressure and pop back up and you won't even notice. On the high end you're limited by air flow bypassing the throttle plate.

if your system has injector data / setup, and you ignore it and start filling in values in the VE table, you're making your life harder than it needs to be

Why dont you just try "autotune"
And on low revs no boost make afr target 14.7.

The timing trick sounds like it should do what I need, I didn't think of that. I will see what happens tonight, ran out of time on mothers day.

Good info guys, I really appreciate the help and I am still learning a lot here. There is always more to learn...

Autotune made it run waaaay worse, to the point of leaning out off the map and stalling. No bueno.

It's also possible that your wideband isn't conveying a realistic lamba signal due to ground interference (auto tune references commanded versus measured AFR). It's my understanding that VEMS has it built-in...is that the case here or are you using an existing controller via a signal input pin?

yes, using the integrated 4.2LSU style setup. I did the open air calibration to the sensor as well. Its now spot on. I still have my AEM wideband hooked up too so I can always monitor what comes through the exhaust.

Both are telling me the same signals however one is lambda, the other air/fuel. They both go rich and they both go lean together.

you may know this (so this is a refresher :) ) but...
autotune sniffs the exhaust and adjusts the VE table to bring actual lambda inline with your lambda table target.

you said it runs well at like .7 lambda which is stupid rich. did you get it to run at that value by tweaking the lambda table or the VE table?

did you adjust your reqfuel yet? might want to start with a higher number (maybe 10 ms) and work your way down. larger injectors need smaller reqfuel ("on time") for a given amount of fuel required for idle/running.

not to say my tune is perfectly set up by any means but it runs darn well. if my 240cc or whatever NA injectors run great at 12ms and you were running at 5 or 6 that would logically match to a 480cc injector or so, right? since you have 380cc you need to UP the reqfuel time and "start over" with the VE table and auto tune.

My req fuel was like 5.2 - 5.4 from memory, running 80# injectors.

Thanks guys, made some progress with it last night, it likes the timing bumped way up (around 18-20) at idle and I messed with reqfuel again. I think I just need more time as I only got to play with it for about 45 minutes yesterday...but stepping in the right direction finally.

Thanks for the help so far!

Not for a 951 but I have a Porsche 944 s2 where did you buy the VEMSvems from online? also, did you need anything additional for the install? I want to break down and do it but I don't know where to start

Email peep@vems.ee

He'll get you sorted with the kit. For S2 you'll need to weld in WB O2 bung in exhaust I think, but that should be it for basic kit.

Thank you just emailed him! was it pretty easy to install? will he know what Ignition Drivers, Primary & Secondary Triggers, MAP Sensor, EGT, and Knock Input i'll need?? I'm mechanically inclined but have never messed with ECU stuff

You won't have to do anything except "plug it in"...Peep takes care of everything else for you.

I've been reading all these threads with interest - I have a VEMS system from Peep sat ready to install.

I'm pretty happy with knowing what I need to do to physically install the system but a lot of the tuning side seems like a steep learning curve. I've currently got an Apexi AVC-R EBC installed and will leave that in place for now, I have got a solenoid from Peep to move to a VEMS-based boost control in the future but not sure where on earth that wires in???

Seems like some good knowledge and experience in this thread so I'm hoping that I'll be able to contribute as well as learn.

That solenoid goes on the end of the stock cycling valve wire, replacing that connector with the one for the solenoid. With auto tune it's not bad to figure out, although some of the finer points such as acceleration enrichment take some fiddling. It helps that Peep came to LA to help my tune and EBC!

Anyone setup traction control yet with the VEMS?

When are you going to install it.

Meanwhile in Red Rock Canyon near Las Vegas :)


https://cimg3.ibsrv.net/gimg/rennlis...3c88837a85.jpg

This weekend hopefully. Should be able to get everything physically in and calibrated but I'm waiting for the Inlet Air Temperature sensor to turn up before I can start the engine.

NCLA951 - that's useful thanks, I'd be interested to see what your boost table looks like at some point? Although ive got a good understanding of duty cycles, etc. from tuning my Apexi.

I finally got my VEMS system installed a couple of weeks ago, and I am impressed so far. I have a custom Evergreen K27/K26 billet hybrid turbo on the shelf, along with a spare engine to build up to cope with long periods of ~1 bar boost on track, but I decided to try the VEMS on the tired original engine (100% stock) to start learning the system.

The main goals were to keep the engine bay as factory-appearing as possible, and to ensure that any modifications were 100% bolt-on and bolt-off (thanks to the draconian "visual" requirements of the CA smog test).

These requirements neatly dovetailed with my desire for a true "cold air" intake, which is fulfilled by the factory airbox. A huge pet peeve of mine is a stupid cone air filter sitting in a hot engine engine bay and billed as a "cold air intake." Of course, in a forced-induction application the air is compressed and heated and then cooled down somewhat by the intercooler, but ensuring that the incoming air is at ambient temperature still helps.

To this end, I pillaged an adapter flange from a 1980s BMW AFM and mated it to a 3" silicone hose section and adapter pipe purchased from Pegasus. A friend TIG-welded an aluminum bung to a spare factory intake pipe to mount the IAT/MAT sensor as close to the throttle body as possible while ensuring minimal heat-soak for accurate temperature readings. The AFM adapter and the airbox itself received some love from a die grinder to eliminate the small rectangular AFM slot and allow a true 3" inlet from the airbox (at least until the J-boot necks down to 2" for the factory turbo inlet; my upgraded turbo has a 2.75" inlet).

https://cimg5.ibsrv.net/gimg/rennlis...8cf7dbf10e.jpg


https://cimg6.ibsrv.net/gimg/rennlis...1d2d43cd89.jpg

After installing everything, the engine started right up with the base map that Peep installed, but it ran quite lean. I started by manually tuning the VE table in the idle/low load range to get a feel for how the system responded to changes, and got it running fairly well. Then, I set my AFR/lambda requirements and basic timing map and let the auto-tune feature do its job for the VE table, and it is running great after a number of tuning drives. The throttle response is definitely better than stock, as is the midrange torque.

I also installed a Tial F38 wastegate at the same time, along with a MAC boost control solenoid. I chose a 0.7 bar spring for the WG, and started playing with the electronic boost control feature of the VEMS, but I noticed that even at 0.7 bar, the factory injectors were exceeding 90% duty cycle at higher RPM. I installed a 3 bar Bosch fuel pressure regulator and re-tuned the VE table, but the injector duty cycles still approach 90% at high load levels. So, I will wait until I can install a Bosch 044 fuel pump and larger injectors before tuning for the bigger turbo and higher boost levels!

Overall, I am quite pleased with the VEMS system, and Peep has been extremely helpful and responsive to any questions that I have had.

Anyone got a base config file they could share? I didn't get one in my kit, so I believe that's the reason why I'm having tuning issues. Emailed Peep currently waiting on response. But we're like 18hrs difference in time. Much Appreciated - Wayne

​​​​

I never send out a unit without a config!

Also sent you instructions how to proceed, please follow those precisely.

Didnt realize config was pre-burned into ECU. Sorry for the confusion. I also want to compare setups with everyone else.

What is the problem with your current setup?
Ign table? Fuel table? Sensor setup?

I was having high idle and stumbling issues, even after getting req fuel lamda set and using autotune. Deleted the idle control valve, adjusted the hard stop and tps to open a bit more and I have rock solid idle at 900rpm at operating temp, but it also idles well at 500rpm. req fuel is 8ms, ignition retard for idle is 15. Fine tuning at the moment and will probably add boost control and launch control soon. Currently using at HKS EVC-S boost controller. Keep the tuning tips, issues and solutions coming. We need to make a dedicated VEMS thread. -Wayne

It seems like this is the thread!

I just installed my Evergreen K27/26-6 hybrid turbo with SFR 3" downpipe and center pipe along with Siemens Deka 80-lb injectors and Bosch 044 fuel pump, and the change was seamless with the VEMS. A combination of manual and auto tuning of the VE table and the engine runs perfectly throughout the rev range.

Next up, working on the PID boost control and installing the waster spark setup along with a cam trigger to go to full sequential injection!

Awesome Chris and I think this thread has alot of great information. Please share install of cam trigger and sequential injection, was looking at this in the future. I can share instructions for wasted spark. How many pins does your coil pack have?

I got the coil pack from Peep, it has 3 pins, same as the one pictured on the left here:

http://www.vems.com/ignition/bosch-motorsport-2x2.html

As for the cam trigger, I had been planning to adapt the Hall sensor and cam gear setup from my old 944S project car, but this is so much quicker:

http://www.clewett.com/index.php?mai...roducts_id=306

3 pins like mine. I have an MSD unit. center pin goes to 12v coil / red wire from factory coil. 2 otter pins goto ec36 pins 34 and 36. then put these settings. base setup>ignition outputs visual>https://cimg2.ibsrv.net/gimg/rennlis...b78ac0413f.png

Great thread and very helpful.
I ordered VEMS for my 88 951s and am looking forward to the upgrade. And thanks to Chris who shared notes on the stock airbox adaptor as I plan to follow that path as well.

Great thread! I ordered a VEMS PnP for my E Production race car. It's a '87 N/A that has been using the stock DME for three years. I'm making some big changes to the motor that will require much more tuning. The only thing I've ever tuned before this was a guitar so I'm hoping it's as easy as it sounds. Can't wait!

Just received my goodies from Peep. I have a friend tig welding the air temp sensor to the turbo intake tube similar to the pics in above, and I will be routing the wideband next.

This thread has been super helpful and thanks to all who have shared their experiences. Even though the VEMs is plug and play there is still a bit of planning with respect to routing the wideband wires other details.

Has anyone installed a MAC boost control with their VEMS? I had Peep add that but not much in terms of instructions there. I plan to make that a "day two" project to replace my manual boost control (38mm Tial wastegate) once I get the car running with VEMS. Ditto for teh oil temp sensor which requires a dedicated oil pan bung or drilling and tapping the oil drain plug. Lindsey sells a tapped oil drain plug bit looks like it is threaded differently from the one Peep provided.

Another day two project.

You’re not putting the temp sensor Pre Turbo are you? It wants to be near as is can be to the throttle body, Post turbo and intercooler.

No - bad wording on my part. I had the bung welded into the hard pipe from the intercooler to the throttle body just as you did. Just got it back.



https://cimg4.ibsrv.net/gimg/rennlis...ccad064b62.jpg

Next step is to install the wideband o2 sensor. I was considering something similar to this pic from Lindsey, but on the outer side of the downpipe as it is easier to get to and weld. Then route the lead up behind the engine to mate with the VEMs harness, which I was planning on depinning so I could get it through the passenger side firewall grommet.

Wondering if that sound reasonable?

Thanks!



https://images.duckduckgo.com/iu/?u=...cation.jpg&f=1

You want the Wideband O2 sensor pointed at a 45 degree angle pointing down to keep moisture off it. To prevent premature failure. Its tight, but many have done it by loosening the bolts at down pipe for extra clearance.

Having a weird issue since installing VEMS and Peep and I cant see to figure it out. Im getting some low idle/low RPM stumbling issues. But it happens every so often. For example, from a cold start my RPM is around 500rpm ( i have no Idle Control Valve ), once it comes to temp, after warm up enrichment, i still have low idle at around 700rpm, stumbling and sputtering at 1100-1200rpm and lots of popping on de-acceleration. But if i shut off the car and immediately turn it on, Idle changes to 1000rpm, no stumbling, no sputtering and no popping under de-acceleration. Its weird. Tried many different config files, had Peep adjust idle etc. Tried auto-tune with EGO closed loop off. I can time it to where every 4 restarts , I get a good idle/start. I usually have a bad start, but after shutting off and turning on its good again. Been thinking TPS, calibrated multiple times, still no help. I have a ford throttle body and tps coming in, so will see if that helps. Maybe speed/reference sensors? This happens cold OR hot start. Other than that issue, care runs perfectly and gas mileage has never been better.

Have you looked at the charge temp table? that corrects for cold / hot engine ?

Could be a problem with your Afterstart settings or RPM based overrun fuelcut or VE table or ignition timing table or MAT/TPS fuel enrichment table or.... I would say, hard to make a diagnosis.
Maybe one could help, if you post your config file.

Hi All,

attached file compressed via RAR.

This issue can be on cold OR hot start. But it seems, if I turn the ignition to on without starting then off 3 times, on the 4th i start, I get a good start. weird. I dont know if theres a setting in VEMS where a condition like that would exist. But take a look. Doesnt get too cold here in Hawaii, maybe 55F early morning and high 90F in afternoon. Which is why warm up and start up looks the way it does. One thing I noticed is EGO is always off and in visual graph it shows EGO 100%. I dont fully understand all features yet, so please excuse nooby questions. Thanks in advance -Wayne

my first thought is that you're not getting enough fuel for the first few tries, but by the 4th time there is enough residual fuel on the port walls that it can now start happily.
cold start needs more fuel, but some of that fuel will just settle on the walls of the intake port rather than go into the engine (cold fuel doesn't evaporate very well) so you need more fuel to compensate for that.

i thought that too but that would only happen during start up. After warming up to temp, it should run normal. the problem im having is from start up, until i shut off the engine. doesnt matter if its a 5min from or 45min drive.

BTW, has anyone with VEMS wired in an aftermarket TPS sensor? Im doing the 65mm FORD TB mod and will wire in a new TPS 3 wire sensor. The factory plug is 5 wires. Looking at the VEMS diagram, only 3 inputs are needed. Would like to just splice into factory harness instead of making another one to the VEMS ecu.


https://cimg2.ibsrv.net/gimg/rennlis...754428f331.png
https://cimg4.ibsrv.net/gimg/rennlis...391fd8d364.jpg

Any startup, either hot or cold, needs extra enrichment during cranking. Is there a table to set cranking enrichment?

Yes there is Dave. Will check that out. Thanks. But like I said previously, the stumbling and low idle will stay until I shut down and start up the car again. So even after driving for a long time the issue is still there.

Why don't you have an idle control valve? This sounds like a street car, so it would be a very nice thing to have! Otherwise your idle speed will vary depending on engine temperature, accessory load, etc. Plus an ICV will allow a higher idle during the warmup enrichment period, etc . . . . .

While ignition timing and fuel mixture certainly have an affect on idle speed, the main determinant is the amount of air entering the engine. Without an ICV, the throttle body bypass screw is the only thing that controls this when the throttle plate is all the way closed. Have you tried playing around with the mechanical idle speed setting? Are you sure the O-ring on the bypass screw is OK? Any other vacuum/boost leaks? You need to make sure that the mechanical components are OK before getting too deep into the ECU side of things.

All of that said, have you looked at datalogs when the symptom is/isn't occurring? There is a lot of data available with the VEMS or most other standalone ECUs, so a close look at that might reveal your answer . . . .

Haven't had a idle control valve for a few years prior to installing vems and was still running a maf and piggyback. Yes it's a street car without Ac.. Car ran fine on vitesse maf and smt6, having a hard time believing that I would need an icv with vems. There's also a few individuals I know who have vems without an icv without my issues. I've logged both conditions, good run and bad run and given them to Peep. However it hasn't shown an answer that he or I can figure out which is why I'm reaching out here. I also thought that it might be the solid state DME relay I have from F9 but it showed issues in 3 spare relays. Beginning to think it maybe the speed/reference sensors. Gonna grab an o-scope from work and check out the signal. Thanks

Btw, can a few of you upload a configuration file so that I could compare settings?i know it will depend on mods, but would like to make sure I didn't mess with an obvious settings. Much Appreciated.

Do you have a scalloped flywheel? Here's another current thread that talks about that issue, https://rennlist.com/forums/944-turb...startup-2.html

You could also use a timing light to verify that it's firing correctly. Just use a silver marker or whiteout to mark TDC on the crank pulley and timing cover for reference. Set the idle timing table to 5 degrees in all cells so you know it's not moving around. Check the timing when it's idling good, then check it when it's not idling good.

