PDA

View Full Version : 2011 S5 4.2 V8 misfiring on Bank 1 only



volvofan
10-03-2018, 04:02 PM
My car has about 91k, and also has an APR Stage III kit although I donít think thatís relevant.

Drove fine one day, then woke up stoopid the next morning. VAG-COM showed all running fine and dandy at idle, then anything more than about 5% throttle would cause misfires to start to pile up on bank 1 (with MIL flashing the whole time) until EPC light eventually came on.

Ruled out injectors, coils, and plugs (all four didnít go bad at once, for sure), low pressure side of fuel system (feeds both banks) and MAF sensor (feeds both banks).

I swapped camshaft position sensors between banks 1 and 2, as well as the variable valve solenoids between banks 1 and 2. Misfires stayed with bank 1 each time.

I have NOT yet messed with the fuel regulator valves (N276) nor any of the less accessible bits on the front of the engine (I know there are some other cam-related sensors there).

I have not checked fuses nor traced wires, but I would think that if it were an electrical issue, then the coils/injectors would either be totally dead, or work just fine regardless of load. Instead, idle is okay, itís just when I start to give it throttle that things go wonky.

Thoughts?

Thanks in advance!

-Jon

Ape Factory
10-07-2018, 08:16 AM
If it's just one bank, you'll have to isolate everything that's specific to that side. So off the top of my head;
-high pressure fuel pump
-02 sensor (pre-cat would be my guess, not post)
-something with the variable intake manifold system (broken or disconnected arm on one side?)
-intake manifold runner position sensors (you can swap those I believe),
-intake manifold runner control vacuum actuator
-vacuum leak on that side of the motor, either in the runner to intake port area or the manifold runner actuator system as there's quite a few lines and check valves to split or stop working

It's hard to imagine any throttle body issue causing problems on just one side of the motor but you never know...

Any codes other than bank one misfire being picked up by VCDS? And it's all four? If you log them individually, they're all misfiring?

3g_bishi
10-07-2018, 09:02 AM
O2 sensor on that bank upstream would be my first guess.


Sent from my iPhone using Audizine (http://r.tapatalk.com/byo?rid=87676)

volvofan
10-07-2018, 09:11 AM
No VCDS codes except for the misfire codes (all four cylinders) and cylinder deactivation. Sheís supercharged, so no more variable intake runners or flaps or any of that jazz. Surprised I would not get a code for cam sensor or fuel pressure regulator or O2 sensor if one of those were the problem, but maybe? Can you swap the N276 valves from one HPFP to the other easily? Otherwise, the HPFPs are the same, yes?

-Jon

Ape Factory
10-07-2018, 06:12 PM
Not sure about swapping the valves but I'd think the HPFP's are side-specific due to layout so you can't swap them. Plus there is a fuel pressure sensor and like you said, if there were something amiss, I'd assume it would throw a code.

Have you done a compression test? Only other thing I could think would be an bad ignition driver circuit on the ECU itself. But that would cause a code too, me thinks.

volvofan
10-07-2018, 07:56 PM
N276 valves swapped (was easy). No effect. No compression test yet. Swapping upstream O2 sensor tomorrow AM.

-Jon

volvofan
10-08-2018, 12:32 PM
New O2 upstream sensor, no luck. Swapped HPFPs. Also no luck.

Will do a compression test, but would be surprised to find I had lost compression in an entire bank of cylinders at once.

Could be electrical, but if so (ie with injectors or coils) then why would I not be getting any misfires at idle?

What outputs should I be looking at or logging in VCDS?

Thanks for the continued assistance, AZ!

-Jon

Ape Factory
10-08-2018, 07:11 PM
Are there two camshaft sensors per bank on the S5's V8 like there are on the RS5? (Edit, there are two)

Is it possible to switch back to the OEM software or reflash the APR tune to rule out some sort of file corruption?

Did you rule out the MAF because it's just happening on one bank? What about the throttle body?

Pressure relief valve on the bad side. Still think that would be caught by the fuel pressure sensor.

There are two camshaft adjustment valves per bank.

Throttle valve control module. Wonder if it's out of whack?

I'm assuming you can visibly hear the car misfiring/stumbling and it's obviously not running correctly?

