gasmanstl

At wits end with a problem RE rich running and failure to go into 'closed loop' mode. Have eliminated about everything else. Not temp II, O2 sensor. MAF tests OK and has been refurbed by John Speake recently. I can't get car to run stoichiometrically even with fuel pressure turned WAAAY down. TPS is not good, but if anything, it should not be prompting brain to enter idle map (closed throttle should be 0 ohms, and tested anywhere between 34 and 55 ohms, but is infinite when throttle open at all, as should be). (Just never entering the idle map couldn't cause this problem, right?) WOT is OK I think - with throttle closed, the circuit isn't completely open per se, but R=14.5 mil ohms - pretty close - and with throttle firewalled, WOT = 0.37 ohms.

I personally believe that the TPS having 30 - 50 ohms instead of 0 ohms in the 'throttle closed" position - which, admittedly, will probably keep the computer from ever entering the idle map - should not be sufficient to keep the computer from working in normal/closed-loop mode in "putting around town" throttle settings...UNLESS the fact that the WOT switch is only 14.5 mil ohms when "open" - instead of infinite ohms, will lead the computer to always be in WOT map. However, I have the disadvantage of not knowing much about it...

Ahhh, scheist. The TPS just flat needs replaced. But I really don't wanna do the intake now...I'd rather swap in a computer to rule that out. Whaddyou think.

Should I be looking for a computer to swap as the next diagnostic, or just bite the bullet and replace the TPS? Any other ideas?

Thanks in advance...

Clay

FlyingDog

Unplug the TPS then jumper the connectors while running to observe changes in behavior. Also check for continuity between the TPS connector pins.

worf928

+1 what Matt said.

The Idle/WOT switch isn't about ohms. It's about continuity. If you are getting non-infinite with the WOT side when not near WOT then the computer is indeed thinking the WOT switch is WOTing all the time. I'm not 100% certain, but, if I was a betting man I'd say that off-idle you'd always be rich because the computer's offsetting the base map for WOT. What it does at idle is (with out disassembled LH micro-code) anyone's guess. But, again, if I was betting I put my money on the micro-code being dumb and trying to trim the 02-loop to stoich while at the same time offsetting the base for WOT.

John Speake

Hi Dave,
A shorting WOT switch will also enrichen the mixture at idle.......

gasmanstl

"""Unplug the TPS then jumper the connectors while running to observe changes in behavior. Also check for continuity between the TPS connector pins"""

Now there's an idea...why didn't I think of that?

Done. With the plug disconnected (no chance of WOT being detected as continuous, even at a resistance of 14 mil ohms), and with the other 2 pins jumped, simulating a closed "idle position" switch, the symptom is the same. I suppose this means I need to find a computer. Am I missing something else?

John Speake

From a couple of mail PMs we have exchanged off list I understand that you have 30# injectors fitted. This will mean you will have to adjust the MAF to get the correct idle mixture.

You also have a narrow band O2 meter wired onto the O2 sensor, plus a voltmeter on the O2 sensor for testing.

Make sure the idle throttle switch is giving zero ohms at the LH ECU when the throttle is closed.

Turn the pot on the MAF anticlockwise and see if you can get the volts on your meter cycling around
0.5v.

If you can get a CO meter on there, before the cats, disconnect the O2 sensor and set the idle CO to 1% (airpump disonnected and pipe plugged)

You may still have mixture control problems off idle with those bigger injectors....

gasmanstl

For completeness' sake, I am posting the contents of my pm's with John, so that you all may see this and reference in the future, and so that those following this thread will be less clueless...so this post should really be before the last one by JDS. thx. Clay

----- Original Message -----
From: Gasmanstl@cs.com
To: info@jdsporsche.com
Sent: Monday, October 29, 2007 5:57 PM
Subject: Hello John &staff:

(John Speake says) """"Hello Clay,
Yes, I have been meaning to jump into your Rennlist thread.....

Have you tried unplugging the O2 sensor to see if it idles OK ? Also check mixture at idle under those conditions and adjust MAF as necessary for 1% before the cats.""""

(gasmanstl says) Well, no...what I did do, though, was to access the wire to the O2 sensor via an access point to the AEM wideband O2 meter thingie, which Mark Robinson had installed before I got the car...and to clip a meter lead to it and to ground. (The AEM is basically the only way I know it's running rich and not going into closed loop, besides the unburned-gas smell of the exhaust. Otherwise, it actually runs OK) The sensor is clearly working, I think, the voltage changes as the thing heats up and when warm, it usually doesn't read much below about 0.8v this way - though I can make it fluctuate by producing vacuum leaks. Incidentally, the car idles with about 17-18 inches of vacuum, so I don't think this is a problem. I should also mention that Mark has installed 30 lb injectors, by his own history. (He was working with me on this via e-mail, but he's been in Italy for awhile now...and I'm slogging through).

About the MAF adjustment...I have not attempted to adjust that mixture screw with the plug in it? I presume that's what you mean. I am not equipped to measure CO before the cats, anyway. You have actually just serviced this MAF, as Mark thought that was the problem initially. It didn't occur to me that it would be very far off, since you had just serviced it. But...the injectors?? Does that change the equation?? (By Mark's history, the car ran fine for 7000 miles with the turbo, and the injectors, and the computer working fine).

If I don't have a CO meter, should I still try to do something with this screw, based on the AEM reading?

