m.wr

Hi guys, new guy on the forums here (and new to Audis in general). Let me preface this by saying that I have spent many hours on google and several different Audi forums trying to learn as much as I can and find the source of my problem on my own. There is nothing worse than a new guy with an obvious problem who wants to be spoon fed a diagnosis and solution. I can assure you that is not the case here. However, I am quickly getting close to exhausting a fairly long list of possibilities so I am finally turning to the experienced guys for help. I have tried to be thorough with my diagnostic checks and troubleshooting, but I’m hoping there is something non-obvious that I have overlooked and can be caught by you. I will try to be as detailed as possible here.

STORY
I bought a cheap non-running 2001 A4 quattro M/T 2.8 from a friend. He had bought it with the intention of fixing it but found a different project instead. The previous motor had skipped either a timing belt tooth or cam chain link (don’t know which) and bent some valves. It would run but extremely rough and was belching smoke. I couldn’t find the exact motor locally (a DBW ATQ M/T) but I was able to find a DBC AHA M/T out of a crashed car (front end collision, no visible damage to any part of engine) and had a friend (who is an Audi enthusiast and mechanic by trade) swap the motor in. This required cleaning out and swapping the intake manifold along with throttle body, injectors, and fuel rail from the blown ATQ to the AHA, in order to preserve the DBW. This swap has successfully been done in the past according to several forum reports. New manifold gaskets, battery, clutch (reused flywheel), timing belt, pulleys, thermostat, and water pump were done at this point.

SYMPTOMS
These are hard to describe accurately because they are so variable, so please bear with me. The car has a severe misfire and will not drive. I’m sorry if the following description sounds imprecise, but it is completely unpredictable to me and I have not been able to find a consistent pattern.

• Sometimes when the car is started, it immediately misfires on one or more cylinders and continues to do so the entire time. However a few times it has smoothed out and stopped misfiring until revved, at which point it misfires again.
• Sometimes the car will start and idle perfectly. It will reach operating temp and purr as long as it is left undisturbed, even 45 minutes or more. Once revved it will start misfiring and inevitably die. It continues to misfire upon return to idle, even though it was running clean before revving.
• Often, the first start of the day is strong and immediate, and the motor briefly sounds healthy. Sometimes it will stay that way until revved, other times the misfires start on their own after a few seconds.
• Usually once it starts misfiring, the more the car is revved the worse it will run – it will feel as if it is getting heavier and heavier and eventually no amount of throttle will keep it from dying.
• First crank of the day is usually strong and immediate. If the car has recently died or been turned off because of misfires, subsequent cranks are slow and seem very laborious. Sometimes it is not possible to start the car.

CODES
The car has never thrown any codes besides P0300 and P0301-P0306. No sensor codes, no timing codes, no fuel codes, no EGR or EVAP codes, no O2 codes. This is part of the reason this problem is so difficult to track down. Every time I make a change under the hood, I clear the codes so that I can see if the change has had an effect. A few important points:

• The misfires do not appear to be consistent. They do not always occur on the same cylinder(s), and in fact it frequently changes. From what I have seen there are no cylinders which never misfire, and there are no cylinders which always misfire. Some misfire less often than others.
• The misfires do not always happen on cylinders which run off the same ignition coil. Sometimes only one of the cylinders on a particular coil will misfire, sometimes neither, sometimes both.
• I have never yet seen an entire bank misfire. It is usually 1-2 cylinders per bank. Sometimes bank 1 will run totally clean, other times bank 2 will run totally clean, other times neither.