This just happened out of the blue? No maintenance, nothing taken off the engine and put back on?

volvofan
10-09-2018, 06:26 AM
Q1 - Are there two camshaft sensors per bank on the S5's V8 like there are on the RS5? (Edit, there are two)
A1 - Yes (one intake and one exhaust per bank). Swapped both. There also "SHOULD" be specific trouble codes for sensor failures that get flagged in VCDS if one of them fails, and I did not get any such indicators.


Q2 - Is it possible to switch back to the OEM software or reflash the APR tune to rule out some sort of file corruption?
A2 - That's not a terrible idea. There is an APR dealer nearby.


Q3 - Did you rule out the MAF because it's just happening on one bank? What about the throttle body?
A3 - Yes x2. Can't figure out how MAF or TB issues could possibly affect only one bank, especially since there's a supercharger in between those two components and the cylinders.


Q4 - Pressure relief valve on the bad side. Still think that would be caught by the fuel pressure sensor.
A4 - What's a pressure relief valve? Is that the N276 bolted directly into the HPFP?


Q5 - There are two camshaft adjustment valves per bank.
A5 - Hmmmmm... I swapped the two solenoids that face straight up right in between the intake manifold and the valve covers towards the back of the engine. Is that what you're talking about?


Q6 - Throttle valve control module. Wonder if it's out of whack?
A6 - Again, anything with the throttle body/valve should affect both banks.

Q7 - I'm assuming you can visibly hear the car misfiring/stumbling and it's obviously not running correctly?
A7 - Oh God yes. It's awful. It bucks, hesitates, barely gets up to the speed limit after ginger application of throttle and frequent resetting of the DTC codes to keep the EPC light off and to keep it from shutting the cylinders down completely.


Q8 - This just happened out of the blue? No maintenance, nothing taken off the engine and put back on?
A8 - Yes, out of the blue. No recent maintenance. I had a long drive up to VT from MA (about 3 hours) and she was running like a Swiss watch. The next morning, she ran like dogshit. Did no maintenance or anything in between (I had just run a Spartan race so it was the last thing on my mind!).

Thanks for continuing to help me troubleshoot! Great questions!

-Jon

Ape Factory
10-09-2018, 07:19 AM
The only reason I mention the MAF and throttle body is because it's German and nothing seems to make sense with these cars. Especially since you're not throwing any codes. It's nuts!

Do you have the Audi factory service manual for ignition and fuel injection?

volvofan
10-09-2018, 07:31 AM
No, I don't have that service manual. Is it floating around on PDF somewhere, or is it something I can access through ElsaWIN and/or VCDS? If so, I have both of those.

-Jon

Ape Factory
10-09-2018, 07:32 AM
Check your PM's.

volvofan
10-09-2018, 07:33 AM
RGR just did, thanks much!

Ape Factory
10-09-2018, 07:41 AM
Have you checked all of the fuses?

Is there anything internally on the APR supercharger which might cause one bank not to get air? I'm assuming there are built-in heat exchangers. There's no smoke coming out the tailpipe, right? Nothing white or black? No coolant smell? Check your coolant level.

Ape Factory
10-09-2018, 07:45 AM
Factory service manual
https://www.audizine.com/forum/showthread.php/817161-How-to-download-factory-service-manuals-for-your-Audi?p=13335841#post13335841

volvofan
10-09-2018, 07:54 AM
Have you checked all of the fuses?

Is there anything internally on the APR supercharger which might cause one bank not to get air? I'm assuming there are built-in heat exchangers. There's no smoke coming out the tailpipe, right? Nothing white or black? No coolant smell? Check your coolant level.

Checked all fuses under the hood by the ECU. Didn't check the ones on the driver and passenger sides of the dashboard or in the trunk, but I looked over the list and can't find a single thing on there that should pertain to a single bank? Will add that to the "just in case" list and will check those later today.

Re: Supercharger, yes, there's a heat exchanger, but I can't imagine that something could have gotten in there and clogged up the entire heat exchanger so perfectly as to block the entire bank. I have some NPT plugs positioned over each cylinder (for inspection and possible carbon cleaning or water injection) that I can pull to double check.

No black or white smoke. No coolant loss or coolant smell. Will double-check the air-liquid intercooler coolant level to ensure no loss there, but IATs don't indicate a problem. It is a completely separate system from the engine's cooling system.

