I recently installed a Motorsports x-pipe on my 90GT, along with the corresponding high-flow cats and a wideband O2 sensor & gauge.

This morning I met up with Dr. Rob (from whom I bought the car originally, and who was on the east coast because our turkeys are better) and we did some Sharktuning on my car. I put the PEMs in it on Wednesday but they had stock maps already loaded.

Over the course of three ~10min runs we collected data about fuel map changes based on the WBO2 data and the standard Air:Fuel targets offered within Sharkplotter.

In the higher-load end of the maps the net additional fuel was over 30% in some cells, and was fairly consistently over 20%. We were only driving on local back roads so no really high-RPM/high-speed loads, but plenty of enthusiastic 3rd gear usage and a bit of 4th.

Anyway after all is said and done, and recognizing there is more to be done on highway trips or on a dyno, the car is quite remarkably different-feeling. It pulls harder and feels smoother (objective, I know). AFR is getting into high 12s now, whereas before it was quite lean -- up into the 15s at times.

Rob also managed to improve my car's high idle somewhat. Remaining symptoms point to the throttle body itself, either the seals on the throttle shaft or the butterfly's at-rest position, but it's still a hell of a lot easier to drive when it's idling just above 900rpm vs 1200pm+.

And he made my clutch feel better too. So all in all, a grand morning, and I really appreciate him taking the time during a family visit!

So thanks to Rob and a big thumbs-up to everybody involved with Sharktuner and its software packages, that's a slick setup particularly given the rather, um, historically interesting design of the Bosch ECUs.

 

Car runs great in this freezing air. How did I live in this Arctic cold?

It was nice to see the old GT this morning, and particularly gratifying to do some baseline sharktuning after the installation of the Motorsports X. Initial AFR's definitely indicate that there's room for significant tuning after adding the X. Certainly more fine tuning to do but got it roughly dialed in quite nicely.

The GT tormented me for years with its intractable high idle. We spent some time this morning playing with the manual ISV setting in the ST2 software, by dropping it manually from 50% downwards. Taking it from 50 to 40 dropped it from 1220 to ~950 rpm, but it would drop no further, and paradoxically increased to 1000 rpm as the ISV was closed down to 30-32%. (Much less than that and the car should stall.... ).

We tried an idle adaptation with the new ST2 software, which improved the idle quality but would still not drop below ~900 rpm. (With the A/C engaged we can get it to the magic ~775 rpm, but today was not a day to run the A/C......) So the ISV responds to manual control, but the engine is getting enough bypass air from somewhere that it can't be stalled by essentially closing down the ISV.

So either there's something wrong with this (sub-10K mile) ISV, or there's an air leak I didn't address (at the throttle shaft seals, or the butterfly, or the throttle stop screw is out of adjustment and it doesn't fully close. One interesting observation- At idle, on the Fuel system monitor indicates the the throttle position to be at IDLE, while on the ignition system monitor it registers CRUISE. Then after a bunch of spirited driving, both read IDLE at IDLE. No real difference in coolant temps before or after, so I don't think it's temp related. Weird. Car indicates 90 mpg during coast-down, so the idle switch is activating.....

The other quirk is that the car never activates WOT, despite the fact that I carefully indexed the TPS on the throttle shaft and validated its operation back in 2007:

https://rennlist.com/forums/928-foru...questions.html

I have seen plenty of TPS's filled with oil, which presumably gets there by being pushed into the MAF elbow, then tracks along the throttle shaft, past the worn shaft seals, and into the switch. Despite having installed a new TPS less than 10K miles ago, I wonder whether the loss of the WOT signal might be due to faulty throttle shaft seals that have crapped up the TPS?


Just for fun, here's a 90GT, 117K miles, dodgy idle, with the 'baseline' stock 90GT LH map (at lower left), and a newly added X--pipe with cats. 93 octane, about 45oF outside. Datalog is from the first run, about 12 minutes of running mostly mid-throttle 3rd gear pulls from 3000-5500 rpm, then getting stuck behind Prii, etc...:



Sharkplotted the results, updated the maps, and went and collected another 10 minutes of data (Run #2). AFR's are a lot better. LH map after run #1 at lower left




One more easy 'cool-off' run #3, superimposed on run #2, re-sharkplotted the LH map.