Cheers

Clay / gasmanstl on Rennlist.

gasmanstl

Hokay... :
Swapped out the LH brain for my working 85 16v S euroshark: this brain actually runs a little richer, according to AEM. And this running 10 PSI of fuel pressure! (I have also changed the gauge to ensure the number is correct).

So, then, I jacked up the R rear wheel of shark (since the MAF lives under the car,) having swapped the old LH brain back in, to adjust the trim on the trimpot that John suggested on the MAF. Before moving the pot, I adjusted the fuel pressure back to the low, but reasonable range, about 24 psi - (stopped there because it started surging). Since the AEM is obviously a glorified voltmeter (and since I re-soldered and shrinkwrapped the whole thing so nicely just a couple of days ago, nixing access to the wire with a meter) - I simply attempted to trim the pretty display lights into the "green" range. No go.

I have turned the trimpot three turns now (counter-clockwise, right, John?) - and I'm scared to turn it any more, lest I be shipping the MAF back to UK. One thing that really strikes me is the lack of resistance to turning this screw (for clarity's sake, we're talking about the recessed 3mm screw with the much-larger-looking head, right)? You can practically breathe on it to turn it. Is this normal? (There was no plug, btw...when I am finished with this, I plan on fashioning a watertight plug for this item.)

If it were a stuck injector, wouldn't the gauge pressure on the fuel rails leak down relatively quickly after shut-off? It does not.

Thanks again in advance...Clay.

John Speake

Hmm... you didn't say at the start of this thread that your 928 had 30# injectors AND a turbo !!

You should be able to turn the MAF idle pot (3mm adjuster) at least another 6 turns or so anticlockwise.

If uncertain if you have reached the end of it's adjustment, then measure from pin 6 to the body of the MAF with a ohm meter. When you get to zero ohms don't turn anymore.

You are OK, the adjuster moves very easilly.

Quite honestly, with 30# injectors and a turbo with all its plumbing, you really need to SharkTune it.

gasmanstl

[QUOTE][Quite honestly, with 30# injectors and a turbo with all its plumbing, you really need to SharkTune it.
/QUOTE]

Yes...I've heard of this Sharktuning thing...never seen it done firsthand, nor even read much about it...but at any rate, from your website, it looks like it only works for LH 2.3, which I don't have...

Thx for comeback...will go out now and crank the screw a little more...

FlyingDog

You're putting out 25% too much fuel putting your air/fuel in the 11s by default. IIRC, the lambda loop (O2 sensor) can only correct around 5%. Using the Euro LH, the lambda loop doesn't exist.

John, What is considered an acceptable range for a MAF? I remember you posting that 300ohms is where they should start life.

gasmanstl

Back from the garage, smelling totally like exhaust puke:

1). My TPS is totally f'ed. I seem to get different numbers each time I measure them.
2). I did this trial with TPS connector D/C'ed, thereby simulating "idle switch open, foot somewhat on throttle," and "no way is it WOT, even if the switch is shorted." Therefore, definitely in map 2, right?
3). Measured pin 6 on MAF (from disco'ed LH plug) to MAF housing, starting from where I was earlier today, 3 turns counterclockwise from the JDS factory refurb setting. 168 ohms...
4). Slowly turned adjustment screw out until 0 ohms, to count how many turns it took (can't keep meter hooked up in realtime while I'm trying to adjust mixture). It took three more turns, for a total of 6.
5). Rehooked LH brain and started motor, back in the "three-turns-out" setting. Using the AEM meter, I turned the screw out slowly, 1/2 turn at a time, doing throttle blips in between each setting to assure steady-state conditions had been achieved....

...and the meter never moved. Still "two-bars-rich," (and exhaust smelling like gas-o-slime). All the way out, 6 turns. I knew the R would be zero. Just for fun, I jumped the idle-switch receptacles on the plug, and the motor settled into a nice idle - but again, the meter never moved. And no closed-loop behavior - (by this, I mean the meter goes to zero and initiates another sample every second or so) - was ever demonstrated.

6). I stopped the motor and disco'ed the LH brain again, and verified that R btwn pin 6 of MAF and its housing was zero. It was.

Gentlemen, I don't know what to think. Let's just hypothesize that the injectors are so big, that the computer will never correct for it. Would running rich like that be enough to keep the brain from entering closed-loop mode...or is something else wrong? If the mixture is good, and it never goes into closed-loop...do I really care? Should I turn the fuel pressure back down and see if I can trim it into running stoichiometrically?

Your thoughts and input are, as always, appreciated.

Lizard928

Johns sharktuner will work with the LH 2.2, though you have to purchase the ST and upgrades/adapters to adapt to the 2.2 version.

As to your issue, if it were me, I would be tearing the factory EFI out and going with standalone.
It is cheaper than getting the sharktuner and you then loose the MAF. You run instead with a MAP sensor, so should you all of a sudden get a vacuum leak, it will automatically compensate for it, instead of all of a sudden getting extra air causing you to go WAY too lean.

FlyingDog

Running that rich can crap up the O2 sensor to the point where it doesn't work.

Low fuel pressure is bad. Bosch injectors are designed to run at 43.5psi.

John Speake

I would have been good at the beginning of this thread to have had a proper summary of the build of the car, instead of basic facts emerging gradually.

I had the impression that this was a stock car which had develpoped a problem.

Now we are told it is a rear mounted turbo car. Has the car ever run properly, or are we redesigning it ?