Thanks will definitely try check timing for both conditions. Haven't thought of that. I don't have a scalloped flywheel. Factory flywheel. It's been consistent with 4th start being a good one.

Update* so after installing a new coil pack and plug still will not start on all 4. Got a hold of Peep and said I blew the ignition drivers for pins 36/34. Gonna scour the island for transistor V3040P.


BIG PROBLEM. After trying to fix stumbling issue, i adjusted some values via my laptop and uploaded therm the next day. Adjusted idle values and ignition firing order. Upload while ignition on and engine off. Upload successful but car now running with only 2 cylinders. I trouble shot and found cylinders 4 and 1 are not getting anyspark. Which happen to be the firing pair on my wasted spark. After trying to swap pins and configuration on VEMS, it seems ECU PIN#36 is dead. I even bought a new coil pack in hopes to resolve problem. Tried pin pairs on 33 and 35, 11 and 12. They do not start engine. Even after configuring " ignition visual outputs". 34 and 36 runs on two cylinders. With anyone running wasted spark care to share their config so that I can compare. But pretty sure at this point my pin 36 is damaged or in-op. Frustrating because all i did was upload a config file. Any help would be greatly appreciated. -Wayne

This one below
https://cimg7.ibsrv.net/gimg/rennlis...47c0606b3c.jpg
some time ago I had two fried V3040P
My solution was an additional heat sink.
https://cimg8.ibsrv.net/gimg/rennlis...e289682cf3.jpg
I'm running wasted spark as well. --> https://rennlist.com/forums/944-and-...l#post14325562
Attached is my current config file. Though it's for a NA and I use the ICV, but Starting/Idle shouldn't be much different.
Unfortunately I can't open your uploaded config file, it seems to be corrupt.

does the VEMS waste spark use an external igniter box to drive the coil pack, or those transistors are it?

those transistors are it, as far as I know and understand.
But you could also use an external ignitor box driven by VEMS to drive the coil pack, but in my case not needed. It depends on the coil pack.

thanks! will try to compare configurations. weird that the files corrupt. i also blew my pin 36 ignition driver. car runs on 2 cylinders, so i believe my 34 pin is good. did you burn both? where did you end up sourcing your transistors? nearest i found is in Texas off Ebay. Mouser doesnt have any in stock until July 2018. LOL. I guess blowing these is a common thing? Is it cause of the output of the coil packs? Would it be better if we move to coil on plug? I also ended up burning the trace from pin 36 to center pin of transistor.

Actually only one of the drivers was dead. I don't remember which one was it. But I changed both, as you can see on the picture both didn't look very well. I sourced them on ebay. Look for ISL9V3040P3 --> https://www.ebay.com/sch/i.html?_from=R40&_trksid=p2380057.m570.l1313.TR0.TRC0.H0.TRS0&_nkw=ISL9V 3040P3&_sacat=0
I think the problem is that the transistors have to trigger the entire supply voltage for the primary coil without heat sink. If we used the VAG Coil 032905106b with integrated ignitor, then we shouldn't have this problem. Notice the heatsink of this coil for the ignitor. COP with built-in ignitor would IMHO be the best solution, but I don't know a suitable part for the 8V engine.
https://cimg0.ibsrv.net/gimg/rennlis...4e4b406ee2.jpg

would love to know how to set up that coil to run off the VEMS.

my original-from-1984 plug wires might be getting tired...:)
so maybe i'll throw WS at it while im in there LOL

like this for a coil with built in ignitors.
https://cimg7.ibsrv.net/gimg/rennlis...b80b94a713.jpg
https://cimg2.ibsrv.net/gimg/rennlis...3e98fde1da.png

i meant id need to figure out how to change the config of VEMS (probably add wiring) to drive the two "ignition pairs"

You should be able to just use one of the spare outputs and configure it as a second spark output. May even have dedicated spark outputs that aren't in use. Then just configure this pin in the tuning software to be your 2nd ignition channel and set it up as wasted spark. The major point here would be that the above coil is logic-level, meaning it has built in igniters/amplifiers. Therefore you don't need to add one of those transistors to your ECU circuitry, all it needs is a low-current signal to tell the coil to fire. The actual high-current pulse to the plugs is powered by the heavy wiring to the coil unit itself. So, setup should be easy. That's how I did my LS1 coils with Megasquirt... it was less work than actually using the stock coil.

However bear in mind that your original (single) coil output will have that transistor as part of the circuitry and you shouldn't have 1 channel using that (the coil has that already) and 1 channel as logic level. Maybe you can find 2 spare pins or unused spark outputs and configure them as your 2 channels, leaving the original spark circuit intact. That allows you to change back and forth easily if you need to at the expense of running 2 new wires instead of 1.

Also make sure to find out the appropriate dwell time for any coil you hook up. Generally it's between 1-5ms and you can find it online by searching. It's probably a lower dwell time than the stock coil and not updating that value can cause overheating or failure.

I was able to source a ignition driver transistor from a local Rennlister who also has VEMS on his supercharged 944. He also blew his transistor. Luckily he had alot of spares. I also ended up burning the trace from pin 36 to emitter pin on transistor. So i placed a jumper wire from pin 36 to new emitter pin. Shes running like before. Still trying to figure out idle/stumbling issue. I adjusted after start prime from 6ms to 8ms, but not change. still need to turn ignition to on position 3 times, then start on the 4th to get a good start. Might add more fuel and see how it goes. I also figured out TPS wiring. Factory turbo TPS connector - White/Red is 5V - Brown is Ground - White/Green is Signal. I installed a 65mm Ford throttle body with TPS, which took alot more effort than I thought it would LOL. Will update throttle body thread with pics and wiring. Was hoping new TPS would fix idle issue, it didnt. I dont think its the speed/reference sensors, since I only need to turn key ignition on position 3x. I also attached a log file from today after new transistor was installed. This was with a good start. Will update with bad start.

Blowing IGBT happens when firmware is flashed and ignition/injector fues is not pulled or config change is made (in Wayne's case I suspect he changed to inverted coil output => instakill for transistor).

The VAG coilpack above can be fired from stepper output (EC18 pins 4, 5, 10, 11), just needs 12s supply and additional ground connection to cylinder head.

Regarding your starting, you need to add some fuel for afterstart table or try to add a tad cranking fuel (go to Starting/Idle menu and add 10% to cranking VE and give it a try).

Yep was trying to fix stumbling issues. BTW, do the VAG coil packs on the VEMS store fit the 8V head? Been looking to do coil on plug to clean up the engine bay a bit. Going to try a wire tuck and hide majority of wires. ( yes, its the current JDM fad, figure I try it for us transaxle crowd )

Yes, been slowly adding fuel on afterstart and pre-crank. But shouldnt the stumbling issues go away after reaching operating temp? Much Appreciated Peep. I figure I bother you enough via Email and try the VEMS community on here. :)

shortyboy - was the Ford TPS as easy as connecting the relevant wires to eachother on the OEM harness, and then re-calibrating in VEMStune or anything else needed?

Raceboy - can a "plug-and-play" single-coil VEMS be somehow upgraded to do WS? another plug-in adapter (something easily reversible if needed)?

No adaptation needed, just route two wires from EC18 pins 4 and 10 for example to coilpack input pins and give coilpack 12v and ground.


Wayne, I refer as coilpack to the wasted spark coilpack, you are referring to coil-on-plug (COP) coils and no, these do not sit well on 8v head. Only one that would is the shortest one from 2.3 v5 engine for example but still not good.

Regarding blowing IGBT's, I have always said that if engine runs well at full throttle and load, there's nothing wrong with ignition system and the issue lies elsewhere. It almost never is an issue, 1% of cases is where plugs are old or something like that.

Yep, I just spliced in proper wires from ford tps to wires on OEM harness and calibrated in VEMStune. Thats it.

Understood. Never touching that settings again. Lol. Looks like I need to get longer spark plug wires to hide wasted spark coil near battery tray then.

cool.
i noticed in the ignition settings under base setup there is an option for "EDIS"...is that the correct software-side change or is there more to it past that?
i tried googling this and got lots of vems.hu results but its all written in computer-people-speak :)

EDIS is something to ignore, it is ONLY used when real EDIS module is used what decodes wheel and feeds VEMS with simple square signal.

There's nothing else to change other than change output channels to stepper ones.

cool...love VEMS.

found this.. :)
http://www.vemssupport.com/forum/index.php?topic=441.0

where rob says:
Configure Ingition Outputs
Ensure that the Stepper Motor Driver is enabled as above. Then configure the outputs, using the Four Cylinder example as above the
Go to Settings->Ignition Outputs and set the sequence:

for a 1342 4-cyl i assume that would mean inputting as: EC18-4, EC18-10, EC18-4, EC18-10 ?

one guy there (and odonnell, above) mentioned tweaking dwell time - necessary?

Yes exactly like this. The VAG 4cyl coilpack works nicely with 2.5-3.0ms dwell.

Seems like your settings for start up and idle helped my issue a bit. I copied over your priming, starting, warm up, crank and aftterstart. Just modified afterstart settings to 100 cycles each. Not sure what settings that is for though. Anyone car to explain?

Getting ready to install mine when I get free time from work. How did you guys route the O2 harness into the foot well?

on mine where the ECU sits under the steering wheel i got a long O2 sensor extension cord (ebay), ran it from approximately the stock O2 sensor location, over to the battery cable grommet, then behind the firewall/cowl over to the windshield wiper wire grommet and down inside.

on a late 944 i'd think you could cut a slit and squeeze the O2 wire into the "DME harness accordion grommet"

I haven't put mine in yet..but I've been thinking of routing it through the shifter hole...any one done this?

Did that with the innovate AFR cable.

This is the easiest solution, I ran my EGT and oil temp sensor wiring the same way.

Though, If I had the engine out/more space, I'd drill a dedicated hole in the firewall and install a large grommet to accommodate any additional wiring. I plan to run a spare +5V reference wire from the VEMS for use with extra temp/pressure sensor sensors for datalogging temperatures/pressures at various points in the intake/intercooler system for testing purposes.

In order to run O2 wire through, you will need to de-pin connector. I did mine this way. I will also be adding in a few sensors also. Oil/ Fuel Pressure and Oil temp.

I couldn't get the connector pins to come out so I cut the connector and bought a new connector/new pins and ran it through and reconnected in once it was in the footwell.

I have not installed mine yet but was able to de-pin the connector on the VEMs side (you need to slide out a small purple retainer clip to do so) and plan to run it through the DME accordion grommet.

My O2 plug fit thru the wiper wire hole...very tight but it *just* fit.

Has anyone wired up the speedometer line to VEMS ecu? If so, did you go directly from sender at trans or tap at the dash?

So just installed VEMS on my 88 951s. So far so good; car started right up and Peep's instructions (although pretty high level) were sufficient.
One thing that I noticed was a lot more fluctuation in the tach, both at idle and at steady highway speeds.
Have others noticed this and wondering if its a simple fix.
Thanks

Did you try stepper output as I suggested?

Not yet Peep. I can only work on the car on the weekend so will take a look then.

Thanks
Sean

https://cimg3.ibsrv.net/gimg/rennlis...6ad29bfdb6.jpg
Anyone have the round gauge up and running? No matter if I try and connect to my computer COM port or the VEMS ECU it will not power on / show up in Vemstune. The instructions are lacking IMHO, I see these wires which one is red/white with a fuse so Im guessing it wants 12v and the other blue wire I originally assumed would be ground except the printout shows rpm input. I don't need any of the extra wires for WB02/input as I am just trying to stream the data from the ECU when the laptop isn't plugged in.

Also is anyone else having to adjust fuel req quite a bit? Seems like once I start the car / comes out of cold start mode - car happy and fuel req good / then after a long drive and the car is hot - the target values are way different (lambda). Are you just supposed to shoot for a happy medium or tune it based on hot or can it be weather/temp dependent? My car does not seem happy timing wise (haven't touched it - on a base map sent with the unit) - I have contacted a local well known tuner who hopefully can dial it in, I think the fuel map is OK as we have messed with it quite a bit but its so lazy coming into boost the timing is just obviously way off - its nothing like the car used to be. I expected to have to dial this all in so not complaining - just curious what others are seeing that have been messing with it themselves.

You probably need to tune the air temp curve. Heat soak at the air temp sensor can cause weird running, so there's a curve to compensate.

Sam, you need 12v for the gauge and also RS232 connected but with twisted polarity, so that Rx of VEMS is connected to Tx of gauge and vice versa.

So I've had VEMS up and running for a couple of months now on a mostly stock 951s (38mm Tial and manual boost control limited to 15 or 16lb) and all seems fine after working out a few minor quirks. Good track day last week but I was stupid not to refresh the sparkplugs as the ignition started to breakup at 5500 up, but that is now resolved with fresh plugs.

I had I've run the autotune in the VEMS software, but I am wondering whether it is worthwhile having a dyno tune so as to maximize efficiency/power. I am more of wrench guy than a software guy and given the investment so far I want to ensure that I am getting all I can out of the VEMS (and also realizing my own limitations and lack of experience with the software).

Have others taken this route? I'm no aware of any VEMS-specific tuners in my are (Long Island NY) but there are a lot of dyno tuners with good reputations in the area. Would the VEMS software be easy enough for an experienced non-VEMS tuner to pick up and easily mange?

Thanks for any thoughts or advice.

You can definitely benefit from a proper tune. The base ignition map is very conservative. If all you are getting tweaked are the fueling and ignition maps, then yes any good tuner can do that.

Im scheduled to have my car tuned on a dyno July 26th by a place up north of Chicago - seems like they tune about anything and are familiar with most standalone systems. $725 flat rate.

Thanks 944crazy and Sam. I will speak to some of the local shops and see if I can find someone who is comfortable with VEMS. Not surprisingly, most of the shops cater to VW, Subies, Nissan, etc and nome mention VEMS on their websites, but I imagine that someone familiar with standalones could work with it.

Sam - please let us know your results. If only you could dial in the low-end grunt of your GTI (my DD is a '16 GTI with performance package).

Sean

$725 to tweek Peeps map, they might do it in a hour or two...

Well my car doesnt seem real happy on Peeps basemap. Driveable sure but its fouling plugs and seems to miss a bit, and lazy as heck coming on boost. Thats fine - its a basemap and I expected to have to tune it / dial it in. I prefer to have someone who knows what they are doing do it and be done with it.

Yes I didnt have much luck finding anyone in town that would mess with the VEMS, the big go-to tuner guy in town said No - guess he just wants to work on what he is familiar with. Ill be sure to report back.

Yes love that GTI from down low :) Great DD!

I guess I should mention that it is entirely possible that you have underlying issues that have nothing to do with the VEMS system or its base map (for example, is the Wideband O2 sensor properly calibrated? Vacuum/boost leaks? Exhaust leaks? Fuel system issues?)-----but let's assume that you do not. The VEMS auto-tune feature is great for dialing in idle/cruising/light load stuff in terms of VE/AFR (save the ignition timing curve for the dyno). I have done a progressive sequence of manual tuning of the base VE map to get into the ballpark, followed by some lengthy autotune drives (datalogging the entire time), then going back and manually fine-tuning the VE table after studying the log. Rinse and repeat several times. I first did this with the bone-stock engine/turbo/fuel system, then with upgraded turbo/exhaust and larger injectors/fuel pump/3 bar fuel pressure regulator (after changing the base/reqfuel setting), and the engine runs better than it ever has with crisp throttle response and minimal turbo lag.

It's the part-throttle/transitional street driving stuff that is most time-consuming to set up, so if you get much of that out of the way, the majority of the dyno session can be spent optimizing a safe ignition timing curve and focusing on high load/full boost maps.

Tuning a speed density ECU is virtually the same between all systems. It's a simple load/RPM map.

Also when you change the ignition map, you will need to re-visit the VE table as there will surely be fueling requirement changes to meet target.

