SQLGuy

Is Temp II the modular plug sensor on the water bridge, or the two blade connector one?

If Temp II, or its wiring, is bad, could this cause an S4 to maintain idle around 1K?

I've checked for vacuum leaks (not exhaustively, and not with pressuring the manifold), but I have visually verified vacuum lines and hoses, and have tried running an unlit propane torch all around areas where the manifold might suck air, and have not found anything yet. It really seems like the LH/ISV are actively maintaining 1K. Idle switch signal verified good at the LH.

Background: this started when I installed my Murf kit. Before installation I had a very consistent 680 RPM idle. Installation of the kit included new injectors and a decent amount of vacuum/EGR replumbing. Idle AFR averages close to Stoich, but bounces around between 13.8 and 15.4 or so. Idle at startup is about 1300, then drops to around 1000 within a few seconds. It's not perfectly steady, but ranges between about 920 and 1020.

Thanks for any further troubleshooting suggestions,
Paul

Lizard928

Looking at the front of the engine left side, same plug (but blue) as the injectors.

SQLGuy

The left side of the water bridge on my car has a yellow plug.

Lizard928

Then yellow, that is not factory.

Speedtoys

Being a standard injection style plug, could have just been replaced with a new plug/etc...but that is Temp2.

What is your sharktuner set to for idle?

SQLGuy

No Sharktuner, just the LH and EZK chips from Tim.

SQLGuy

Well, I measured Temp II and tried running with it disconnected. Cold it read 2.5K or so for each pin to ground. Ground to other ground was about .02 Ohms. Disconnecting it while running and warm made no obvious difference.

I also tried spraying starter fluid under the manifold and near the injectors. Didn't notice any obvious difference there either.

Measuring vacuum (at the front FPR), shows 11" at idle. I'm at 6500 feet elevation here, but that's still not enough vacuum.

Thing is, if the ISV thinks it's being told to regulate at 1K the vacuum doesn't really prove anything does it? The ISV itself would then be the vacuum leak. Is there an easy way to disable the ISV so I can see whether that makes a difference? If there's a real leak somewhere, idle should still stay high... but if the ISV is being told to open, I would expect the car to stall or at least idle really low with the ISV disabled.

Thanks,
Paul

David L. Lutz

After my SC install I had a very similar result. I just could not understand if that before the install my idle was around 650-700 and after right at 1000. Also to add to my confusion I had just replaced all items under the manifold with new.

After talking to Tim several times he kept pointing me in the direction of an air leak somewhere! Telling me with the kit/chips/etc. my idle should be normal.

Only after I pressurized the top end I found a couple of very small leaks. Once fixed made all the difference. Maybe for some reason with the SC the idle is more sensitive if you have an air leak.

Also check your blow off valve and make sure it holds vacuum.

jcorenman

Did you measure the Temp-II resistance at the sensor, or at the LH & EZK connectors? Measuring at the sensor terminals verifies that the sensor is good, measuring at the LH/EZK also verifies that the harness wiring is good. See WSM page D 24-14, shows how to check the Temp-II sensor from the LH connector.

But I don't think that is where your problem lies...

At 6500 feet, Everything is a vacuum But you are thinking like I am, that there is a leak somewhere.

If the LH is trying to regulate idle at 1K rpm then you are correct. IF...

Yes, it's called a Sharktuner.

There are two ways to attack a problem like this: The component approach, versus the systems approach. From your questions, I am guessing that you are a systems-oriented guy, who likes to collect data, analyze it and then declare what the problem is and how it fix it. And just often enough, you are met with resounding success and are encouraged to continue this path to success. I am the same way, my background is science and engineering and computers, which often respond well to that approach.

The other method is to carefully make sure that each component of the system is working correctly, and once that is done properly then the system will work the way that the designers intended. This is certainly the method embraced by Porsche when they wrote the WSM, and by most mechanics-- in fact by most folks, in any profession, who actually make stuff work. It's well-proven, I think, that engineers are the last folks you call when you need something to actually work.

So what we've got here is a system with a number of components, some of which are rather inaccessible (e.g. the ISV). And a system controller (the LH, in this case) that you have no idea of what it is doing or why.

In short, you are stuck.

So with a Sharktuner, the first thing I would do is check the Temp-II and other sensor readings, as seen by the LH, then have a look at the ISV control signal. For an S4, it usually hangs around 45-50%, up and down as needed to regulate idle speed. If it is low, or unresponsive, then there is likely a leak-- the engine doesn't need the ISV to maintain ia 1K idle speed.

The next thing I would do is disable the ISV-loop and control the valve manually... crank it up to 60% and the idle should increase, down to 40% and the engine had better stall. (Most ISV's are fully closed by 35-38%). If it doesn't stall, then there is a leak-- or a faulty ISV. By peering into the system controller you can diagnose the problem, or at least narrow it down. This is the analytical approach, and it works.

The "component approach" is more straightforward: pull the intake, inspect every hose and part and replace anything that is in any way suspect ("original" being a prime suspect), reassemble, do a pressure test, and hopefully the problem is fixed. If not then check the harness, replace the LH with a rebuild (a good idea anyway if it hasn't been done), and hopefully it is fixed then. This is the path that is outlined by the WSM, and it works.

