damonoresky

Hi All,

I have a 1991 S4 that is failing the emissions tests after a tune up. The car is in California, Sacramento area and has about 55K miles on it. I literally drove the car from the work shop after the tuneup was done to the emissions test facility and it failed. The car was warm as it is about a 30 minute drive. In the past 4 years (2x testing) the car passed with flying colors.

The NO (PPM) is measuring 1167 when the max allowed is 706. This is at 25mph. The car passed at 15mph with a reading of 134 where the max is 767.

And even though the tune up was done (plugs, wires, cap) the car still runs a bit rough at idle even when warm. It will idle fine, "chug" a bit after a few minutes, the battery gauge dips, and you need to rev it up a bit or go into neutral or it will stall. This randomly happens at red lights or even when just backing out of a parking space after driving.

I don't fault the shop as the car does run better than before and the owner admitted he is not a 928 "expert." He does work on a lot of new Porsche though.

Any tips or ideas where to start would be appreciated. Oh, lastly, a few years ago the car did register a bad Hall sensor but it passed emissions 2x with that.

Thanks in advance!
Damon

jcorenman

High NO generally means a lean mixture. Interesting that it is OK (by a wide margin) at 15mph and then fail (also by a wide margin) at 25.

Can you post the complete results, as well as the previous test? There are likely some clues there.

And the rough idle, stumbling and stalling is another good clue, which also points to a lean mixture. These engines idle great when a bit rich, but fall on their face when lean.

It could be an air leak, a split hose under the intake or similar that would allow unmetered air to bypass the MAF. Has the intake been up to check those hoses?

It could also be an aging MAF, that would tend to cause a lean mixture. Aging is both a function of miles and stop/start cycles, it is actually the burn-off cycle when the engine is shut down that causes the most aging. But generally they are good for 60-100K before falling far enough out of spec to fail testing.

What about the O2 sensor, has that ever been swapped? That will also cause a lean mixture and is simpler/cheaper to swap. Typically those are good for about 60K miles.

A bad hall sensor would retard the timing, which would actually reduce NO a little.

damonoresky

Thanks for the information, this is a good start. The shop did mention that they "cleaned" the MAF (is this even possible?) and maybe that messed it up? Can these be replaced? I believe he said they are not available anywhere to replace.

I can dig up the old emissions results but I am pretty certain the numbers were fairly identical for both speed tests in previous times.

OTR18WHEELER

I'm no emissions expert, but I believe the grade/octane rating of the fuel, and the type of spark plugs can affect the lean condition.
you should have standard copper plugs, and avoid the lower octane gas.
Lower octane ignites quicker, hence leans the mixture.

jcorenman

Yes, and yes. Your friendly local auto-parts store sells spray-cans of "Mass Air Flow sensor cleaner" (CRC and other brands), but it doesn't do much for these types of "hot wire" MAFs. There is also some risk of breaking the hot-wire element by being over-zealous. But in that case the engine would barely run (if at all). So that's not it.

New MAFs are indeed unavailable (and would be unaffordable otherwise), but rebuilt/recalibrated MAFs are available.
Louie Ott (louie928 here) is the agent for JDS Porsche in the UK, his website is here:
https://www.performance928.com/products.html
I am handling shipping for Louie for the next few weeks.
I can also test your MAF ff you want to send it here, see Louie's website for details and PM me for my shipping address.
Also I believe that Roger has MAFs in stock, not sure about testing.

I think the most likely issue is a split hose somewhere. Have a look around for any disconnected hoses, but most of those leaks hide under the intake.
The shop should have a smoke-tester for finding intake leaks, I would start by asking if they can have a look with that.

If that's OK then O2 sensor and MAF would be next on my list, but with only 55K miles I am not confident of that being the problem. Testing the MAF however would answer that question.

Yes, that's what would be expected. But it would also be useful to see the how the HC and CO numbers compare from before to now.

worf928

Mass-Air Sensor likely (very) needs to be rebuilt: JDS Porsche, Injection Labs.

Dying at warm idle - assuming single failure mode - in an automatic is usually the MAS.