Finally got around to installing the VEMS. Car starts and runs...barely. It's off the chart lean right now. I have attempted to pair the VEMS with my laptop but it's requesting a pairing code. Wondering if you guys have an answer? TIA

Pairing code is 9111. You need to calibrate the O2 sensor outside of the car before you start it. Then you can let it warm up and adjust fuel-req , and then after that you can begin auto tune to adjust the VE map.

Here are Peep's instructions for the initial set up of VEMS:

1. Remove AFM or just disconnect.
2. Install air temp sensor into intake pipe between intercooler and throttle body.
3. Weld WB sensor bung into downpipe after turbo.
4. Route the WB cable of the adapter through the firewall rubber grommet (can be through the one that DME harness goes through or drivers side one, your choice) and connect KLR boost line to VEMS ECU.
5. Download latest VEMSTune software from here and install into your computer: http://vems.hu/download/v3gui/NIGHTLY/
6. Connect VEMS to laptop computer. Please see tutorial here: https://www.youtube.com/watch?v=ipbimqIGgX4
If your VEMS is equipped with Bluetooth you need to pair it with laptop (pairing code is 9111) and it will install two "Serial over Bluetooth COM ports" and you just select the lower numbered one in Vemstune and click "OK" (not "Detect ECU"!)
7. Calibrate WBO2 sensor with sensor connected to VEMS but not screwed into exhaust (this is important)! See the tutorial here: https://www.youtube.com/watch?v=D6dVF_r5d3o
8. Verify the sensor readings and calibrate TPS if needed: https://www.youtube.com/watch?v=aHGh-e7j1qM
9. If everything is as it should be (temperature readings sane, TPS reading corresponds to real movement etc), then start the engine.
10. After engine has started, let it warm up and adjust req_fuel constant to aim for lambda 0.95-1.0 at idle. This can be done by going to Base Setup -> Engine Setup.
11. If req_fuel constant has been set, use autotune feature if necessary to fine tune fuelling. Please see tutorial here: https://www.youtube.com/watch?v=U7fyfhY_FhI

Thanks so much! I'm going to try this today, will report back.

Well I followed 944 crazy instructions worked just like it should.

Although for some reason the tachometer in the dash is not working. The tach on the laptop is working and correct so going to have to figure that out.

Drove the car for a little bit it's going to need what appears to be quite a bit of tuning but at least the thing starts, runs, and drives!!

that's a big hurdle for me I have absolutely zero experience in tuning.

How do you place VEMS into auto-tune is my next question?

Tutorial here: https://www.youtube.com/watch?v=U7fyfhY_FhI

gpr8er: here is what peep sent me when my tach wasn't working:

Sam,

How did it go with your dyno tune? Hoping that it was worthwhile for you.

yeah me too???

It was late when I picked it up so I just threw it on the trailer and brought it home - I finally drove the car today for the first time for about 25min. it seems to run very well- they use a mustang dyno and tuned it based on torque. It only made 280hp / 309tq @ 18psi / 93oct but I know every dyno reads different so that doesnt really tell the whole story - boost comes on hard like it used to on its old MAF setup which was missing from the basemap. They said it needs a bigger turbo -I think it needs a new engine >2.5L and a bigger turbo ;)

My car has only been on a dynojet previously so its hard to compare to previous setups. They feel they dialed in the fuel/timing and stated I wouldn't have to mess with it at all anymore. I really want to put a few more miles on it / cold starts before I claim success but I think they know what they are doing. It was their first Vems tune and they stated it was easy/straightforward apart from the nightmare that the "round" gauge is, they got farther with it then I have in getting the computer to talk to it but we still never managed to have the ECU stream to it - to be continued, could be my wiring error. Now I can finally mount the VEMS unit, hide it and enjoy the car again.

Only real mods are to4e .57, bigger injectors, tial wastegate, 3" testpipe/exhaust. They said the turbo was smoking a bit but it cleared up after a few pulls - its 11 years old with ~ 6k miles on it.

Sam. would you care to share the name of the tuner?

Dynosource Performance - their shop is impressive and they have a separate dyno area / room and seem to tune just about any standalone system / any make and model. Locate in Antioch, IL north of Chicago.

In order to have your round gauge to work with AIM stream it needs: reversed RS232 connector (otherwise it would connect Rx with Rx and Tx with Tx while it needs Rx -Tx and vice versa). Also Vemstune has to be configured to have the AIM stream enabled.

Thanks for your help Peep. Yeah I have done that (made a connector with Rx-Tx with one pin grounded) - we actually could not get the ECU/Vemstune to Enable AIM. On the menu screen the pull down option for Enable AIM says No and No where I would expect to see No and Yes.

Thank you for the info Sam

You mean this place? :) Allows easily to enable AIM mode (on the ECU, not the gauge, it works by default, just have to have proper firmware flashed onto it).https://cimg8.ibsrv.net/gimg/rennlis...dc86bcef0c.jpg

Yes that menu shows me "No and No" in the dropdown. I dont have a "Yes" option to enable AIM. Ill have to get a screenshot and send it to you. Is there a way you can take my file and change something to allow the Yes to show up in the dropdown? Thanks.

Please email me your vemscfg, I'll take a look.

I've been running VEMS logs and my stock (blueprinted) injectors (1988 turbo S with 3 bar fuel pressure regulator) are hitting 100% duty cycle at high boost (15 to 16psi, or a bit north of 210 kPa). Also seeing a spike in lambda so it looks like I am leaning out.

What are folks running as far as upgraded injectors? I've read through a lot of older (pre-VEMS) threads here and on Pelican but wondering what the current state of the art is for 951/VEMS.

Thanks
Sean

ive got 2 sets of low-z larger injectors (55lb/hr and 72lb/hr) if you're interested. stock 951 is around 36 lb/hr.
$125 shipped either set.

they are the delphi fat-body type, physically interchangeable with the stock 944 injectors but have a better spray pattern.

VEMS can adapt to any kind of injector, just tell it the new flow rate and the injector type (low impedance vs high impedance firing patterns).

they look like this:

https://www.bmotorsports.com/shop/im...00x800_mat.jpg

Spencer - thanks for the reply. My intuition is that 72 lb/hr is overkill for me at 2x stock. 55 lb/hr should be adequate as those are 50% greater than stock and I'm not looking to do much more to the car as far as mods go.
I will shoot you a PM.

Those are good injectors. If you want only slightly larger than stock, check out Ford "green giant" 42# injectors. They are plug and play as far as mechanical fitment and the deadtime/latency values are known.

Spencer if Porvair won't take them I may be interested in. What model they are? Do they need any tweaking with resistance, latency etc? Sorry for the OT...

I just installed a set of 55 lb/hr (which I believe calculates to 577.5 cc/min) injectors vs the stock 36 lb/hr (378 cc/min). Now I need to reset req fuel in VEMS. Basic req_fuel calculation:req_fuel = 6.49 * (D / N / I)
I injector flowrate (cc/min)
D engine displacement (cc)
N number of injectors (eg. 4 for a 4 cyl port injection)

So I would have expected a req fuel of 10.7 with the stock injectors (6.49*(2500/4/378)), however it was set at 9.6 from Peep, which I presume is correct. The calc with the larger injectors comes to 7.0 based on teh above formula, however I set it to 6.3, which is my original 9.6 setting *36 lb/hr/55lb/hr. That is yielding good lambda at idle (i.e., 1.0) so it seems Ok.

What am I missing with the calculations?

And thank you Spencer for the new injectors.

are you running @ 3bar or 2.5 bar fuel pressure?
they are 55# rated at 3 bar, less at 2.5 bar stock 951 pressure.

i remember having to set my reqfuel a little different than the equation would suggest too, didn't care to ask why.

Running 3 bar. I'll drive it around and run some logs to see how those look. Just want to make sure that I'm not missing something obvious here.

Make sure your dead time (injector latency) is set correctly. Hopefully you can find a datasheet online. Your low rev regions will be a lot smoother and easier to tune, with less AFR variance at idle.

There are a bunch of "lag" time tables for a variety of injectors listed at http://injector-rehab.com/shop/lag.html

So I'm having an issue trying to get my tach in the dash working. I've checked with Peep and everything appears to be correct in the settings and also the harness. Hoping someone here can take a look at their settings and clue me in as to what might be wrong . https://cimg0.ibsrv.net/gimg/rennlis...5f44f22e4f.jpg

https://cimg7.ibsrv.net/gimg/rennlis...f94c074821.jpg

sorry for the discombobulated posting, trying to do it off the phone.

I just checked my config and they're exactly like yours. Possibly incorrect wiring or bad ground?

Has anyone come across erratic power and loss of lambda signal under hard acceleration? I experienced what felt like a rev limiter along with erratic lambda readings (lambda maxed out at 1.83) under hard acceleration, with rpm north of 5k and higher boost (180+ Kpa). Car drives fine otherwise, but it will not pull past 5k or so.

I tried to recalibrate the 02 sensor and ultimately just replaced it with a new Bosch unit. I just came back from a drive and am experiencing the same issue - it feels like a rev limiter is kicking in at 5k or so and the data log indicates that lambda is spiking up to 1.83 (which Peep said is equivalent to the 02 sensor no longer communicating with VEMS). FWIF, I also upgraded the injectors to 55lb/hr after replacing the o2 sensor but that does not seem to help. And I have confirmed that the rev limiter in VEMS is set at 6699, so that is not the problem.

Any thoughts are most welcome.

Thanks

try adjusting the RPM range that the ECU sees lambda input at WOT.

Spencer, how is that done? Is that under one of the Base Setup menus?

Base Setup -> EGO Closed Loop Control, about 2/3 the way down the popup.

Thanks Spencer. I gave that a try and no difference.

At Peep's suggestion, I re-gapped the speed and reference sensors (great fun) to the high end of the spec (I'm about 1.2mm now) but no difference.

Originally I thought this was a fuel issue as I was focused on the lambda, as that spikes upon this problem occurring (and recovers in 30 sec or so) but I am now thinking that the base problem may be an ignition issue. The engine revs to redline in neutral, but as I approach 5k under boost, I see the revs (per the VEMS log and the dashboard tach) suddenly drop to zero for a brief moment, with a corresponding loss of power and a few ugly exhaust pops.

Plugs are new and wires (Clewett) and cap are 3 yrs old with < 10k miles. If one of those were an issue I would expect a degradation of ignition, but not an abrupt cut.

Possibly a coil issue? Anytime I had a bad coil, the car just stopped as opposed to acting like it is hitting a wall from an ignition standpoint.

I'm corresponding with Peep who is very helpful, but also wanted to throw this out to the community as any suggestions are welcome.

Image of one of the logs is below.

https://cimg7.ibsrv.net/gimg/rennlis...4819f0c735.png

What is the RPM at the 50 timestamp? It looks like you hit the rev limit at that point. It's also odd that TPS is tapering downward, maybe you are tripping the BOV to open and causing a surge?

...i wonder if your throttle plate is being pushed slightly shut by the boost?

Looking at that log, everything drops to zero at the exact same point (ignition, fuel injectors, and TPS), which to me indicates an issue with the DME power supply and/or sensor ground circuit.

I would check the typical 944/951 bugaboos, the sensor grounds at the bell housing, DME power supply circuit, etc.

The first time I browse the 951 forums and this is at the top. Some good convos in here for sure.

I like the line of thought here; have seen that exact same symptom and cause at higher load (Granted an entirely different setup but I think this premise deserves some merit).

Good morning all and thanks for all of your ideas.

The rev max @ the 50 timestamp was 5,392, and the drop is the TPS was me backing off the throttle as the ignition cut is very abrupt. I have similar logs where I was at 100% TPS and it typically cuts somewhere between 5,000 and 5,200. I back off the throttle when that happens as it is very abrupt and not as smooth as a "normal" rev limiter given that this appears to be a full cut of the spark and not just holding it at the limit.

I will check all of the grounds, and I'm running a solid state DME relay so will swap in the original to see if that helps.

All that said, I hear from Peep this morning and he believes that It is a trigger error, and when VEMS sees the trigger error it cuts the ignition. He sent my info and a request for a solution to VEMS hq because he has seen that it happens totally randomly on 944 turbos, around 2-3% cars have this issue, others work just fine.

I do have a track day @ Pocono coming up in a few weeks. I'm not racing or out to set a lap record, so I can either short shift @ 5k, or may revert to stock, which I prefer not to do as its a PITA and like all of the folks on this thread I've invested time and $ and hate the idea of undoing that. Or replace the all season tires on my MK7 GTI and run that, but I really prefer the 951 as it is so visceral on the track.

Can you display the trigger error on your log? It’s an option to display that as well as the other stuff so you can easily see a correlation.

Which s and r sensors are you using?

I am guessing this happens to you in all gears at that RPM?

Yes, VEMS can display trigger errors in the VEMS log (File/Open Log file).
https://cimg5.ibsrv.net/gimg/rennlis...c9f2c9cfe3.jpg

I wonder, why in the posted VEMS log above (post #201) there is only 19.00° spark angle @ 4445RPM and no further advance @ 5000RPM. Therefore I would check the RPM/MAP Ignition table, if there is any error.
Edit: The VEMS log above (post #201) shows the view mode "File/Open Log file to separated frame" which doesn't show trigger errors by default, but you can enable this in the edit mode.

Sounds like the dreaded VEMS trigger issue above 5k rpm. I had the same thing...only fixed by going to a TTV 60-2 flywheel.

R and S sensors are brand new Bosch. I swapped those in a couple of days ago and I regapped them to approx 1.1mm per Peep's advice. That exercise had basically no effect on this performance issue as I had similar behavior before and after the replace and regap. The issue is occurring in 2nd and 3rd gear; I cant try it in 4th or 5th without fear of losing my license...

I changed the view of the logs of my post in 201, and there, in fact, no trigger error. Below are two screenshots - one right before the cutout, and another immediatly immediately the rpm dropping to 0. The spark angles do look way off vs the base tune map which is embedded in the lower right corner of each pic. I'm perplexed. Maybe overwhelming the coil and it is giving up momentarily?




https://cimg6.ibsrv.net/gimg/rennlis...88bb306093.png
https://cimg8.ibsrv.net/gimg/rennlis...a069129684.png

Did you swap out the coil/coil-pack? Could be a worn/bad coil, or it's under a ton of heat.

I have not tried a new coil yet. I may do so, but I feel a bit that I am throwing parts at this as I swapped in a new WB02 and speed and reference sensors in the last few weeks. The existing coil is stock and likely has a lot of miles, but I was hoping to get some better direction before buying new parts.

Thanks again for the log info.
In the latest logs I see MAT at negative 100C, which is basically the same as having no sensor connected. This is an electrical issue. Check the wires, connectors and sensor. An easy test is to connect the logger, position your laptop so you can see it while you're under the hood, key on-engine off, then wiggle each wire, connector etc until you see MAT spike/ drop off.
Also, the engine rpm dropped to zero, which should be impossible if your description of events is accurate. I suspect another electrical issue between the VEMS and your SR sensors.
MAP sensor also dropped to zero kpa, which is absolute vacuum and not a real reading. Is the map and mat sensors powered by the same wire?

I agree. The 944 is prone to sensor ground issues at the best of times (I have seen all kinds of issues with poor bellhousing grounds on stock 944s and 951s at my shop over the years), and I suspect that something with the VEMS system combined with typical 30+ year old 951 wiring issues manifests itself in unique trigger issues with the 2-3% of cars mentioned by Porvair.

My VEMS-equipped '86 951 with 170K miles and all original DME wiring harness runs great, both with older Bosch speed/ref sensors and recent cheapo FAE replacements (the Bosch sensor connectors were broken). However, I am leery of the old wiring harness; I recently placed an order with Lindsey for a few parts and had them throw in a new speed/ref sensor wiring harness to have on hand.