DIAGNOSTICS

• Compression is good, 160-175 on all cylinders. Leak down not checked.
• Fuel pressure is 50+ psi at all times, even when misfiring. Fuel pump passes Audi FSM spec for flow. Swapping fuel rail and FPR (but not injectors, which are different) from donor motor has no effect.
• Cycling the fuel pump several times before cranking (to build up fuel pressure) has no effect. Applying artificial vacuum or pressure to FPR nipple (to decrease/increase fuel pressure respectively) has no effect. FPR does not leak vacuum.
• All injector impedances within spec.
• Fuel rail lifted, positioned over 6 solo cups, and engine cranked. All injectors fire neat conical spray patterns. After 10-15 seconds of cranking, all cups have same amount of fuel inside (eyeballed). No apparent sticking open or closed during this test.
• Smoke test shows no gross intake leaks. Have not noticed any change in idle while spraying carb cleaner all over intake path while motor running. I did not try doing this on the lower injector seats, however…maybe I should.
• When not misfiring, motor has very steady vacuum in throttle body. I believe it was around 15 inHg but I can check again to find the exact value if this sounds low.
• Vacuum reservoir, EVAP purge valve, manifold changeover actuator, and EGR valves vacuum tested with engine off. All good except right side EGR valve, which leaks vacuum, but I am not yet sure if it is at the diaphragm or the hose nipple. I disconnected the hose from EGR solenoid and capped the nipple (basically disabling EGR) with no effect.
• Unplugging MAF does not improve misfires. MAF reads 4-5 g/s at idle.
• Unplugging front o2 sensors does not improve misfires.
• Running with unbolted catalytic converters (in case they are clogged) does not improve misfires.
• Throttle body is clean and moves as it should. Throttle body calibration done every time battery is disconnected.
• Got the engine to idle clean and went through all accessible wiring in engine bay – moving it, shaking it, bending it, etc. in order to expose possible shorts. No effect.
• No oil in spark plug wells.
• Air filter in good condition.
• No noticeable pressure or suction at oil cap opening at idle.

WORK DONE / PARTS REPLACED

• New NGK spark plugs (the 4-prong type which cannot be gapped). Not sure what the exact number is, I can find out.
• Three sets of coils have been tested: 1) the aftermarket coils that were on the ATQ when I got the car, 2) new aftermarket coils that I bought, and 3) OEM coils that were on the AHA donor motor. Car misfires on all of them. Resistance of all coils is within spec.
• New aftermarket spark plug wires. Resistance within spec. Double and triple checked that they are not crossed.
• Ignition noise suppressor tab trimmed and insulated from ignition coil frame.
• All old gas pumped out of tank and replaced with fresh 93 octane plus a bottle of Techron.
• MAF and IAT sensors cleaned, no effect.
• All vacuum lines in engine bay replaced with 1/8” (3.18mm) silicone tubing.
• Valve cover/PCV pipe breather hose inspected, cleaned out with gasoline, dried, replaced.
• “Suction pump” and green check valve were bad, both letting air past check valves. Replaced with new OEM, no effect. New hose clamps used all over these parts, brake booster hose, “F-hose”, etc. Holds vacuum.
• Manifold changeover actuator removed, cleaned/lubricated, vacuum tested.
• EGR and manifold changeover solenoids vacuum tested.
• Both sets of cam position sensors (ATQ/AHA) installed and tested with no effect.
• New OEM crankshaft position sensor, no effect.
• New aftermarket coolant temp sensor, no effect. OBD reader shows coolant temp reading reasonable values while cold and rising smoothly to operating temp.
• Used ECU swapped in, same P/N out of a manual trans car, unsure if Quattro. No effect.
• UPDATE 11/6: replaced bent valve in #3, engine still misfires. See post #8 for details.

WORK NOT YET DONE / THEORIES / SUSPECT PARTS

• Wiring/harness/short issue. I really hate this possibility but it is slowly rising in probability.
• Try different spark plugs?
• Bad MAF? Seems unlikely because of normal idle reading and no improvement when unplugged.
• Non-OEM plug wires may be arcing/shorting to block.
• Injectors may be sticking/flooding when misfires happen. Lower seats could be leaking vacuum, but not likely since symptoms are roaming/intermittent.
• Flywheel reluctor ring could be damaged/defective? Not sure what it looks like or how this could happen. Unlikely because symptoms are roaming/intermittent.
• EGR valves could be dirty/stuck.
• Valves could be leaking. Unlikely because symptoms are roaming/intermittent.
• Fuel filter could be clogged. Unlikely because of fuel flow and pressure tests.

If you have read this far, I deeply thank you for your patience. I am trying to solve this mystery as intelligently as possible in order to avoid throwing more parts at it than I already have, which I know is not the best way to go about it. I welcome any input/ideas/suggestions and will try to answer any requests for additional info to the best of my abilities. I will also update this post whenever I find new information.