This map is by no means perfect but it illustrates the amount of fueling that can be added into the midrange of the stock map after adding an X-pipe. We didn't touch the ignition timing, probably more gains to be had there (Though for some reason the knock retards we did see were primarily from cylinder #7, at medium load in the torque max around 3900 rpm. Not sure why that would be, maybe some carbon in that cylinder, dunno)

So, X-pipe owners, this is what you're missing by not sharktuning after your X-install! This was only ~1 hour of work, there's obviously a lot more refining to do and more WOT logging to do. But I'll let Cardinal risk his license to do that part.

 

This, I think... The one thing I bet you didn't touch was the throttle-stop screw. It is factory-set to allow a small calibrated air flow, which adds to the ISV flow and the brake-booster venturi to make up the idle air.

So if your average PO is trying to fix a low-idle condition, and spots a screw adjustment for the throttle stop, what's he going to do? That fixes the problem, whatever it was, and the next poor guy drives himself crazy....

The fix is to adjust the throttle stop until the car idles nicely with ISV% reading around 50-55% on Sharktuner. (You can get to it-- just-- with only removing the upper/lower air box. It requires a mini borescope camera, a 1/4" drive u-joint socket, and a skinny screwdriver). Or maybe there is a feeler-gauge spec somewhere, but I haven't found it.

It can't, it is the same switch wired to two different ECU's in parallel. They can't read different, they just can't!

It's got to be a dead WOT switch contact, or wiring. If you tuned with the WOT fuel map zero'ed then it won't matter. Did you also disable the EZK WOT map?

Cool plots, thanks for posting those. Things are lean-ish in the mid-range for stock maps, looks like more so with the x-pipe. The O2-sensor will fix that to a point, until things go open-loop as RPM and load increases.

There is certainly more timing to be had mid-range, with good effect. (Butt-dyno numbers are 40% mid-range torque, 50% acoustic, and 10% seat-foam compression).
Does a compression test show anything different for #7?

Thanks for posting the results, that is really great!

 

Compression results during Doc Brown's PPI earlier this year showed the following:

1 - 175
2 - 162
3 - 172
4 - 175
5 - 180
6 - 176
7 - 180
8 - 182

I can definitely confirm that the LH and EZK screens within the software showed different states for the TPS position concurrently. LH showed CRUISE while EZK showed IDLE. This was with the engine off. Later, with the engine running, they both correctly showed IDLE.

I suspect a TB-ectomy is in my future. I don't have a borescope so Jim if you do and find yourself near DC any time soon, great quantities of your preferred libation will be on offer if you want to help troubleshoot this further.

 

I have a USB endoscope/boroscope that I can mail to you, but honestly it'll cost you less to buy one than the roundtrip shipping to/from me. GB says its picture quality is better than his Snap-On BK5500.

http://www.ebay.com/itm/2M-10mm-USB-...item27daa9004b

Not sure if it'll work on a Mac product, maybe there are others on Ebay that will.

Didn't zero out the WOT map, but I sent Cardinal home with my ST2 on the off chance he'd need it (first hit off the pipe is free!!!) so he can do that.

 

I think it is important to note that the x-pipe did not cause a lean issue (as you indicated it was in the 15's prior to STing.) The fact of the matter is that the stock fuel maps are a tad bit conservative in pretty much all respects. This si why the Autothority chipos were so popular back in the day.

Without boost or being a stroker, high 12's for AFR at WOT is probably a bit rich. You may find that you will get slightly higher HP at low to mid 13's - but figuring this out on the dyno is the safest and best place to experiment.

The x-pipe does help the car breath and make a little more power - for sure.

As for the idle switch, if you have idle contact through the LH but not in the EZK, then there's something wrong. There is only one wire that comes in throught he harness for the idle contact switch. Like Jim says, it can't be active on one and not the other - UNLESS there is an issue with the wiring.

I had great luck with some of the USB endoscopes off EBay. They are small, and will fit through the spark plug hole. It's amazing what you can see in the bores with those things.

 

Andrew, it was your post a while back about the USB endoscope that saved me from buying a used Snap-On, when I showed it to Greg, he was like, 'Crap, that image quality is better than mine.'

True, I have no idea what the AFR's were with the stock exhaust, or whether the X-pipe would have changed the AFR's vs. the stock exhaust. That would be a good experiment for someone to do. It just seemed that AFR's of 15.5-16 under high load in 3rd gear using the stock maps with the X would be way leaner than the factory would have preferred, and I know the MAF calibration is spot on, plugs are new, timing is Spot-on 0|0, etc. etc.