I would check and clean all relevant power and ground connections, starting with the bell housing grounds, but also battery, alternator, coil, fuse/electrical box, etc. It is entirely possible that the original Bosch Motronic system is more forgiving in terms of clean electrical signals than is the VEMS . . . .

Thanks all - I will check the grounds and sensors.

More urgent issue this morning; I decided to join some Porsche buddies for a 2 hr drive to Lime Rock for the vintage festival. I left the house with my 2 sons (age 9 and 23) at 4:45am with a plan to meet up in Queens, and got about 2 miles from home and the car just quit. Tach to 0, and no spark. I was on the Northern State Parkway on Long Island, which has some oppressive towing rules with exclusive towing franchises, so AAA could not tow me. We caught an Uber home (note: the Uber app does not really expect folks to be picked up on the side of a limited access highway) and got my Cayenne and a tow strap, and the older boy towed me back home (man, I'm too old for this crap...).

Got home and confirmed no spark. Good 12v at the coil, and I checked the coil resistance, and primary and secondary are in spec, When I test the spark however,( https://clarks-garage.com/shop-manual/ign-04.htm ) I get a very weak, pathetic spark. So perhaps the coil is part of the issue. Not sure if I can find one locally today so I may need to wait a couple of days before diving back into the morass.

And I swapped DMEs while on the side of the road but no effect. And the VEMS display on my phone does indicate 170 rpm when I crank the engine, which I would think means that VEMS is sending a signal to the coil?

Coil and distributor

It shows that the VEMS is getting a good signal from the SR sensors at that rpm. I'd focus on diagnosing the erratic sensor inputs into the VEMS. garbage in = garbage out. Your weak spark could be due to a weak battery from cranking it repeatedly on the side of the road.
BTW I tested the cranking speed on my car with a new battery and new starter and it reads 210 to 220 rpm, so you're not far off.

I have a known good coil you can have, replaced mine just as its old and I was in there but it worked fine.

Thanks SamGrant951, but I think I am good for now. I ended up picking up an MSD blaster from the local speed shop.

Got home, installed the latest $part, and a big nothing. No spark. Nada. Nil. So I ordered a speed and reference $ensor harness (as that was very brittle) and new cap and rotor.

In frustration, I decided to pop in the stock (Lindsey chip)Motronic boxes and ... it started. Ran like crap as no AFM and missized injectors, but it ran.

With that, I just grabbed more wrenches, screwdrivers, and implements of destruction and reinstalled the AFM and stock injectors. Took it out for a spin and it pulls like it should up to 6500+.

Not sure of next steps. I hate to give up on the VEMS, but after leaving me and my kids on the side of the highway I'm gonna have a hard time feeling comfortable with that system in my car, especially as the factory parts work reliably.

Thanks for the update. It sucks that the VEMS hasn't worked out for you, but I really feel like the issue is not something inherent to the VEMS itself, but a combination of its requirements and the marginal 30+ year-old 951 wiring. The reason that VEMS has become prominent in this application is that it is one of the few standalone engine management systems that plays nice with the unique speed/reference sensor setup of the early Bosch Motronic systems, which require an inordinate amount of processing power to accurately interpret. The VEMS seems to work great for most of us, but a small yet significant percentage of cars seem to have the exact same issues. Hopefully we can all collaborate and figure out what the issue is.

In the meantime, I went back to the basic VEMS installation manual, as I remembered it had some specific instructions about separating the ECU sensor grounds from the "power grounds" (electrically noisy components such as fuel injectors, idle control valve, etc), as outlined on pages 5-7 of this PDF manual:

http://downloads.vems.com/documentat...etupmanual.pdf

My interpretation of the factory 944/951 wiring diagram shows that the sensor and power grounds terminate at the same point on the bell housing. To anyone who is having these issues, I strongly recommend cleaning these grounds to start, and looking into replacing questionable sections of the DME wiring harness. Perhaps running a separate ground for the sensitive speed/ref sensor circuit would cure some of these issues . . . . . I also wonder if "noisy" ignition wires or other EMI issues are at play.

Thanks Chris. I'm hoping that a refresh of the speed and reference sensor harness helps, and your points on noise and possible grounding issues make sense.

Before running across the inability to rev over 5200 or so I had a really annoying issue with a bouncy tach, both at idle and at speed. It was not due to varying rpm but rather a noisy signal, which I thought was perhaps a grounding or other interference issue somewhere along the line. Peep did not have any advice other than to swap the tach feed pin from EC36 pin 4 and to EC18 pin 4. That cured it, but I still notice a slight pulse in the other gauges (factory boost and fuel) which very well could be a grounding or noise issue.

I will probably leave the VEMS harness in place including the MAT and WBO2, so swapping between factory and VEMS would then be a 30-minute process. I will attack the bellhousing ground when I install the new R and S sensor harness. And then, when I feel the need to punish myself, I can try to get this working like it should.


.

MAP = 0 triggers a Fuelcut by default. Check Base Setup/Rev Limit/Overrun/Safety-->Overrun Fuelcut/Fuelcut below xx kPA.
So I think since your MAP is 0 VEMS cuts the fuel as you can see Lamba reading >1.83 / Pulse width = 0.5mS.
MAT -100°C is probably like the MAP reading a sensor or wiring issue, which doesn't help either. Or your MAP hose has a sharp kink.
My conclusion, too many parameters out of default scope. Maybe VEMS crashes because of the implausible values and displays RPM = 0.
Based on the screenshot above, I don't think it's a coil issue.

hmm....really don't like hearing about issues like this. I have held back selling my last set of DME/KLR boxes just in case. On a lighter note I was able to finally figure out why my VEMS would not communicate with my laptop. Turned out to be a com port issue. Anyway I was able to FINALLY take the car out for a tuning drive and learn how the VEMS responds to live tuning. Drove around for 15-20 minutes and got the car running ~decent on light throttle. I felt like that was a success for the day and parked it. I move at a glacial pace. LOL

With trigger error rpm should not rop to 0 and that applies to fuel injector pw also. Open log file and add a pane where you choose to display some flags also (afterstart, trigger error flag etc), if you see afterstart flag right after cutout it means that you get arcing at high load and ECU resetting. With the 5k rpm trigger issue wverything works like it should also after it happens, you should not get no spark situation.

I wanted to share my latest update re VEMS on my 951. I very much appreciate all of the input and focusing on potential signal noise, I have just finished installing a Lindsey speed and reference sensor harness as mine was marginal, and also took the time to clean the bellhousing grounds. Just wish I had two sets of hands as an extra small set would be helpful at times...

So I just wrapped that up and the car runs great with the factory ecus. I still have all of the VEMS harnesses in place so I popped in the VEMS box and, nothing. As soon as I synched the laptop I realized that VEMS is in "Safe Mode Active" which I believe kills spark and fuel, which may be very safe but of course not much fun.

I have a message into Peep, but wanted to also get this out to the community here as there is a lot of knowledge and the more dialogue that is out here the better it may be for others who run across challenges.

So this may be a simple question, but how does one shut off "safe mode"? I don't have any anti-theft connected so this sounds like it should be simple.

Thanks all.

I think it's this your'e looking for
http://www.vems.hu/wiki/index.php?page=SafetyMode

But to be sure, wait for peeps answer also :)

Thanks DLS. I saw that but I have no idea what is meant when the documentation says

"from firmware version 1.1.83 (I'm running 1.2.32) the Injector and ignition outputs are disabled by default right after firmware upload, until they are enabled by either:

• a special firmware-version dependent configuration word (named prohibit, 2 bytes) re-enables it
• naturally, after firmware upgrade, this word will be wrong, and need to be set. VemsTune sets correct value automatically after upload of full (.vemscfg) config (to leave "safe-mode")

Sounds like I need a special configuration word, and I would try one of the above ones but I have no idea where to even input the configuration word.

Have you done a firmware update recently?

Otherwise i think this is the answer
To reenable the ECU:

• easy way: In "firmware upload" press enter (not "cancel" !) in the "upload saved config" step.

That's all. Nothing special needs to be done in the normal case. For advanced users:

• if you cancelled that step, you can also do it later: upload a proper config and VemsTune (newer than 2011-01-14) sets automatically.

but best to wait for peeps answer :)

Any news Porvair?

DLS - I was traveling for business and just got home. I will play around a bit tomorrow to see if I can get it running well on VEMS, but I have a track day coming up and may just run the stock ECU as I'd rather be driving than working on the car:)

I was able to figure out why my tach wasn't working, Thanks to this thread from Tom,

https://rennlist.com/forums/944-turb...-question.html

Turns out that when I did the VEMS install I also changed out the engine harness with a 89 65k mile harness I have had on the shelf for years. For whatever reason IDK the 4 pin square red plug only had the male side. So what I did was connect the VEMS tach output to the Green w/ black stripe = tachometer signal. After using the trick Tom provided here and verified the issue.

I believe my car may need an exorcist to drive out the demons. I installed a new harness for the speed and ref sensors, and cleaned the bellhousing grounds. I took it to the track on Sat and got in 75 really good miles with the stock ECUs with the car running very well. Then, halfway around the circuit, it started surging and backfiring and losing revs and power. It died just after getting it into pit lane and no spark. Towed it home (I trailered it there), checked the coil and installed new reference and speed sensors and still no spark. Guessing that the DME may be shot, but I don't see any obvious burning on the circuit boards

So I am back to sorting out this VEMS.

It is stuck in safe mode and I cannot seem to get it out. I tried a firmware update but as soon as I go into the wizard I get a Comm error and my laptop stops communicating with the VEMS. I skipped tried skipping the wizard, and it looked that I was updating the firmware, but as it ended the count down of the installation ETA (approx 10 minutes) it disconnected with another communication error, so I never got to the step where I can upload a good, saved config file.

Any ideas?

Try another laptop or a new WEMS.

What is a WEMS?

http://www.abc.net.au/news/2018-09-1...78?pfmredir=sm

First, be 100% sure there are no vacuum leaks (smoke test if possible). Then, get your base fuel/VE map in the idle and low RPM/cruise range sorted before messing too much with the idle settings. A fresh valve job will make the cylinders seal better then they did before the job, which changes the VE. Auto-tune is an easy way to get this dialed in quickly.

Also, make sure that the idle contact switch on the TPS is adjusted properly (this can easily be verified in the VEMS software, it will say IDLE ON at the top of the main screen if the contact is closed). If all of the above are OK, then start playing with the specific idle settings.

Hi Chris
Thanks for replying. It's fixed(see above as I edited my post)
My mechanic sprayed WD40 to check for vacuum leaks and all was good.
After fixing idle issue by deleting the ICV, I just need to reload(again) the original config Peep supplied with the unit and do a decent autotune. The car is running great and I can't wait to get the autotune done.
Bruce

Still chasing my start issue where I get a good start and bad start ( lots of stumbling at idle and all rpm ranges.) in Vems which sensor is it getting timing from? The speed or reference sensor? My theory is that I have a bad sensor which is why I would need for it crank over a few times to see TDC and have a "good" or normal start. I've tried numerous settings for cranking, afterstart and warm up. Again issue shows up cold or hot engine. Also DME temp sensor comes to mind also. Car runs great on that good start. But if I get a bad start and try to run it, it feels like I'm running very lean, lots of sputtering and back fire. It almost seems like it's loading a different tuning map, like an A, B setting. Also suspecting fuel system, not enough pressure first few cranks. Or injector settings like latency or dead time. Gonna install an 044 soon. Waiting on the adapter.***did tons of reading and im leaning towards fuel system. Either incorrect injector settings like dwell time/latency, voltage drop to fuel pump, bad check valve or even fuel damper. FPR is good as gauge is steady at idle and reads 3bar/43 psi with vac line off.***

Is anyone running delphi low z 55# injectors? If so, case to share your injector settings?

Have you logged good and bad running sessions and compared data? It seems that with a running issue this drastic, something obvious will show up in the datalog (AFR, MAP, CLT, MAT, etc).

Hi Chris,

Ive done some short logs as I dont want to run it that long with a bad tune, which it looks like it. It shows lean afr across the board. Also, whenever Im cranking the starter, if rpms on first crank is below 1k, its a bad start, but if it goes past 1k, around 1300rpm it always starts good. But since adjusting injector dead times or latencys, Ive been having good results, car is starting much better and throttle response too. Probably need to adjust it further for cranking at a specific voltage. Still cant find dead times/latency specs for my delphi 55# injectors. Initially VEMS was set to 550cc, where in fact they are 578cc flow rate.

I don't see injector latency settings being slightly off as the cause of an intermittent lean running issue. If that were the case, it'd be lean/bad running all of the time. How lean is the AFR when it acts up? If the sputtering/backfiring happens during low speed/low load conditions, I doubt you'll hurt anything.

Are you saying that you see either 1000 or 1300 RPM while the starter is cranking, before the engine starts? That isn't right, starter speed is ~400 RPM at the best of times. Maybe you are having a version of the trigger errors I keep hearing about? Does the engine RPM reading in VEMS seem to correspond with actual RPM? Hook up a digital timing light to a spark plug wire and compare the readings to what VEMS is showing.

It runs lean or rough across all rpms and even after warm up. I've done the VE analysis when it runs rough and it's about 5 - 10points lean. It's very weird. Cause, when I have that "good start ", then shut it off then immediately start it I would have the issue, rough/lean condition. I can see on my windows tablet and tach that rpm matches. If I see the rpm on startup don't pass 1k,i know I will have the bad condition. If it goes past 1k around 1300 it's good and smooth. The car starts before 1k but I'm talking about the needle moving after the engine starts. Next will check timing. Starting to think maybe it's my speed reference sensors. For example, it doesn't see TDC properly and gives the rough condition, then a restart would correct it. I also have a fuel pump waiting for parts to install, maybe a voltage drop at pump on start not supply enough fuel? Would you mind sharing your config, so that I can compare. Maybe a setting I did messed things up or was incorrectly set when it was sent to me. I appreciate your help on this. 🤙🏽I'll post a vid of both conditions and logs.

Req_fuel is set at 8. And lambda for idle is at 1. Here's a video of the conditions.

Check your throttle position sensor and throttle cable adjustment. Even with all of the glare and blurriness during the tablet screen shots, I can see that the VEMS cannot tell the TPS is seeing idle; if the idle switch circuit is good, the "IAC ON" icon at the top would light up green at idle with the throttle closed all the way. Therefore it is likely that the idle control valve is sometimes playing a role in the engine running, sometimes not, and could even be sticking open.

If I am totally off base and/or you have deleted the idle control valve (or even if not), the culprit is likely an intermittent vacuum leak. Smoke test the entirety of the intake and boost circuits. Check your bypass/recirc/"blowoff" valve. It could be that it is hanging open intermittently and causing a lean condition.

Real sunny day today, sorry. The tablet connects via Bluetooth and it takes a while to reconnect. I'll recheck my Tps, it's a Ford Tps on a 65mm throttle body and cable tension. If you read back on my old posts this happened also on the stock TB and TPS, I swapped it out hoping to fix. The issue, but it still persists. I've deleted idle control valve and my bov is an Hks ssq that vents to atmosphere. Will pressure test and smoke test vacuum system. Since it's running lean, could it possible be fuel system related? Like a voltage drop causing fuel pump or injectors to run lean? Battery is 5 years old. You can hear it when I'm gonna get a good start/tune cause the engine revs higher. And I've never had this issue on vitesse maf/piggyback setup with an idle control valve. Thanks for your input.

I have had same issue since I got VEMS. Usually by second or 3rd start it runs properly. I am wondering if it's either voltage related or maybe sensor gap on speed and ref sensors. Car runs great when it clicks over and is clean. I have 3 br, FPR, 80lb injectors, reasonably new fuel filter, so doubtful it is fuel delivery. I have also removed ICV now, no change, although car idels properly when I get clean start. Nothing I have done has made any difference so have just got used to having to start it a few timees. Very annoying considering how well it runs when its running cleanly

I may be mistaken, but I think Peep told me VEMS doesn't rely on the idle circuit in the standard TPS.

