Questions about failed emissions test, high NOx and turbos

[richardodn]

Car is a 99.5 A4 Quattro Avant automatic with a stock AEB 1.8T motor.

I took this to have the emissions testing done and the car failed on high NOx. HC was well within the acceptable limits.
The numbers...

               LIMIT
HC GPM  0.5366 1.2000 PASS
CO GPM  6.4360 15.000 PASS
NOx GPM 2.7936 1.5000 FAIL

Looking back at the test history, the PO had a failure back in 9/2013 with HC 1.3471, CO 10.8389, NOX 3.5341. This one looks like a misfire condition. I know the PO replaced at least one coil as well as the ignitor and I'm guessing it was due to this. Additionally I was told the catalyst had been replaced. Assuming the PO didn't lie, it was likely done at this time too.

Typical reasons for high NOx with good HC are high combustion temps or failing catalyst. This is typically due to:

A. Bad EGR function - Am I correct that the AEB does not have an EGR system?
B. High coolant temps - OBD reported temps of 93-95C during the test. This seems fine for a 87C thermostat, but I may try an 82C if it fails again. I don't suspect any other cooling system issues.
C. Failing oxygen sensors - This is what I suspect the problem is. I recently had to clear a related sensor code. I've got 2 new Denso units on order.
D. Lean A/F mixture - Can this be read out of the measuring blocks on the AEB?
E. Excessive carbon deposits - Seafoam?
F. Failing catalytic converter - My other suspect. New Bosal unit on order.

Now here is where we get to the turbo questions. As the PO said the catalyst had previously been replaced and I suspect it was done about 3 years ago, I'm thinking that failed a bit prematurely. Of course it could just have been a crappy unit which wouldn't surprise me one bit. (Many shortcuts taken by the PO.) I plan to give the turbo an inspection while the cat is off. My suspicion is that oil from the turbo might be contaminating the cat. I don't see excessive blue smoke in the exhaust, but the turbo is likely original and now has about 190k miles on it. If I do decide to replace the turbo, here are my questions.

All units are new Borg Warner parts.

K03-005 - ($640) This is the factory unit for the non-drive by wire AEB.
K03-027 - ($640) This is the factory unit for the later drive by wire engines. Is there any benefit or detriment to using this on a stock tune? If I later get a tune for this, what's the upside?
K04-015 - ($830) This is the bolt in replacement for the K03. My understanding is that there are potential issues without supporting mods. Is this correct?

And finally to temper your responses, other than making it less of a dog, my goals for the car really don't amount to much more than a reliable beater.

Thanks!

[RENOxDECEPTION]

No EGR on any 1.8t.
AEB is narrowband so the o2 sensor only says if rich/lean/stoich not how far between, but I guess you could try.
Only use an OEM thermostat OEM temp.
What state do you live in? out of curiosity, guessing California.

[andyrew]

Low HC's indicate that the Cat has been replaced.

A. No EGR
B. Temps look normal.
C. Possible
D. Yes, if you have vag com or free vag com log block 31 for actual vs specified 02, and 15+16 for missfire's
E. Seafoam or water, but this will get in your cat.


If you want to pass, put 1 gal of Tutoline or Xylene in your tank to 5 gals of gas. It'll drop your Nox down to pretty much 0. Also speaking of gas, what gas are you using?

[richardodn]

No EGR on any 1.8t.
AEB is narrowband so the o2 sensor only says if rich/lean/stoich not how far between, but I guess you could try.
Only use an OEM thermostat OEM temp.
What state do you live in? out of curiosity, guessing California.

Wasn't really planning on changing thermostat. And after thinking it through a bit more, a lower one wouldn't change a thing.
Colorado > 5000 ft

Low HC's indicate that the Cat has been replaced.

A. No EGR
B. Temps look normal.
C. Possible
D. Yes, if you have vag com or free vag com log block 31 for actual vs specified 02, and 15+16 for missfire's
E. Seafoam or water, but this will get in your cat.


If you want to pass, put 1 gal of Tutoline or Xylene in your tank to 5 gals of gas. It'll drop your Nox down to pretty much 0. Also speaking of gas, what gas are you using?

D. I have this which is supposed to be able to read live data as well as much of what VAG-COM will do for these older cars. I have yet to try it for anything but a code reset. https://www.amazon.com/gp/product/B0...?ie=UTF8&psc=1
E. Sounds like a dose of seafoam before I change out the cat is in order.

As for gas, I run premium which is 91 octane at this altitude. Gas itself is either Conoco or from a grocery store chain with a new pump station.
Tutoline == Toluene? I may try this if it fails again.

Thanks!

[andyrew]

Possibly. Xylene is easier to get your hands on and I know it works.

Your reader probably doesnt do logging so you wont be able to read certain blocks like Vag-Com. You can get an ebay cable and the Ross-Free program to do this with. Not a bad $20 investment.

[Avant Nate]

I would also check for any vac leaks.

[richardodn]

Low HC's indicate that the Cat has been replaced.
D. Yes, if you have vag com or free vag com log block 31 for actual vs specified 02, and 15+16 for missfire's

It looks like this tool will read the measuring blocks although at this point I don't think I can actually log them. Reading 15 & 16 shows no misfires on any cylinder. While reading block 31, the tool presents it as pre and post values (option 1) and not actual and specified (option 2). That doesn't mean the tool is correct with its labels though. In either event, what should I be looking at? How do I interpret what I'm reading?

Option 1
31	Oxygen sensor voltage	1- bank systems	
	Bank 1, sensor 1	Bank 1, sensor 2		
	[V]			[V]		
Option 2
31	Linear oxygen sensors	1- bank systems	
	Lambda actual value	Lambda specified value		
	[]			[]

[richardodn]

Here's a followup to my test failure.

Initially planned on swapping just the cat and O2 sensors. I had also done a Seafoam application and had a bunch of smoke come up from the exhaust side so I figured some gasket replacements were in order. After disassembly I had a chance to inspect the turbo and found the flapper was in awful shape. The shaft was loose in the casting and the flapper had partially disappeared. So the old turbo was toast after 191k miles even though it wasn't burning any oil. I'm guessing the disintegration of the valve was due to a crack in the casing where the valve seats causing a hot spot to form. Finally I found a bunch of hairline cracks in the exhaust manifold which prompted me to replace that with a good used unit.

I took it for a test yesterday and here are the results:

        OLD    NEW    LIMIT
HC GPM  0.5366 0.1547 1.2000 PASS
CO GPM  6.4360 1.5491 15.000 PASS
NOx GPM 2.7936 0.3087 1.5000 FAIL

HC and CO both dropped by about 75% while NOx dropped by about 90%. Needless to say I'm pleased with the results. My theory on the failure is that due to the non-functional turbo, the tester really had to flog the engine to keep it on the test sequence. This in turn resulted in higher combustion chamber temps (high NOx) and inefficient burning (high HC and CO) of the fuel. It sucked dropping all that cash on this car but the end result made the car better than it ever was since I bought it.