Thank you!

m.wr

Update: yesterday I warmed the car up thoroughly and was gently varying the RPMs while watching misfire counts. Cylinders 3 and 6 (closest to firewall) were the ones most often misfiring. Spots of 2 and 5 (middle) occasionally. Almost never on 1 and 4 (furthest from firewall). Is it possible that a malfunction in the EGR valves, which are back by the firewall and feed closest to cylinders 3 and 6, are causing this?

m.wr

So if I understand the Haynes manual correctly, it looks like only the older 12v motors had EGR. The newer 30v have SAI, which does not run into the intake manifold and shouldn't have any effect on combustion, just emissions. Does anybody know if that's correct? I'd rather not spend time removing those two valves if they can't possibly cause what I'm seeing.

m.wr

UPDATE 10/2: Lately the misfires have stopped roaming and are happening pretty consistently on cylinders 3 and 6, which are adjacent to each other in the firing order. They are the two cylinders closest to the firewall and are ignited by different coils. I plan on swapping injectors with 1 and 4, and if that doesn't affect it then I will do a leak down test. Maybe valves are leaking?

m.wr

Not much activity here Still hoping for some help so I will continue to update on the off chance somebody reads and has any tips to contribute.

UPDATE 10/7: I decided to check leak down and started with #3...things got weird. 100% leakage. Yes, made double and triple sure it was at compression TDC. All other cylinders normal. Checked compression next and got 0 on cylinder 3, all other cylinders normal. This is strange because when compression was tested a few weeks ago it was normal across all cylinders. After much testing, sniffing, probing, and removing the valve cover I have isolated it to a massive leak from the intake side. I removed the fuel rail and held a small down feather in the #3 injector hole. When trying to pressurize cylinder 3, the feather waved around in the draft. I tried it in another cylinder and the feather stayed completely still.

I put a borescope in the cylinder and saw nothing unusual. Then I put it in the injector hole and looked at each intake valve individually, while spinning the crank and watching them move up and down. It didn't look like any of them were sticking open, although the resolution was not great so I may have missed it. Maybe carbon on the valve seats? Is there any recommended way to clean/seal the valve seats without removing the head?

4Driver4

Here's what I would do:

Before you do anything else, change the coolant temp sensor. It's a long shot, but worth a try.

If that doesn't work, check the timing on the chains and belt. If good, remove the head and fix the bent valve at 3.

m.wr

CTS was already tried, no effect.

Since my last post I did a new test. I waited until it was dark and put the borescope in the #3 injector hole as before. Then I rotated 3 to TDC and shined a flashlight into the spark plug hole. I could see a thin crescent of light coming in past the middle intake valve. This means I have a valve that is not fully closing for some reason. I confirmed this by rotating the crank while watching the valve move up and down, never fully closing.

The possibilities I can think of are: bent, burned, or sticking valve, broken/weak spring, stuck lifter. Borescope confirms no valve/piston contact (unless it was light enough to not mark the piston in any way) so I doubt the valve is bent, although it is possible. Burned seems unlikely because the valve still looks perfectly circular. Maybe it's just my optimism speaking, but I'm thinking broken spring or stuck lifter.

In any case I have to remove the camshaft and inspect the valve and spring. Hopefully head removal will not be necessary. If anyone has seen this before, please do chime in.

m.wr

Well, middle intake valve on #3 was indeed bent... I spent all weekend taking the head off and replacing the valve. Checked #3 piston but saw no evidence of contact so I don't know how it happened. Nothing unusual in the head. Cams, lifters, tensioners all good. New HG, checked timing on both cam banks, timed the car using crank pin and locking bar, sealed everything up tight.

Still misfires randomly. Idles smooth as butter but starts to misfire when revved. Roaming cylinders, intermittent. Sometimes I can rev it a few times and it'll be ok, then something sets it off and it starts to misfire badly. If I let it idle like that the misfire eventually fades out and returns to smooth idle. Does anybody have any ideas? I feel like it's a SIMPLE, SMALL thing that I am missing and this car would run great if I could just figure out what it was.

clancy

You sound way more knowledgeable than I, but I'll comment nevertheless. Have you looked at fuel trim numbers? You said you checked fuel pressure and volume?

m.wr

Any input is appreciated! Fuel pressure is normal at all times. If by volume you mean the fuel pump volume test, that came back normal too. Fuel trims look normal at idle but to your point, I don't think I have checked them WHILE the car misfires. Would they even change then? LTFT would probably not change, but I suppose STFT would. Does this car do STFT? I'm familiar with Subaru ECUs, which do both short and long term trims.