Surprised to know that someone else with the same issue. What have you done so far to correct issue?I thought about speed reference sensor gap and quality. Then interference and kickback comes to mind. Did you notch the bell housing by any chance or have the older type speed reference sensor bracket without the updated sleeve?I've notch my bell housing when I rebuild my motor a few years back and believe I have the bracket without sleeve.

Have you datalogged both scenarios to see if any of the relevant sensor readings are drastically different when running well and not? Looking at Shortyboy's video, it is clearly running lean when it acts up; does yours do the same? In both cases, it could be a sensor ground or other wiring issue that the VEMS is more sensitive to than the factory DME. Make sure your bellhousing grounds are clean and tight; if not this can cause all sorts of strange issues, whether VEMS or bone stock.

Same thing for the power side; check for voltage drops at all of the junction points from the battery (starter, bellhousing, alternator, etc). Make sure the sensors are getting a full 5V reference.

Well, it does if you have it set up for an idle control valve like I do, but it can be set up to run without like these two guys have done.

I supposed it could be something interfering with the speed/ref mark signals, but the odd part is that your engine runs so lean when it acts up (and it sounds/acts lean in your video). The original Bosch Motronic DME has a pretty robust, specialized chipset to process these speed signals, which are quite noisy. The VEMS is one of the few aftermarket ECUs that seems to play nice with this type of inductive/ring gear setup, but it seems like it is perhaps not as tolerant of noise/wiring issues as is the factory DME. How is your speed/reference mark wiring harness on the car? Any crappy aftermarket ignition wires or anything else that could be causing electromagnetic interference (EMI)?

I thought the map Peep supplies uses the ISV but not the TPS idle circuit ?

The only way the factory DME or VEMS ECU knows to activate the ISV is via the idle contact in the TPS.

You might be confusing the fact that jumpering the connector in the engine compartment is no longer used to set base idle---you have to use the VEMS software to deactivate the ISV to do that.

It's not using idle contact in factory TPS, only uses variable resistance part of the TPS (3 pins).

Regarding starting or not, you should observe/log the injector pw when idling and then adjust cranking enrichment and/or starting VE so that it is close to what engine likes when idling. Obviously something is off when it does not want to start, you should check the injector pw then also.

Well I stand corrected, then! Thanks for the clarification. Any particular reason for that?

I have no idea about the VEMS but always wondered if the separate switch was simply a function of limited processing power in the old 8051-driven DME's. With ample processing power, how hard would it be to set a flag in software when the pot reaches the idle area?

I was able to get my low speed and cruize semi dialed in with autotune. As of now I'm not able to decipher how to set up the ebc config in VEMS. If someone could provide a screenshot of their settings I would greatly appreciate it. I plan on running about 16psi. So for now I plan on using a MBC. Also wondering how you guys are using autotune to set fueling under boost? Autotune requires that the cell being tuned remain constant for about ~2 sec. Thanks for the help.

For EBC, you will need to install an aftermarket boost control solenoid like this (factory cycling valve won't work):


My VEMS wasn't configured correctly for EBC when I first got it. Peep changed it for me after I sent a config file to him. Use the factory cycling valve wires, and in the "output channels" menu of VEMStune, set the boost control channel to "Injector output 4 (7A) Grounding, InjGroup." Then set your boost curve how you want it. The PID boost control works great.

Let me know if you need advice about plumbing the boost control solenoid (I assume you have a dual-port wastegate).

Yes, I have the solenoid valve and Tial 38. Had it all installed and plumbed. My problem is that the setting menu in VEMS are beyond my comprehension. So for now I'm using a MBC.

Here ya go. Make sure you change the boost control output setting to injector output 4 as I posted above.

Then set your boost control target vs RPM table and have fun! The PID boost control quickly learns the optimum boost curve.


https://cimg5.ibsrv.net/gimg/rennlis...8cb882c9b.jpeg

Thanks!!
Had to put off messing around with the car today, found a screw in the work truck tire. Heading out to see the Thunderbirds off the coast of Huntington Beach

Hi Peep,

My car runs great, when it gets a good start But it takes a few cranks to get that good start. If I run the car on a bad start, it seems like it's running lean. Please see video I posted will try and get better videos of problem. Another person is also experiencing the same exact problem as me.

Sorry for slow response. I checked mine this morning. Normally I just restart until it runs properly, but today I waited till wideband had woken up and it was lean. After a restart it was fine AFR's normal, and purring. I did a few trips today and at one point it was just a little off, I didn't think to check AFR's but noticed at lower revs it bucked , but once revs were a little higher, it was fine

Again, you guys need to datalog the poor running conditions and compare them to when it is running well. Post screenshots here, or send log files. We can help!

Got some logs. Its labeled good start and bad start. You can see the bad start is lean across the whole tune. I got a good start, ran the car until warm up. then shut down and restarted, then got the bad start condition logs.

Just to bump the thread, my car is running great. Low speed driveability is damn near perfect. I still need to get the boost control situation figured out. Drove around for a couple of hours today with the MBC I have currently installed and hit overboost protection, it's like throwing out the anchor.

What type of injectors is everyone using? I know that vems can use low and high Z injectors but I'm currently using low Z and was wondering if it could be the culprit of my issue?
​​

I am using Gen IV High Z 80# injectors that came with Rogue M tune, So I don't think its the injectors.

I'm running high impedance dekas, but was running the oem low z with the PWM card channels. I have volvo injectors if you want to try do start and idle tests with them. I had them running when I was in NA mode.

Hmm. Back to the drawing board. Do you know if you have FAE or OEM Bosch speed reference sensors? And if you have the updated bracket with sleeve?

Thanks but I found out that one of my injectors were bad and swapped in a spare set. I also didn't know the set I had in was 72#s, I thought they were 55#s. And the set I put in I thought were 52# are actually less, 52#s at 3.8bar. I'll just retune it for now while I save for a set of deka high Z 80#s. Regarding the pwm channels, do you need to change it when switching from high Z to low?

You need to either enable or disable PWM in Base Setup -> Injector Settings (For high-Z: Injectors PWM DC 100%, Injector PWM Peak Time 25.5 ms, for Low-Z: Injectors PWM DC 35%, Injector PWM Peak Time 1.0 ms).
And also adjust req_fuel due to the flow rate difference in Base Setup -> Engine Setup.

Thanks Peep, Ive adjusted req_fuel for the time being and did some auto tuning. Would these fit?https://fuelinjectorconnection.com/c...ies-80-lb-deka I noticed they have another same model different height at 60mm. Thanks, BTW, waiting for email instructions for EBC solenoid install and settings. Im also going to replace and re-gap speed reference sensors with OEM bosch.

60mm overall height won't wit.

Hi Wayne,

I just checked out your log files, and there definitely seems to be some sort of trigger/interference issue with the engine speed sensor(s).

I attached screen shots below---in the "good start" file, you can clearly see that the engine is "idling" at about 2000 RPM, while in the "bad start" file, the indicated RPM is about half of that, all across the board.

The erratic "worst knock value" reading (jagged black lines all across the top field) also seems to indicate some sort of interference. In my own experience with VEMS, this "worst knock value" reading seems to spike when there is an ignition misfire (I still have the factory coil, can't keep up, have Peep's wasted spark setup on the shelf).

I would start by cleaning the bellhousing grounds, followed by checking the speed/reference mark sensor condition and gaps. Take a close look at the engine side of the speed/ref mark harness, and consider replacing it with the Lindsey repair harness section.

If all of the above seems OK, it could be something with the ignition system itself----maybe crappy ignition wires causing electromagnetic interference with the speed/ref sensors? Non-resistor spark plugs?

"Good" start:


https://cimg1.ibsrv.net/gimg/rennlis...baa652a829.png

"Bad" start:


https://cimg5.ibsrv.net/gimg/rennlis...212e548815.png

What boost level is your MBC set to?

Go into the "Base Setup">"Rev Limit/Overrun/Safety" menu and see what your overboost fuel cut is set to. I left mine at the default value of 250 kPa, which is 1.5 bar of boost pressure (~22 PSI). If this is set below your target boost level, then you can raise it. And yes, a hard fuel cut is like hitting a wall!

Wow nice work Chris. I can't see any of that looking at those screen shots.
I have LR harness, and know grounds are good, but I only mention that for anyone who reads this down the track. I have all the gear ready to go for wasted spark and it sound like that will fix the issue ?.
Cheers

I'm not sure that installing the wasted spark system will be the cure----Wayne has this installed and seems to have the same intermittent issue. Post some datalogs here of good and bad running and we can see if it is the same.

In Wayne's case, it appears that something is interfering with the engine speed sensor output. I am by no means an electronics expert, but I do know that anything electrically "noisy" like a spark plug wire/ignition coil, noisy alternator, or a fuel injector is capable of causing interference with more sensitive circuits like the Bosch variable reluctance (VR) speed sensors. The original Bosch ML3.1 DME had a custom-designed chip to convert the analog 130+1 signal into a usable digital square wave, and it was quite robust. While the VEMS seems to be one of the few aftermarket ECUs capable of working with the 130+1 setup, it does seem to be much more sensitive to interference and/or wiring/ground issues, sensor air gap, etc. A conventional 60-2 or 36-1 tooth setup is simply much more forgiving in this aspect.

Proper diagnosis of the issue would require an oscilloscope and experimentation with isolating any noise sources, but we can start by getting some reliable data in the form of logs and going from there.

First thing to do is to get the rpm signal clean! It absolutely should not read 2000 rpm at any state while idling, I would suggest saving your config, then upload an unaltered vemscfg for starters (I can send you one).

Hi Peep,

Any insight as to why the "worst knock value" signal would be so erratic in Wayne's car? It obviously seems to be some sort of interference, but why would it show up in this particular value? Like I mentioned above, this reading seems to spike if/when I have an ignition misfire under heavy load with the original coil. What does this particular value actually represent?

I haven't had any issues----my VEMS-equipped 951 runs great and always has since I installed the system (now almost 3,000 miles on it). But it seems like several of you are having the exact same issue, so if we put our heads together we can likely figure it out. Make a datalog of the engine starting/running well, and try to log the same conditions after a "bad" start and post them here. Based on Wayne's logs (see screenshots above), it seems like something is interfering with the speed sensor/RPM signal. The reason that the AFR is different in both states is that when the VEMS thinks that it is running at 2000 RPM instead of 1000 RPM, it simply changes to a different cell in the VE table where it happens to run leaner in the case of Wayne's VE table (which looks to be an un-tuned base map for the most part).

So don't read too much into the AFR----log all of the relevant inputs and see what is different when the engine starts runs well and not. It seems like many of you guys are not taking advantage of the logging features of the VEMS; use it and analyze it before making any changes or replacing parts. If you don't know what you are looking at, post the files and/or screenshots here and we can try to help!

It shouldn't be idling at 2000rpm, that was either me cruising around or revving it up to 2000rpm. But clearly something is wrong when the bad start happens. Same sputtering as mentioned. Might be interference with speed sensors on start, like starter kick back. I have the high Torque starter and I'm not sure if I have updated shielded speed sensor bracket.

Do you have the updated speed sensor bracket with shielding?

I also have a hi torque starter and remember that at one stage my former mechanic removing the starter motor shield, as a bolt had stripped or something and couldn't be refitted without rattling. I don't have an updated sensor bracket.AFIK. I haven't played with making logs yet, so not sure I can do that, plus barely driving car ATM. I still need to do a proper autotune, but haven't had the chance yet. Maybe later in the week. I may be able to d a screen shot though.

Interesting. The neat thing about looking at someone else's VEMS log file is that you can also go through all of the menus and check their config and tune. In doing so, I noticed that Wayne's VE table appears to be very "untuned," looks to be some sort of base map. Maybe he loaded another base map/config while attempting to solve his starting issue?

I was under the impression that you guys all had the VEMS for a bit and at least had the basic low-load tune dialed in. I am hesitant to suggest any kind of tuning if there is potentially a fundamental trigger/interference issue, but if the the base VE map is completely untouched and there are huge gaps between cells, I could see certain engine conditions/temperatures causing an intermittent low/lean idle, especially if the idle control valve is deleted.

Do an autotune at idle and at low/light loads just cruising around. You won't do any damage to your engine if you stay off the boost. Once that is dialed in and the throttle response and idle are crisp, then go back and see if your intermittent bad start issue persists. It may be that all of you guys having this issue are freaked out by it and unwilling to go any further with basic tuning, which is understandable.

However, I still do not like the look of the erratic "worst knock value" reading on both of Wayne's logs, that seems to indicate some sort of interference.

Also, making a log file is easy: start the engine with the VEMS software running, then click "LOG" at the upper left of the menu bar. The icon will change from red to green to indicate that it is logging, and will do so indefinitely until you click "LOG" again or the power is cut from the ECU. The file will automatically be saved with a date/time stamp in the filename. Try to log a good start and a bad start under as similar of conditions as possible (let it idle for a bit, rev the engine in a similar manner, go for a short drive around the block, etc).

Hope this helps.

Chris, although I have had VEMS for a while I also had another issue I was chasing. I ended up having to get head redone. So I am starting from scratch. I loaded the original config from Peep, and today I adjusted req_fuel and found 5.6ms got a 14.7 when warm. I then went out and did an autotune prior to seeing your post above, so still no log. This is the screenshot of autotune. I wanted to keep going but Melbourne cup was about to race, so only got about 30 mins of driving in. This is obviously a clean start.

https://cimg9.ibsrv.net/gimg/rennlis...934cf7a13.jpeg

You're correct regarding my tune. Placed another set of injectors and just tuned area under boost. I also performed auto tune with good start condition.i too haven't had much time to do much troubleshooting.i had mine in since January and just got used to it and just restarted the car a few times till i got a good start. I thought I was a special case but it doesn't seem so now. Good to know that we both have high Torque starter and not the updated shield bracket. When I have the time, I'm gonna swap out the starter for the stock unit and do more testing.

Chris, worst knock value spiking is probably some noise, in any case rpm reading should be steady not jumping up and down..
When you see worst knock value smallish spikes (that reflect real world movement) it can be misfire in fact. You can use knock chip as measuring individual power of the cylinders, just choose Individual power in knock input menu. Then you can see in the logs the misfires (power going down on cylinder) and also knocking (cylinder power spiking up). It is calculated by CPU as it is known how much time it takes at each rpm for the crank to reach TDC (ref pin is the reference as the name suggesting) and it is displayed as graph in the log.

I shot out to grab something before store closed and I got a slight bad start on way back. It got worse and that's when I noticed volt gauge was jumping around wildly up around 14+ volts. It is normally rock solid around 12.5 - 13and even with system cranked it only drops slightest bit on bass hits. Alternator is brand new 200amp unit, but the regulator has the issue wear it needs to be revved on start to kick in Alt. Anyway wasn't data logging so only an observation.

Hello! I just had a LR push clutch installed by a local mechanic and now my 951 will not start. Prior to the throwout bearing failure, the car was running well on Peep PnP VEMS. I am hypothesizing that the issue lies with the reference sensors, so I attached a reference sensor recording from the car attempting to start. I also attached the VEMS log file for the trigger log file form the start attempt. The attachments are located below the pictures. I am hoping that this will provide some insight as to what went wrong. Please let me know if I can provide anything else to help diagnose my issue!
https://cimg0.ibsrv.net/gimg/rennlis...7a3cb0a9c4.jpg

https://cimg6.ibsrv.net/gimg/rennlis...d8c2b635a5.png
Attachment 1295808
Attachment 1295809

I verified the reference sensors were attached to the correct wiring harness connectors but have not verified the sensor gap. Attached another longer starting attempt with the VEMS and Triggerlog.
Attachment 1295814

How do you access the trigger log function of the VEMS software?

I don't see any sort of voltage scale in your screenshots, but I would start by checking the speed/reference mark sensor gaps. It is possible that the mechanic who worked on the car removed the adjustable sensor bracket instead of pulling the individual sensors before the bell housing was removed.