An intake 'smoke test' will show false air.

O2 sensor may be dead or dying.

Need HC and CO results for better diagnosis.

Poor emissions on an S4+ is usually at least two of: bad MAS, O2, or false air.

jcorenman

Good thoughts. The anomaly here is the big difference between the 15mph and 25mph tests. I would think octane or spark-plug issues would be relatively constant. A split hose on the other hand would behave differently with different engine vacuum, which would be RPM-dependent.

But that's a good question: Damon, what kind of pugs are in there? Does it show on your receipt? What you want are Bosch W7DCR+ copper plugs, nothing fancy.

V2Rocket

Oxygen sensor dying will cause fluctuating warm idle.
As said previously lean mixture tends to lead to higher NOx and also affects idle smoothness.
Don't know the specific RPM on the 928S4 DME, but in earlier Motronic the O2 sensor has a "minimum RPM" to activate closed loop, with coolant in a particular temperature range. 1500 might be below the threshold but 2500rpm or so for the 25mph test is right in there.

Richard S

Hey Damon,
Hope everything else is going well, its been a long time since we've touched base.

If you need to smog it in a hurry, you can borrow my MAF. I had it rebuilt in December. Mine passed smog just last week.

BTW, EuroSunday is this weekend in Roseville if you can make it.

Rich

dr bob

I'd add the LH controller to the candidate list of causes, along with dirty fuel injectors. I can speak to low-miles seldom-driven cars as candidates for varnished injectors as cause for higher NOx readings.

You can gently pressure-test the intake system to find if there are air leaks. But leaks generally show up worse at idle, as manifold pressure is lower and an air leak will be a higher percentage of total intake charge.

Basic checks of fuel pressure would be a Good Idea before you get too far along.

Does the car have fresh fuel in it now? Old gas can be a problem.

There are some good in-the-car injector cleaning systems that have demonstrated value in helping this situation. It's hardly worth the cost of trying to assemble a DIY solution IMO, so ask your shop if they have a system they recommend. Someone here may jump in with a recommendation. The new plugs should readily show where there are lean cylinders after a bit of normal driving under load.

As others share, the Bosch plain copper plugs have a colder electrode than the premium platinum plugs pushed by many shops. The original fitment was the Bosch WR7DC, but that sells under a new number now. Your parts seller will know the four-digit number. I've had best luck getting them in sets from one of our regular 928 parts places, vs getting mixed batches from a few local parts places.

Get with Rich S for the easy things to test, like the MAF and the LH controller. Go from there.

damonoresky

Thanks all for the info and advice, this is truly one of the best forums around.

Rich, I may take you up on that offer and perhaps also see you at the Euro Sunday if my car can make it all the way there

damonoresky

Hi Jim,

The paperwork says that the plugs are WR7DC+. Other than that all they did was spray some cleaner the MAF sensor and said there were no visible line leaks.

Also there was a suggestion that there could be carbon build up on the valves and I could run something (C-foam? (sp?)) thru some vacuum lines into the intake that would clean them? Sounds a bit hokey to me...

Chalkboss

Show up on Sunday! It would be good to meet you, I'll be there. If Rich comes we can quickly swap the MAF out. Rich, I'll bring the HF screwdriver

John Speake

You can check for possible intake leaks by testing the vacuum at the front fuel damper. At warm idle you should get about 20"Hg. A much lower reading indicates an intake leak.

GregBBRD

High NOx are very common in the California test....becoming more and more prevalent as these cars get older.

It would be far better if we could see all the test results....there's a bunch of information in those other numbers.

Obviously, from your description of how the car performs, there is something very wrong that a very basic spark plug change didn't fix....and repairing that issue first may be the best way to proceed.

General information:

The O2 sensor (if working) in these cars tends to drive the fuel mixtures (CO) very low. And the information that the O2 sensor interprets and reacts to is an average of all the cylinders. If one cylinder is lean because of an intake leak or a poorly performing injector, the NOx will be driven up.

The very first thing we check when we get a car with high NOx (common) is the entire intake system for leaks. We use a smoke tester (best tool I've bought in years) to do this.