I am sensing that you prefer the analytical, systems approach. But that only works if you have the right tools. For a circuit engineer, that's a scope. For a software guy, a computer with the right software tools. For a 928, you get the idea.

Are you going to be at SITM?

Cheers, Jim

steve_p

I'm probably way off the mark here, since I've only owned my 928 for a few weeks, but I had a cold idle of around 1200, coming down to 1000 when warm. To fix a stalling problem I replaced the CPS, and since then the idle has held steady at 600, hot or cold.

SQLGuy

Hi Dave,

I checked the blow off valve nearly first, since it was one of the new things added during the SC install. Since then I've also checked the FPR and the fuel dampers, and a few other things. One item that surprised me a bit is the hard black vacuum line that goes to the device attached near the coolant reservoir; it's actually pressurized when the car is running. I guess it taps off before the throttle plate.

Where did you find your leaks? I didn't have my manifold off, so I shouldn't have disturbed too much, but you never know. I did try disconnecting the S/C from the MAF and the idle is the same. I then tried pinching off the inlet, and the engine pulled a very hard vacuum and quickly died. If there's a leak, it's nowhere near enough air for the engine to idle off of.

Hi Jim,

I'd love a Sharktuner, but I am also pretty torn. For that price I could also remove both ECUs and swap to MS-III with full sequential injection and even COP support. I used Microsquirt to convert my 1982 Yamaha XJ-750 to fuel injection. Although a lot of that project was a pain, especially since I bought one of the first 10 Microsquirts, I like having full control of all the aspects of system. As it is now, that bike is running pretty nicely, with Microsquirt and MS2Extra running semi-sequential port injection, and dual coils for wasted spark.

Sorry, no, I won't be able to make it to SITM. One of these days maybe...

By the way, how messy/complicated is it to open the LH plug to allow access to the signals at the contacts?

Thanks,
Paul

jcorenman

Sure, that works. You'll need to redo the harness of course, and I think you need a different CPS arrangement-- if I recall correctly the MS isn't compatible with the factory CPS/timing-ring arrangement. And sequential injection is neat, with a separate WBO2 on each header-pipe you can tune the fuel for each cylinder individually. (Absent that, it seems like just one more variable to try to tune properly).

I don't understand the COP thing, though-- the factory twin-coil ignition is pretty nice, and also pretty robust- it is rare to hear about troubles of any sort. The thing that I really like about the stock EZK ignition is the knock detection & control. I don't know if the MS systems have that, I would hope so.

In the end, it is all about how you want to spend your time. If you like to build stuff, then you should build stuff. Cars are a lot like boats: You can build boats, or go sailing. It is very hard to do both. And I think it is the same with cars. To me, working on these cars is the means to an end, to be able to drive them.

It's not bad, a couple of screws to pull the back off. Getting the contacts out, if you need to, is a bit tougher-- there is a locking strip on each side, that's easy, but you need a special tool to release the contacts.

Cheers, Jim

SQLGuy

Thanks Jim,

I'll probably try pulling the LH cover off then to log some of the signals I'm interested in.

As for MS...

The stock harness would be fine. I'd just need an LH and an EZK connector to plug in the MS box.

I'm pretty sure MS will work with the stock CPS... at least MS2Extra would. It works with almost anything. However, a lot of people like to run Ford EDIS, which requires a 36-1 wheel. I wouldn't do that.

Sequential and injector trimming are related, but separate. Sequential injection, once you get the timing right, is supposed to give a smoother idle and better fuel efficiency in low RPM ranges. Injector trimming is where you'd probably need to open up those individual EGT ports an add a sniffer. However, the guy from Innovate says that he's been able to reliably trim individual cylinders by watching the output of a single LC-1 on a scope and looking at the raw pulse data for each cylinder's exhaust. Converting from batch fired to sequential, though, would require redoing the injector harness.

I agree that the stock 928 ignition system is pretty good. OTOH, Porsche has going to COP themselves in their newer cars. That would require a decent amount of additional harness work, though, and the Cayenne COP's probably aren't cheap. Main benefits would be no caps and rotors to worry about and no HV running across distributed lines (with corresponding EMI). You should also have some better dwell control... but, again, this is not something I've heard complaints about in the 928.

Yes, the newer versions of the MS firmware include knock control as well. Also one feature that's missing from LH 2.2 - flood clear.

Cheers,
Paul

Tampa 928s

The mega Squirt-3 is compatible with the 60 Tooth!

I would double check your Maf if there is a leak, that's where you will find it.

The piping tends to place a load on it as you tighten it; creating a small leak. It will not be completely flush.

What I did was to get the piping connected to the Maf hold the Maf in place then tighten until your snug then connect the piping at the turn.

You can remove your LH, take off the cover to the 35-Pin connector then test the pins, not that hard!

Mrmerlin

another area to look at is the MAF boot there are 2 plastic connectors that if they are old usually will crumble and they can leak from where they are pushed into the boot,
some others have figured out ways to remove these connections from the boot and replumb them

One thing to consider dont spray any flammable liquids or gasses onto a running engine you can find yourself in a small explosion,
as an errant spark plug wire will neatly ignite the liquid or gas,
have a charged water hose at hand