The trigger log function is in the tools menu. Checking the sensor gap will be difficult without an old sensor to glue a washer to. I do believe that it is possible that the sensor bracket was not reinstalled with the correct gap but I was hoping to verify that this is the primary issue prior to ordering a used sensor to use to check the gap.

Just use Elmer's glue or similar on one of the existing sensors, will stick enough for using a washer to set the gap but is easy to clean off.

Thank you for the idea, unfortunately that presents too much risk for losing custody of the washer in the clutch housing.
Does anybody have a trigger logs from a running 951 to compare to or see any issues with the logs I posted?

Ah. Good idea.
I had to switch to the LR Spec Steel flywheel with the push clutch setup: https://www.lindseyracing.com/LR/Porsche/SP-44S-2.html

Oh no! I did not realize that would cause an issue. How did others resolve this?
Can you reply with any additional information, posts, or threads you have related to this issue?

Here is the thread I have found so far:
https://rennlist.com/forums/944-turb...startup-3.html

Wow. That is very unfortunate; I don't believe the LR push clutch setup will work with that flywheel. I saw on the other post, they were attempting to resolve the trigger with a crankshaft trigger here: http://www.clewett.com/index.php?mai...products_id=30
Any thoughts on that idea?

Got it! That's the same one that I found. I'll be placing an order as soon as I get confirmation on this plan with Peep.

That's too bad about the Spec flywheel, but good that NCLA951 caught the issue before you wasted time attempting to set sensor gaps, etc . . . . I wonder what the issue is. Does the Spec flywheel not have a provision for the reference mark sensor stud or is it in the wrong place? I feel that if the starter ring gear were not located correctly you would have starter engagement issues . . . .

Anyway, for posterity's sake I took some screenshots of VEMS trigger logs of my 951 (with stock flywheel and speed/reference sensors) cranking and at cold idle, pictured below (might not have caught the slower starting cranking speed before the engine fired, they look very similar). Maybe they will help someone else with the same setup who is chasing a starting issue.


https://cimg0.ibsrv.net/gimg/rennlis...ab6628c453.png

https://cimg3.ibsrv.net/gimg/rennlis...df0c69f9c9.png

Here is a couple of seconds of cranking, then turning the key off before the engine fired. The entire cranking period looked like the first half of the graph, nice and evenly spaced and sized. nick.cfrancis's trigger log shots show erratically sized and spaced trigger events while cranking.


https://cimg6.ibsrv.net/gimg/rennlis...e1cefd5f5b.png

In my educated opinion, the issue is that the 3 scalloped lobes on the spec flywheel will trip the sensor under certain conditions.

Droops83: Thank you for posting screenshots of your trigger logs. I can definitely see my issue when comparing to your clean trigger log.
I reached out to David Ewell via email and received confirmation today that he was able to resolve the startup issue with the LR Spec scalloped flywheel on a 951 using the 36-1 Clewett Trigger wheel.
https://rennlist.com/forums/944-turb...startup-3.html
Peep also confirmed that this idea would work.
David did mention a potential issue regarding the wiring of the Clewett trigger sensor requiring modifications to the VEMS board. I am hoping to avoid this issue by utilizing the stock reference sensor in the Clewett bracket described here:
https://rennlist.com/forums/944-turb...r-wheel-3.html
I also reached out to Lindsey Racing regarding regarding my issue. Dave explained that in the past they had issues with the spec scalloped flywheel causing false triggers.
https://www.lindseyracing.com/LR/Parts/SP44S-FIX.html
This issue was remedied in May 2017 with a smaller diameter flywheel from Spec. Dave proposed the idea of raising the studs on the flywheel and backing the sensor out to compensate in an effort to avoid detecting the scallops on the flywheel. Given that the flywheel is already installed, there would be great difficulty in setting the studs to an equal height and backing the reference sensor out to correspond to the change. I am going to go with the Clewett 36-1 trigger wheel at this point. Unfortunately, even though Clewett is based only a few miles away from my location, they are out of the office for the Thanksgiving holiday, so I will be on hold until they return on Monday, 26 November.

Is this just with VEMS, or have people had issues with the factory DME/Spec flywheel as well? My (somewhat) educated opinion is that the whatever the VEMS uses to interpret the AC waveform generated by the factory VR sensors simply isn't as resilient as that in the factory DME chipset, and it is therefore more prone to interference. Though it would take someone much more versed in electronics than I to explain exactly why!

Almost done with installing the Clewett trigger wheel and mount but I am somewhat lost on the wiring. Given the reference sensor harness below, which pin is the sensor, ground, and shield?
https://cimg2.ibsrv.net/gimg/rennlis...de44a16fc8.jpg

That's exactly what I'm running in my SC set-up blue 80# dekas.....exactly what Peep said for settings :)

Also, where would you recommend I place the missing tooth with the motor at TDC? The Clewett instructions only contain setup instructions for the 60-2 wheel.
Here is my current configuration. With the motor at TDC the sensor is reading tooth 7-8 after the missing tooth.https://cimg6.ibsrv.net/gimg/rennlis...929faec056.jpg

I don't know much about trigger wheels set ups, but if you research it say with MegaSquirt or other 36-1 set-ups. The missing tooth is typically placed so many teeth past the sensor at TDC and 8 is typical given the amount of rotation. Looks like you are limited to the choices given the set fasteners and pulley positions. That looks right to me to try as long as the config matches.

This weekend I installed the Clewett trigger wheel and Clewett magnetic sensor adapted to the factory harness using the following pinout:
https://cimg1.ibsrv.net/gimg/rennlis...7bfc26733e.jpg

I'm not sure how to adjust the trigger settings in VEMS so I emailed Peep on Saturday to ask for help with changing the config but have not recieved a reply. Does anybody know if he is on vacation or out of the office?
Edit: Just got a reply from Peep. I'll be trying to configure it this evening after picking up a timing light.

When using autotune is it nessecary to set EGO closed loop control lean and rich limits to 0%?​​​​​​​
https://cimg9.ibsrv.net/gimg/rennlis...dbdee08eab.png

Yes

Thanks! I was just verifying as I saw it on a forum post here on rennlist but didn't see it anywhere in the VEMS help documentation.

Happy New Years to all. Gonna start work on the 951 again. Gonna try and solve this issue once and for all. Gonna start by checking all grounds, check grounds and power at vems ecu, and swap out hi Torque starter for stock unit.. The move on to sensors, gap and updated bracket with sleeve. Did the 60-1 clewet setup solve the problem??

The 60-1 clewett setup did solve the scalloped flywheel problem.
One side effect I am experiencing is that if the vehicle is turned off, the cooling fans turn on, and then the key is turned back on, vems will read the cooling fan as the trigger wheel and believe the motor is running at around 600rpm and start firing the injectors. The only workaround I have found is to wait for the fans to turn off before turn the key back on. The risk with this workaround is that when my wife was driving, she stalled the car in stop and go traffic, didn't hear the fans running, and tried to restart it. This flooded the motor and took about 5 minutes to get started again. I was just thankful it didn't happen in the middle of the freeway.

That's not good! Is there any way you could move the sensor bracket away from the fan motor(s)? You might even want to consider thinner, modern cooling cooling fans (SPAL, etc), to keep the motors and their magnetic fields further away from the crank sensor. This would have the side benefit of saving weight and creating a bit more space for working on the front of the engine.

In hindsight it seems like the TTV 60-1 flywheel is the ultimate solution for those having trigger issues with the VEMS, but since you already invested in the SPEC setup and installed it, you'll have to make it work!

EDIT: You should contact Clewett and see if there is a Hall sensor available for this setup to replace the standard inductive sensor. If so, a simple wiring change and VEMS setting changes and you should be good to go.

The sensor bracket can't be moved with the Clewett setup. A different fan setup is probably the best option to prevent the interference. I couldn't find much information on the forums about a plug and play aftermarket fan and shroud option. I will contact Wizard Cooling to see if they have any fan+shroud setups available for the 944.
The TTV 60-1 flywheel setup would be the best solution but unfortunately, I am already committed with the push clutch.
Clewett does make a hall sensor and he said that it would solve my issue. I contacted Peep for details on switching from a magnetic sensor to a hall sensor and he stated that it would require VEMS internal modifications as well as wiring modifications to work.

Maybe try a shielding of some sort? In between the sensor and fans.

As for my attempt. I've clean grounds, replaced. Battery ( mine was dead after a few weeks sitting and it's 6 years old.) swapped injectors and replaced fuel pump with bosch 044, fuel filter, inlet and outlet lines and sock filter. I also noticed some clicking in the ecu when I'm having the rough start/running issue. Next, check gap of s/r sensors on flywheel , replace sensors and updated bracket. I've also replaced wasted spark coil and plugs in the past, trying to fix issue.

Swapped out hi Torque starter for stock unit and still same issue. Ohm out speed reference sensors and both around 924 ohms, will grab an oscilloscope to check output but I also did a trigger log of both conditions. Please see attached pictures of both trigger log png, marked good and bad. Also uploaded two videos of engine running and video on vems screen. Good condition, lambda is good, afr is good and idles a bit below 1k. Bad condition lambda is not spot on, afr is lean and idles below 900. You can clearly hear the difference in exhaust ton, even though it's straight pipe, you can hear popping in the bad video. I've got 72# delphi hi impedance injectors, saving up for deka 80s. New fuel pump filter, tank outlet filter, bosch 044 fuel pump, CEP fuel rail with automotion adj. Fuel pressure regulator ( this is probably the next step in troubleshooting, this or the fuel damper?) anyways check out the pics and videos and share your thoughts. I'm inclined to send my ecu in for repair also after burning the ignition transistors, but this issue I'm having was before I blew them. Or if someone is willing to test my ecu, I'll send it and see if issues follow. Thanks in advance.
https://cimg5.ibsrv.net/gimg/rennlis...412da81b0f.png
https://cimg6.ibsrv.net/gimg/rennlis...75f89dc1cd.png
https://cimg7.ibsrv.net/gimg/rennlis...c59ef5cdc2.png
https://cimg8.ibsrv.net/gimg/rennlis...698e529580.png
​​​​

​​​​​​GOOD
https://youtu.be/xE4pX3-Ld6w

BAD
https://youtu.be/C0zPIP8QQQ8

Forgot to post these pictures. It's the VEMS ecu main board. As you can see I burned a trace when u blew the ignition transistor. I placed a jumper wire to fix it. I also notice a resistor out of place, wondering if anyone else's is like this also.

https://cimg5.ibsrv.net/gimg/rennlis...86474f6167.jpg
https://cimg4.ibsrv.net/gimg/rennlis...f62d491434.jpg
https://cimg6.ibsrv.net/gimg/rennlis...fff418537e.jpg
https://cimg7.ibsrv.net/gimg/rennlis...f3e5d607d3.jpg

I have no Idea, but could that be an improperly soldered 'cold' joint. On my board (left) it doesn't look too good either, but no issues so far.

https://cimg7.ibsrv.net/gimg/rennlis...8c7de5c767.jpg

Gonna go at the whole board with a microscope and soldering iron.

More testing today. And although I have a set of new Facet brand speed reference sensors coming in, I think the ones I have in right now are good. They ohm out good at 924ohms, when using a volt meter, I get 2.9v at pins 8/27 when I crank. It over. Took off sensors and found a bit of metal on the speed sensor, cleaned it off, installed still same issue. Used an oscilloscope to measure waveform of speed sensor and I'm getting more than 4v, almost 10v peak to peak. ( see attached picture ) . So now I'm back to thinking what else to test. I tested my aeromotive adjustable fpr and according to the gauge, on start up, fuel prime, I get about 30psi, about 38psi when running, 43.5psi with vac line off. But when I shut off engine, the pressure bleeds off very quickly. No more than a minute or two and pressure goes to zero. Not sure what else to test. Any thoughts?
https://cimg8.ibsrv.net/gimg/rennlis...366487813d.jpg

That is a definite fail for a solder joint. Any idea what the circuit is part of?

should be the WBO2 heater. Apart from that, In my case the connection to the WBO2 sensor was not wired correctly. I noticed that as I could not calibrate the sensor.

https://cimg9.ibsrv.net/gimg/rennlis...910cd777a1.jpg

30psi when running at idle on a stock fuel system is ok. Leak test minimum pressure after 20 min --> 14.5 psi.

So it might be my Aeromotive AFPR.It goes to zero in a few minutes. Had it a few years as part of my CEP billet fuel rail. Will swap it out and report back. Thanks

Swapped out speed reference sensors with new facet brand and swapped out aeromotive adjustable fuel pressure regulator....same issue. Have not regapped sensors due to good output on oscilloscope. Running out of ideas. My next guess is ignition switch. Since there's contacts for starter and dme/ecu there if they're dirty or corroded, a few clicks on and off would clean the contacts thus making a good start condition. Off to spend more money and wait another week or two for it to come in. Anyone interested in a good used set of reference sensors and adjustable fpr?

Put the oscilloscope on the reference sensor and got 4v, almost same as speed sensor.
​​​​​​
https://cimg2.ibsrv.net/gimg/rennlis...1f30707604.jpg

Shortboy, I haven't been following along and not sure your symptoms, but those pictures of the circuit board are pretty ugly. In addition to toasted traces, it looks like it overheated to the point of melting solder and dislodging components, and looks like that one TO-220 component overheated pretty badly -- (MOFSET?) second from the right on the lower side (did you test it?). Also, as you work to patch the visible damage, do you know if this is a 2 layer board and/or if there are also internal trace layers? If this is a 4+ layer board, then without a schematic and lots of multimeter time on your hands, it would be virtually impossible to know if the board is still ok.

Separately, it seems like Aeromotive FPR and replacement rails have caused issues on these cars in the past, despite good pressure etc. I've never investigated why and/or if there's true cause and effect, but people have reported flat spots and hiccups, etc.

​​​​​I burnt up the ignition transistors, which from what I read is a common occurrence. I'm not sure if other components was damaged in the process. Basically, the issue I'm having and at least two other people, is.. We would need to start the car a few times to get a "good running "start condition. When it doesn't start good, it idles poorly, lean afr and sputters. So far, I've replaced fuel filters, bosch 044 fuel pump, injectors adjustable fpr, reference sensors, ignition coil pack ( wasted spark ) , starter, battery, tps ( aftermarket Ford tps ), clean up all grounds. I just ordered a new ignition switch is hopes to solve issue. My thinking is that the contact in the switch is ether poor or dirty, and the few times it takes to turn the car off and on, cleans the contacts thus making a good start/connection to starter. You can clearly know when I have a good start because when cranking over, if it goes past 1300-1400 rpm it's a good start, if it's anything below it will idle or run poorly. ive uploaded some videos on YouTube. Any help or ideas would be greatly appreciated. Will try and inspect all solder components. I hope Peep can chime in regarding the components and layers of board. Looks like a dual layer board. I replaced the traced with a jumper wire ( red)

It is not a common occurrance to burn out FETs!!! When something like that happens it means that there has been constant GND signal to the coil while ignition is on, usually happens when people update firmware without disconnecting coil and injectors (even though manual strictly says to do that).

Hi Peep,

Any insight on my issues after what ive tested and replaced so far?

Peep, if all is well with your installed map and autotune, is there reason to update the firmware?

Hopefully I stumbled across some progress in solving this issue. Began to wonder if it's a voltage drop issue at ecu or reference sensors. I have a few accessory electronics that tie into the same +positive and - negative lines of the vems ecu. I began to unplug them and the frequency of good starts got to 2 starts instead of 4-5. To further test this theory, I had fans running on high and could not get a good start condition. No matter how many starts it's was rough and lean. I've cleaned the grounds at bell housing and chassis. I have new cable lines I got from lart that were sharky cables with the quest alternator and hi torque starter. If this voltage drop is the issue, I'm hoping the new ignition switch will cure it. Any other ideas? I might have a voltage drop somewhere else also, maybe seats? Gonna unplug radio and amp. Also, for my reference sensors, I have a new Lindsey Racing harness.