I agree that the dyno is probably the safest (and legal-est) place to do WOT tuning, but when I was WOT tuning the stroker on the street, I got a fuel map that was pretty much spot on Greg's target AFR map everywhere. Then when we took it and did a couple of dyno pulls, the AFR's were choking rich (~9.8-10's) in the mid-range and borderline lean at the top end. Perhaps tuning on a load-dyno would be a better replicate of how the thing is going to behave on the street.


I know that it _shouldn't_ happen, but Cardinal and I both saw the mismatch in the idle position indicator. Clearly something amiss, somewhere.

 

Yeah I'll pick up one of those borescopes. I had no idea they'd be so inexpensive. If I could find one with a self-contained display head that wasn't silly money that'd be even better. I wonder if I can feed it into a tablet rather than having to haul a laptop out?

Aero drag is a significant effect in load tuning which a simple inertial dyno obviously can't replicate. I don't know if that's what you observed Rob in your street-vs-dyno experiences but this is a big heavy car and accelerating its mass will definitely take some energy, so I'm not at all surprised you had more fuel than you needed for static conditions.

I don't have any data from the WBO2 prior to the x-pipe install. It just seems unlikely the factory map targeted 15s. We tuned using the default targets built into the software, so while rich, they are quite safe until more rigorous optimisation can take place.

 

In almost every car I've seen with no WOT activation I found too much slack in the throttle cable. I take up slack until the cable is snug but the idle switch still clicks reliably. I assume you guys checked this but maybe a few others haven't.

 

It's maybe a dumb Q, but was the WBO2 calibrated properly? And when you did the baseline, were you running open or closed loop? Are you running a NBO2 in parallel or pulling from the WB?

I just saw that the idle switch discrepancy was only with the engine off. If so, who cares.

 

I am surprised you add to add so much fuel at WOT. Porsche set the LH up to somewhat overfuel at WOT (as a safety margin for aged MAFs ?) and then rather lean in the mid rpm/load range when not closed loop (economy).

I don't think the X pipe would influence the mixture at WOT, to make it lean ?

 

My guess is a lot of the added fuel was due to the absence of the built-in WOT enrichment since we weren't tripping the contact. So clearly at this point my incentive is *not* to fix it, or it'll get far too rich...

Masking hardware problems with software, FTW!

The car has two O2 sensors wired independently, so the NBO2 is still feeding into the LH. The WBO2 is an AEM unit which doesn't require separate free-air calibration. We did confirm the voltages and AFR limits on its output signal were configured correctly and saw good agreement between the display on the laptop vs. the display on the gauge.

If anything I actually think the throttle cable is a hair on the taut side. The other interesting thing somebody noticed at Frenzy was that there is no distinctive metallic 'clink' when the throttle reaches the end of its travel, it just stops like it's suddenly crashed into something very dense but ultimately not rock hard. The effective spring rate suddenly increases very sharply for the last couple mm of travel.

 

Ah yes, I'd overlooked the non-op WOT switch. As Jim says, that doesn't matter as Rob will have zeroed out the WOT cells.

 

[QUOTE=CardinalCar;10941772]Yeah I'll pick up one of those borescopes. I had no idea they'd be so inexpensive. If I could find one with a self-contained display head that wasn't silly money that'd be even better. I wonder if I can feed it into a tablet rather than having to haul a laptop out?

I believe Costco has one from Whistler with a self-contained display for $150. Plus, almost lifetime money back guarantee if it ever breaks.

 

The x-pipe shouldn't have much impact on the needed fuel map calibration. The car has a maf sensor after all. If the x-pipe makes more power, you should be getting a higher maf signal. The computer would then add fuel based on the stock maps and it should all be close enough for government work.

The car running lean is much more likely due to the throttle position switch not working, than the x-pipe anything. I wouldn't be shocked if the base map calibrates to about 14.7 afr and then the WOT map adds a couple of points with the pedal to the medal. If the switch doesn't signal WOT because it's broken, tha car appears lean. The maps are not the true reason however, it's the faulty switch.

If one wants to run the car with broken a tp switch, then zeroing out the WOT fuel map and tuning the base map richer seems like an acceptable temporary solution.