There is no reason. Two of the cases where transistors were blown were caused by updating firmware in attempt to dagnose completely unrelated issues and the updating was performed without consulting with me. That's why it is important to listen to the suggestions and do things written in the instructions not by guesswork.

Fans running at high can cause interference, has happened sometimes (but never with sensors in factory locations, only when using front crank trigger and that is also rare).

If you look at the other side of the board then you see that the FET leg of the WBO2 heater is soldered. They are soldered from the bottom.

Have you actually tested for voltage drops using a volt meter?

Will be testing with a multi meter once the weather permits. Thanks. I also have the updated reference sensor bracket with sleeve coming in also.

Had some time to work on the car. Went to install updated bracket with sleeve and found out that I already had one installed. Went ahead and regapped speed sensor using tape method and made it closer to 0.7 gap. After testing, still same issue. After my fiasco with ignition switch swap out. I found a replacement ignition switch pigtail to the fuse box, soldered it in place. Installed new ignition switch. Still same issue. Voltage drop theory out the door as I have constant 12v+ when cracking at ecu. No clue what to test next. Probably pick up a new set of high impedance injectors. Maybe gap the sensor even closer. Does anyone know at what gap does the reference sensor come in contact with eh pin?

I just re-read this and noticed that you had tapped into power AND ground of the VEMS. The ECU needs its own dedicated ground circuits (with separate branches for sensor circuits and noisy devices like fuel injectors), with nothing else tapped into them. This could be your issue, especially if you are tapped into the sensor ground circuit.

Yup, in the process of removing the Hks ebc that was tied into it. That's basically the only thing tied in, that and a use charger, will move that also. Does anyone have instructions to wire in a boost solenoid? Emailed Peep a few times and no response.

Regarding the speed reference sensors. Curious if anyone you who are having same issues as I am is using the Lindsey Racing Reference Sensor Harness? I noticed the factory has two ground points, one pin of its own and a ground that shares on the oxygen sensor. I'm wondering if adding a supplement wire to the sensor grounds would help.

BTW, I wanna thank everyone in the thread that's been helping me to solve this issue. Much Mahalos 🤙🏽 - Wayne

You wire boost solenoid in place of factory cycling valve, easy.

There are many guys using Lindsay sensor harness, no issues with it. But I would suggest measure connections of the sensor harness just in case, it is possible to reverse the polarity when making the swap. Adding additionl ground wire does nothing good, you should verify that you have ground wires on the bellhousing all in good working order and tightened.

Hi Peep,

Thanks for the reply. Rechecked all grounds with my DMM, from sensor connectors, and bell housing ground, to front chassis grounds to battery grounds to dme grounds, all are 0.2 ohms. Next step will be setting gap closer at reference sensor, that way speed sensor will be as close as possible. Since I'm also tracking a fuel pressure loss issue that's intermittent, the final piece is new siemens deka high impedance injectors. I'll be sure to change pwm settings accordingly. I'll also try to find some type of EMI shielding for starter and sensor cabling.

Regarding boost solenoid, what do input in vems configuration? Thanks again sorry for all the trouble trying to fix this, hoping to fix my issue and others in the future.

New Siemens Deka 80# high z injectors came in from www.fuelinjectionconnection.com swapped them in. No dice still same issue. Readjusted gap again on speed sensor. Same. Does anyone know how close I can get on the reference sensor before the speed sensor touches the starter ring gear? I read that there is a high spot on the ring gear which the speed sensor reads off of. I'm beginning to think I'm just chasing my tail. I'm wonder if the whole issue is the notched bell housing. Regardless of the updated sleeve. Who here has the notched bell housing? I'm hoping that an aftermarket flywheel is not the only solution. Since someone is having issues with the clewet front mount crank sensor.

Peep,
Is it possible to use a cam sensor for a trigger? Also, does vems use both speed and reference sensors?
​​

Just curious if you're using the spacer (item 32) on the speed sensor? That spacer has the effect of pushing the reference sensor closer to the stud on the flywheel.
I updated to this bracket recently on my 85 NA racecar after notching the bell housing and did not put the spacer in. My car starts now w/o all the kickback drama but I might add the spacer and check for clearance to reference sensor just because I'd like to know.

https://cimg5.ibsrv.net/gimg/rennlis...35671cf8a4.jpg

​​​​​​I have the spacer. And per your experience, I removed it and regapped the sensors again. Just to be sure the reference sensor didn't hit, I found the pin and made sure it didn't hit, while I hand cranked the motor over back and forth. I also swapped my new facet sensors to my older fae sensors, same issue. Gonna probably buy bosch sensors next. I also rerouted the battery cables to starter and made them farther from sensors. The only other thing I can think of causing interference with sensors is the cable from battery to alternator. But there isn't another way to route it. There are a few here with the Nissan alternator mod and no issues. Maybe it's my ice shark cable kit? Looks like the shielding use is not for emi or rf but for heat shielding. I want to exhaust all options before going to a crank trigger route. Appreciate your input.

Also, can someone with a turbo config share theirs, I want to compare a few things. Thanks

Posted my config if someone wants to check it out.https://untsorce.cool/metric/?mid=&w...=1549771575609https://untsorce.cool/metric/?mid=cd...=1549771575612https://untsorce.cool/metric/?mid=90...=1549771575614https://untsorce.cool/metric/?mid=6a...=1549771575617https://untsorce.cool/metric/?mid=90...=1549771576546https://untsorce.cool/metric/?mid=cd...=1549771607719https://untsorce.cool/metric/?mid=6a...=1549771607723

@EGO Closed Loop Control Settings
Why is your Lean limit and Rich limit = 0 % ?

F1 (Help)
Lean Limit
Allows this percent of enleanment.
Rich Limit
Allows this percent of enrichment.

I assume 0 % could disable enleanment/enrichment .

Did you already try 5% and step size 1

I was told by Peep, when using auto tune / ve analyzer to set those settings to 0%.

ok, interesting auto tuned VE Table in your config. No values above 55. Seems weird to me.
https://cimg6.ibsrv.net/gimg/rennlis...1daa01f301.jpg

NCLA951's table is pretty different, second post in this tread:
https://rennlist.com/forums/944-turb...l#post14090010

Req_fuel fuel is set at 9.5. Have not made alot of few WOT runs, cause of bad weather here. So I manually input alot of tables based on 2 full wot pulls.

Btw, does everyone's tach bounce when starting the car like with the stock dme / Klr? Mine doesn't, just wondering.

Um, how did you arrive at a Req_fuel value of 9.5? I have the same Siemens-Deka 830CC injectors that you do, and my Req_fuel is set at 5.0. I think this explains why your VE table has such strangely low values. Here's a copy-and-paste from the VEMS webpage:

Basic req_fuel calculation:
req_fuel = 6.49 * (D / N / I)
D engine displacement (cc)
N number of injectors (eg. 4 for a 4 cyl port injection)
I injector flowrate (cc/min)
This is a simplified equation, if you are interested in the reasoning behind it you can read more here: http://www.vems.hu/wiki/index.php?pa...njectorOpening

I am not convinced that this is causing your weird starting issue, but it should at least help with tuning.

I initially did that method, using formula, but was later instructed by Peep to adjust Req_fuel until I have spot on lamba at idle. My req fuel is high, be table values are low, I see that everyones req fuel is low, and table values are high. Main point that lamba values are achieved, I think it's ok. I might have a lead on solving this issue. After adjusting and swapping out brackets and sensors, I began to look at pin height of the reference sensor. I noticed it to be too far in. Going to try and get the reference gap as close as possible. So I removed the Allen screw, bought a new one and reset the height with red threadlocker. I might go and find a longer screw for good measure, stock is m6x12, maybe go m6x16. Tried a cap head bolt but it hits the bell housing. I believe the critical sensor is the reference compared to the speed sensor. As it determines TDC.

Hey, you might be onto something here (though I think that several of us have asked if speed AND reference mark sensor gaps were good). The reference mark signal is really only used during cranking to determine when TDC #1 is approaching (at least on the original Motronic system) to enable engine startup. After that, the DME is capable of calculating engine position by counting engine revolutions via the engine speed sensor, thus rendering the reference mark signal redundant while the engine is running. This would explain why your engine runs well and continues to do so if it starts well, or runs crappy and continues to do so until the next "good" start.

Let's hope that is it before you replace any more parts!

Nope, just put her back together and same issue. Speed and reference sensor gaps are very close now, both at 0.8mm or closer. As much as I want to send in my ecu for evaluation, I've had this problem since installing it new. Got a new set of bosch sensors coming in. Swap out Nissan Quest alternator? For a new or stock?Replace my Ice Shark Battery / Alternator / Starter cables? Idk, thoughs?

Have you ever used a timing light to check the timing values at various RPMs and compared values after "good" and "bad" starts, and compared it to what the VEMS thinks it is? If the values are different between the two modes at the same RPM, that would confirm some sort of trigger issue, though it seems like you've tried everything related to that . . . . .

If there is some trigger issue (missing trigger events, noise etc) then it should throw trigger error flag with information which type of trigger error there was. And would be useful to record triggerlog (Tools -> Record triggerlog) for both good and bad starts, that would help to eliminate trigger issues and wasting time if there is no problems.
Also regarding req_fuel, it doesn't matter what it is as it is just a multiplier for VE table and you cannot compare VE table values directly anyway unless engines are identical. You can compare VE table shape but that is also affected by injector dead times etc.

hoping Peep or others here can help me with the validation errors I have.

For some reason IDK the drop down menu for boost control PID channel is not available to make changes?

https://cimg4.ibsrv.net/gimg/rennlis...e743ce6888.png
https://cimg8.ibsrv.net/gimg/rennlis...162068c124.png
https://cimg0.ibsrv.net/gimg/rennlis...adba7683cd.png
https://cimg1.ibsrv.net/gimg/rennlis...697c51b12b.png

How come you have 1.2.38 firmware at all? VEMS is shipped with 1.2.32 and valid config, updating firmware is ONLY for cases when it is really needed (under strict instructions) or person is experienced enough to experiment with firmwares and features.
Unless you plan using some features from 1.2.38 (like analog input multiplexer) I strongly suggest flashing back to 1.2.32 and using your working config.

I have not tried using a timing light yet to verify timing with vems. How would I do this, have not checked timing yet on my 951.

Hi Peep,

see above for my post a while back regarding trigger logs. The smooth or flatter pair is a good start condition, the jagged or wavy pair is bad start. I will also report feedback after bypassing resistor on ECU36 Pin 27.

Yes you are totally right I mistakenly followed the prompting to update the firmware my mistake

I'm a total neophyte when it comes to this my thinking was that just like other electronic devices updating the firmware was a good thing.

I understand you must disable the coil and injectors when doing the firmware update is it enough just to disconnect them in the engine bay?

https://cimg9.ibsrv.net/gimg/rennlis...da0affd830.png
https://cimg1.ibsrv.net/gimg/rennlis...fd0ca24cad.png
Bosch sensors came in today, got them at Rock Auto for $25 each! Installed, but issues still there. Is anyone here using Ice Sharks cable kit, battery, alternator, starter? How about the Nissan Quest alternator? Maybe its a combination of them aftermarket parts Ive got? Hi Torque starter, Ice Shark cable kit and Nissan Quest alternator. Here is a trigger log png and log file, you can clearly see it going lean when i give it gas, ill upload two new videos of good and bad running. Back to the drawing board. Or liquor cabinet.


BAD

GOOD

OK, the one thing I notice from those videos is that the TPS % barely goes up when you open the throttle in the "bad" video----I assume that you're snapping the throttle by the same amount as in the good video? Don't you have some odd throttle body setup? Have you checked that its TPS is good, and that it correlates to the actual throttle opening?

That's all I can tell from the data presented in those short videos, everything else seems normal. If you can get the car to my shop in Santa Barbara, I could fix it :)

In the bad video I'm trying to Rev it but it's so lean and alot of popping. It's a Ford TPS on a 65mm TB

That's what I am saying----if throttle/TPS were OK and something else was causing a lean condition (large vacuum leak and/or lack of fuel), you'd still see the TPS % go up by the normal amount as you snap the throttle. The injector pulse width, MAP, etc look similar between the two videos, but TPS gets up to ~25% in the "good" video, and only about 5-6% in the "bad video."

I could be wrong, but this would be pretty easy to check.

I was revving the throttle the same (at least I think I am) , in both videos. I highly doubt it would be a vac leak or lack of fuel, because in 3 to 5 restarts, I would get a good start condition again
​​. Seems almost like a huge voltage drop, causing settings to change. If you've been following what's been replaced so far. Tank outlet fuel filter, bosch 044 fuel pump, outlet filter and lines, aeromotive fuel pressure regulator, siemens deka injectors. Fae, facet, bosch speed/reference sensors. Ignition switch. If my voltage drop theory is right, then it would either be my battery cables or Nissan quest alternator. Will swap in my stock alternator and wire vems power directly to battery with a 3am fuse. But I'm still stumped at what would cause a good start condition every few starts. I really appreciate the brain storming
​​. Will test your theory once weather let's up. 🤙🏽

I don't think you are understanding the point of my last couple of posts: the outlier is the TPS% data. The point is, your TPS reading should be the same if you are operating the throttle the same level, regardless of whatever else is going on with the engine. For example, if your actual problem was a fuel-related issue, the TPS readings would be the same in both videos, again assuming that you are snapping the throttle at approximately the same rate.

As for your voltage drop theory, there still could be a voltage drop somewhere (5V reference circuit, etc), but the voltage reading at the VEMS display is virtually the same in both videos, which means that the ECU itself is getting the required voltage.

Because of the wonky TPS readings and the fact that you have changed the throttle setup, eliminated the ICV, etc, I would at least eliminate that as a possibility. I am not 100% convinced that this is the problem, but that is the only obvious anomaly in the data that I can see in the short videos provided.

Your MAP reading (video bad running cond.) seems a bit strange to me. TPS 3%, RPM 821, MAP 37. I never get values like this.


https://cimg8.ibsrv.net/gimg/rennlis...47dcd517ca.jpg

Before starting the car, I verified vems is seeing tps correctly by tapping the throttle. Then after starting it, on the bad start, it won't read it correctly, idk if it's cause it's seeing rpm is alot lower than it is but it's obvious that when this condition happens it has to do with speed. Reference sensors no telling vems the proper rotation speed. There is also no trigger errors when the bad condition happens.


Something not communicating right between speed sensors and vems. But every few starts, it gets the correct info. I'm stumped. A local guy who. Has vems on his supercharged n/a has agreed to try out my ecu with his config. We verified our injector and ignition pinouts are the same. As far as alternator theory goes, im getting 14v+ at the battery and 12.6v when car is off. Starter sounds the same every time I crank her over. Just when it catches is different. When I see the tach start below 1100 I know it will be a bad start. If it goes up to 1200rpm, it's always a good start. Injectors flooding cylinder causing slower rotation? I'll grab a tachometer at work to verify speed, just need to find an adjustable timing light to verify timing. And instructions on how to perform it.
​​
​​​​​​
I also assume that this issue the 3 of us are having is some what related. I think it's something we have in common causing this issue. Which is why I brought up Nissan quest alternator and hi Torque starter.

Your speed sensors are totally fine, trigger signals are clean as a whistle.
But TPS should read identical in all conditions, regardless of rpm etc. It HAS TO reflect actual throttle movement.
In the bad running video it is clearly running too lean, you can hear pops from the intake and lambda reads also very lean.

Where have you taken the vacuum signal? It has to be taken from intake plenum, as close as possible. On factory cars KLR boost line has correct signal and no noise but for example factory S2 models have FPR line tapped together with HVAC vacuum and they can occasionally run way rich even in stock form...

Vacuum reading between different engines does not mean much. It depends on way too many variables like cam, cam timing, engine temperature etc etc.

TPS is a new Ford tps on a 65mm throttle body. I calibrated it in the tps calibration screen. When I get bad start condition, it does not respond well, like lag when I press the gas and when it actually revs. On good starts it works perfectly.

Vacuum line has its own port off the intake manifold straight to vems ecu with nothing tied into it.

If there is an extreme lean running condition (large vacuum leak, etc), an engine will barely rev up and respond to throttle input----BUT the TPS reading should be the same with the same throttle opening, regardless of how the engine starts or runs. It's not the TPS reading that lags when the engine is really lean, but how the engine actually responds to the throttle input. As in, you can hold the throttle 50% open (which will read 50% on TPS), but the engine itself will barely rev up. I find it odd that your TPS reading does not go past ~6% or so in that video, even though it seems like you are snapping the throttle at the same rate as in the "good" video.

It will be incredibly easy to verify this. Shoot a "good start" and "bad start" video and in each condition hold the throttle at the same opening (say 25% or 50%) and make sure the TPS reading is the same for each. If this is consistent, you can move on.

https://cimg1.ibsrv.net/gimg/rennlis...6b1fec3b9.jpeg
In other news, I am happy to report that tuning VEMS to pass CA tailpipe emissions with 80# injectors is a snap! Just had to change VE table slightly at the idle, low load range due to the more restrictive "street spec" exhaust :).

It pays to keep your engine bay looking stock. Find the catch can!

this is great news! I figured that it would be possible. How can it not? ..ability to tweak the fueling.
You're the first congratulations!! FYI all the others, CA smog is a PIA

Yes, it's the visual inspection that is a killer. I was a bit worried due to the lack of an airflow meter, Laust vacuum manifold, and the elimination of the branch manifold under the intake----and the stealth catch can mounted in the A/C receiver/dryer spot. I had a co-worker take the car to a place down the street from my shop; he verified that the factory cat was installed, poked around the engine bay a bit, and deemed it OK. Luckily most don't know what they are looking at in terms of '80s German cars. And it is a happy coincidence that I do not like flashy engine bays anyway!

As for the tuning with large injectors, if you have the correct injector dead time settings and are already tuned to stoichiometric at idle/cruise range (which any engine should be), then you are 98% there. It would be a little tougher in an "enhanced" area of CA that tests on a dyno under light load and measures NOx, but if your cat is in good shape that should also be OK.

EDIT: Props to Lindsey Racing for stocking the awesome silicone hose sets for the boost/intake/J-pipe and for the breather hoses. Eliminating the hard pipes underneath the intake manifold makes everything so much tidier!

Completely off subject here but I have installed a VEMS plug/play and need a little help. This is a new N/A motor build for E Production SCCA racing.
It has 12:1 Compression, big cam, solid lifters, 951 injectors, Molnar rods, Wossner pistons, crank mods/scraper, light flywheel, oil exchanger and many other bits.
I used the VEMS last year in a "stock" motor and it worked great. Peep helped me get this motor fired up last week so it's running and getting broken in. :D
My problem is idle while cold and warm and throttle lift. I've set my fuel pressure to a safe level so I don't lean out OR wash down my cylinders. I've also set my cold and warm idle to 1500 and Ign. based idle is disabled.
The urgent problem is that it won't follow the 1500 all the time, especially after I rev to 4000 and lift. The RPMs drop down to dangerous levels. The danger being low oil pressure due to the larger capacity oil exchanger.
I haven't had a chance to drive the car or Dyno yet so Live VE analyzer hasn't run yet. It's not street legal.
So, my question is: For now, how can I keep my RPMs from dropping below my settings?

https://cimg8.ibsrv.net/gimg/rennlis...3ac594c53d.jpg

Will test tps once I get my ecu back. Gave it to another local who has a supercharged n/a with vems. Had him test it and it runs perfectly. Now we know ecu is good. Gonna replace dme temp sensor, oxygen sensor and TPS. Does anyone know what part number oxygen sensor they're using. I'm using a bosch 17014 LSU 4.2, from my old wideband setup, could possibly be the problem. Just want to know what else everyone is using. I ordered mine without an oxygen sensor ( since I had one ) afm to IAT harness, since I rewired it myself coming from a maf. And I wired in my wasted spark coil from my vitesse wasted spark setup, which again, might contribute to my issues. Gonna try and get close to what everyone has. Gonna need wideband part number, and wasted spark coil part number. Much Appreciated.

Possible vacuum leak? Do you have an idle control valve?

Not vacuum and I tested the ICV. I'll be getting the car to a proper dyno tuner soon. Just thought I'd try to get the idle safe before I go. Thanks anyway!

Installed new bosch 17014 LSU 4.2 wideband oxygen sensor, calibrated. Replaced new Ford 3 wire TPS, calibrated. Installed new bosch DME temp sensor. Still same issues. I can confirm that when bad start condition happens, TPS does not display properly or respond properly, like a lag when the gas is pressed to when it actually registers. This is with old tps, new tps and with factory 951 tps. Changed over a while back hoping to fix this issue.
​​​​​​
Next would be IAT and Knock sensors, but a friend of mine pointed out my knock config retard is 0,so no effect would be noticed. And IAT would not change the way it starts.

Looked at bosch wasted spark coil offered on VEMS online shop and it's similar to what I have already. I originally had an MSD 8239 unit, bought a cross referenced similar unit from Autozone, no change when I changed over.

Rechecked wiring on O2 Sensor and everything checks out fine. ( the reason I checked was I soldered connector harness from old wideband setup to vems harness to skip on fishing new wires through firewall.)

Let's start brain storming again guys.

try adjusting the idle bypass screw and throttle stop, maybe recalibrate the TPS.

MAP tuning is very tolerant of vac leaks but if the throttle sensitivity isn't right it can make weird drop-throttle situations in my experience.

also, don't worry about oil pressure.
remember a 944 pump can make 45psi at 900 rpm with hot 20W50 oil.
i ran for 7 years with 2x Setrab 13 row coolers in series and about 10 feet of -10 size lines (a LOT of added oil capacity) and never had oil pressure problems even with 110F ambient temperature.

OK here's a few ideas. How quickly do you move the key to 'start' position? I recommend that you pause in the 'on' position and give the ecu a second to boot up.
Have you used a timing light to verify actual timing? There's nothing like actual data and info to help you make an informed decision.
Do a 'wire wiggle' test. This is done when the engine is running normally at idle, just go around the entire engine bay, wire loom, and dme and gently wiggle the wires, sensors and plugs. If there's a bad connection the engine should sputter or cut out or do something odd. Don't forget to tap on the ecu case in different directions.

Did alot of searching on here and Google. How do I check timing? I've got the speed sensors out and I'm gonna paint the scribe mark on the flywheel. Do I just clamp #1 spark plug wire, point the strobe at the timing mark and verify what I'm reading at idle is the same as vems is reading? Would a incorrect set timing cause the issues I'm having?

In short, yes. The best practice is to set the ignition timing to a fixed level, say 10 or 15 deg BTDC. Then clamp around the #1 plug wire and check your timing. Should match your fixed value. If it's wrong you will need to adjust the tooth #1 offset (or whatever it's called in VEMS) until it's correct.

It's pissing in the wind to use a non-fixed timing value, as minor engine fluctuations will cause the commanded spark timing to change per the map.

Referring back to post 378, why would you not assume your issues relate to the fried motherboard? I didn't follow the back story, but you clearly had a serious overload on the ECU, enough to fry parts and melt solder. There's just no telling what else got damaged. You could have cold solder joints all over just from the heat they experienced; semiconductor parts could be fried or nearly so; hidden traced on a multilayer board could be burned out inside; etc. That the car occasionally runs great, and other times has readings that are all our of whack, is also consistent with these kinds of faults -- much like when a Motronic DME has bad solder joints. No guarantees obviously, and maybe I missed what you did to confirm the ECU is good, but if it were me I'd want to test that board on another car, or test another board on your car, before spending money on all those other parts...

https://cimg6.ibsrv.net/gimg/rennlis...6cc96eed6d.jpg
What else melted? Did it cool poorly and create bad joints?
https://cimg6.ibsrv.net/gimg/rennlis...7045b90e20.jpg
Very hard to know if you patched all bad traces and what parts may not have survived... Some boards have multiple trace layers you wouldn't see to fix...

OK, thanks will try this and report back.

I had friend the ignition driver transistors due to changing some settings in vems, which I wasn't suppose to do. Replaced them and had same issues. In my recent post and testing I had a local guy run my ecu in his 944 who also has vems, with his config and it ran perfectly. He restarted it multiple times and not a single hick up according to him.

I wish I could go back to dme / Klr easily but I've swapped out tps and permanently soldered in IAT sensor plug where maf used to be.

Was advised by Peep to add more time on ignition dwell. Gonna take small steps with this as it could fry my coil or board again.

Thanks
Keep them ideas coming guys!

Checked timing at idle, vems is set at 14, use timing gun and set at 14, I verify in timing window. Checked all timing marks also, cam mark, flywheel mark and balance shaft mark all show its correctly set at TDC. Checked knock sensor and found it showing 4.9M ohm at Klr pins, went and found a spare older model knock sensor, checked resistance and its 300k ohms within spec. Swap it in and noticed connector is frayed, chopped it and spliced in new connector. Fired her up and still same issues ( could have wires reversed but from what I read it doesn't matter ) so now I replaced all sensors in vehicle, associated harness/plugs. Entire fuel system ( though I'm still suspicious of aeromotive fpr as few are having issues with this brand ) as a last resort I also have a 36-1 crank trigger and bracket I got off ebay. But I'm not sure if it'll help much since I don't have trigger errors. Will keep for another project. I also gave Peeps advise a try and add or subtract coil dwell times from 1.5ms up to 4.5ms, not change. My coil is the only thing different from everyone as it's based off Vitesse Racing Wasted Spark kit. But it still doesn't explain why it would rub perfectly after a few restarts. Idk, maybe I can buy the coil supplied by vems and see if that changes anything.

What I know from the few of us having issues is that we all have a modified or aftermarket alternator installed. We also have hi torque starter. But there are those with these mods that don't have our issue. The only possible thing I can think of next is my IceShark cable kit, which includes battery, alternator and starter cables. My original cables were damaged so I will either need to get a new set or borrow from someone for testing. It's either that or something different in my engine harness thats causing the issue. My vems ecu works fine in another car.

​​​​

Does your Ford throttle body have the hole drilled through the throttle plate? Ford added the hole to help idle problems

Yes, its bigger than the hole found on the stock 951 TB.

installed a new fuel damper and new saddle adapter for 3 bar bosch fpr. No help but I hated that I kept loosing pressure when I shut off my car, apparently all Aeromotive FPRs will do that. I also was told CEP fuel damper leaked so I replaced it too. Still no change, but holds fuel pressure for a few hours like it should. Finally got around to install GSF wheel and trigger setup. I asked some on initial hardware setup. So I have motor at TDC and installed missing tooth after trigger sensor. Missing tooth is exactly 5 teeth from sensor, 5th one being on the sensor. Installation was very easy, it took longer to cut the belt covers than it did installing wheel and bracket. For a temp. solution for wiring, I got an injector plug and chopped up old speed sensor to plug into stock harness. Now I just need help in configuration in VEMS. I noticed theres a default config for 36-1 wheel in VEMS, but not sure the other settings is compatible with out motor. Any help would be greatly appreciated. Side note, I also ordered a new coil exactly like the one offered on VEMS website, maybe the coil I had for Vitesse isnt firing enough, causing lean condition? idk, it was $40, and I had a Ebay gift card for Christmas, lol. Here are some pics of install.


https://cimg6.ibsrv.net/gimg/rennlis...0437f55ad5.jpg
https://cimg9.ibsrv.net/gimg/rennlis...0a71bc20db.jpg
https://cimg6.ibsrv.net/gimg/rennlis...a5ad6791eb.jpg
https://cimg0.ibsrv.net/gimg/rennlis...1d605d4cc6.jpg

Tried default settings for 36-1 wheel and didnt work. Changed TDC after trigger to 50, since its 10 degrees a tooth and its 5 teeth from TDC. Could not get motor to start. Anyone here using a 36-1 trigger wheel? Need to verify pins on ford sensor is wired correctly ( 2 pin sensor to 3 pin harness, dme then vems )
https://cimg0.ibsrv.net/gimg/rennlis...a3645d1007.png

Looking at the values you've entered, you are saying that as you spin the engine, first the gap passes under the sensor, then after 4 teeth your 'first trigger tooth' passes under the sensor, then after 5 more teeth (50 degrees) the engine is at TDC? If this is not happening when you spin the engine, then your settings are wrong.

You can find the approximate TDC after trigger angle by counting the teeth as you have done, then when you get the engine running, use a timing light to fine tune the angle until the timing light matches the timing that you have set in vems ( use the ignition lock tool).

I found it helps to use the visual trigger settings menu.

https://cimg3.ibsrv.net/gimg/rennlis...a88ec01041.png

First Trigger Tooth should be 1,since tooth after gap is 0. Then 3 teeth after FTT would make it 30 degrees. Using ignition lock, do I set to zero?

So you're saying that TDC occurs on the 5th tooth after the gap? If so, then setting FTT to 1, and setting TDC after trigger to 30 degrees puts it at the right spot, but it is not within the recommended range of angles. TDC after trigger should be between 60 degrees and 127 degrees. Can you clock your trigger wheel to make a larger angle between the tooth gap and TDC?

Don't lock ignition at 0, it will need some ignition advance to idle properly. Idle is usually around 15 degrees advance. Keep in mind that if you have a non programmable timing light, like I did, you will need to put a mark that is so many degrees before the TDC mark.
I checked my timing by locking the ignition to 12 degrees, and then because I have a 60-2 flywheel, I simply put a mark on the flywheel that was 2 teeth before the tdc mark on my flywheel. When that mark lined up with the gun I knew the trigger settings were perfect.
You can also use the cam pulley and shine the light through the inspection window. Keep in mind the cam pulley turns at half the speed of the crank, so your timing mark has to be half the amount of degrees of the ignition lock before the tdc mark.

I can clock the gap farther. New gap will be 14 teeth, teeth after gap is 0, FTT is 1, so now it will be 120 degrees. Ok, will lock ignition at 15.

Ok yep, that should work if you're getting TDC on the 14th tooth after the gap. If you find that when you are fine tuning the TDC after trigger angle with the timing light, and the angle gets too close to the max value of 127 degrees, you can just move the FTT to 2 or 3 and reduce the angle by 10 or 20 degrees respectively.

Just tried my settings with 30 degrees after tdc, and 0 FTT and was able to dial it in to 36 degrees. No trigger errors. Ran and caught the motor on start way faster than on stock speed and reference sensors. I just had it rigged up temporarily with wires unshielded. But the issue is still there BUT it's not as bad now. It used to stall and sputter like crazy, now I can still Rev it past 3k good, just off idle stumble and a bit of tps lag. I'm betting all this issues might be ignition related, since trying to get my timing down, the motor sounded and ran similar with the timing off on the stock sensor setup. So I had my new bosch coil pack come in, only to find they sent the wrong item. Instead of a coil pack for four cylinders, they sent 1 coil of plug coil, 😂. On to the joys of waiting for a package to arrive in Hawaii. Gonna also try and set the missing tooth farther to get the 120 degrees setting to see if it changed anything. But this is great progress after spending a ton of time and money trying to solve this. Much Appreciated 🤙🏽
​​​

Guys, sorry for the noob question but I've searched and haven't found any information - what size tubing/hose fits onto the MAP sensor on the VEMS.

Photo below - I have the threaded `compression' fitting out of the shot.
https://cimg7.ibsrv.net/gimg/rennlis...dcd593c562.jpg

The OE boost line doesn't fit, and silicone tubing I have on hand in the garage isn't working.

Ultimately - I would love to know what kind of fitting this is so I search for the right part.

Thanks!