-Jon

volvofan
10-10-2018, 10:58 AM
SUPER helpful and detailed info in those manuals... thanks so much! It looks like I only swapped one of the camshaft positioning valves / solenoids (the intake sides, sticking straight up), so that's still one more possibility. My "to-do" list is now (because I didn't do anything yesterday):

1) Check air-liquid intercooler fluid level.
2) Swap exhaust-side camshaft positioning valves.
3) Check ALL fuses.
4) Inspect supercharger intake tract.
5) Re-flash ECU.
6) Compression Test.
7) ????? Back to drawing board?

-Jon

Ape Factory
10-10-2018, 02:02 PM
I donít know the construction of the APR blower (didnít even know they made one until you mentioned it!) but I would check the blower chassis for cracks. Was looking at photos of the PES supercharger and it had developed cracks in the top cover. Just another random thought. Need to rename this the red herring thread.

volvofan
10-11-2018, 03:29 AM
Chassis is solid as a rock, thank goodness!

Ape Factory
10-11-2018, 09:07 PM
Take a look at the pressure valve on that bank that I mentioned earlier. It's in the FSM. Not sure how expensive that part is or how easy it is to change out with the SC...but I keep coming back to that. I have a feeling as they say.

volvofan
10-12-2018, 04:02 AM
Pressure valve? Are you talking about fuel pressure? I think thatís controlled by the N276 valve on the HPFP and I have swapped those already. Under the SC, I think there are pressure sensors but no valves.

-Jon

Ape Factory
10-12-2018, 05:16 AM
Go to page 7 of the fuel/ignition FSM, to 2.1.2. It's number 10 on that diagram. Just to make sure we're talking about the same thing. There's no sensor hooked up to it. I'm guessing if it failed and pressure was either too high or too low, the fuel pressure sensor would throw a code.

Also, you've checked the oil to make sure there's no coolant in it? Just thinking of the other forum member who just had his S5 SC'd and blew a head gasket. A small leak might not show any evidence out the tailpipe. Not sure how all four cylinders would then start misfiring but this whole thing is a mystery.

Have you tried to log various parameters with the engine running? You could do that bank for fuel and timing and see if something weird is happening. Maybe do knock as well. It really sounds like it's a fueling issue though. Everything points in that direction. Wait...our hpfp's are gear driven right? Any visible issues with the drive gear? Still think it'd throw a code though.

volvofan
10-12-2018, 06:10 AM
Aha. Nope, definitely NOT talking about the same thing. Yes, that would definitely be a pain in the ass to get to in order to check/swap!

I have logged knock, low side fuel pressure, and high side fuel pressure. Nothing out of the ordinary, I think? I have also compared timing bank 1 vs. bank 2 and I'm pretty sure they were consistent? Maybe I'll go back and double check that. Of course, the computer only logs what the sensor tells it, so if one of the sensors is wonky (without being completely open circuit or dead short) then maybe that's the problem.

HPFPs are cam-driven. I already swapped both HPFPs with no effect. I also inspected the cams and cam followers and noticed no abnormal signs of wear or cause for concern.

Weather has been shit for the past three days so I haven't gotten to anything on the to-do list. Good point about coolant in oil, but I'm not losing any from the reservoir so it's highly unlikely. I did check the SC coolant circuit and it was full to the brim.

New list:

1) Check oil for evidence of coolant (head gasket leak).
2) Swap exhaust-side camshaft positioning valves.
3) Check ALL fuses.
4) Inspect supercharger intake tract.
5) Datalog and post A/F ratio, intake and exhaust timing for both banks while idling and driving.
6) Re-flash ECU.
7) Compression Test.
8) ????? Back to drawing board?

volvofan
10-13-2018, 04:52 AM
1-3 done. No oil in coolant, ALL fuses okay, and swapping cam positioning valve for exhaust bank 1 (swapped with intake bank 2) had no effect.

Also did #5... checked a bunch of engine parameters from A/F ratio to intake and exhaust cam timing to knock and timing retard. Compared their requested (by ECU) and actual (reported by sensor) values and compared bank 1 to bank 2. Found no obvious anomalies.

Stupid question but at this point Iím not ruling anything out... bank 1 is on the LEFT side of the engine as you pop the hood, stand at the front of the car, and look down at it... ie the side with the air intake, NOT the side with the coolant reservoir, correct?

Off to do a compression test...
-Jon

volvofan
10-13-2018, 07:30 AM
Compression test shows a consistent 215-240 psi across all cylinders (after 6-10 cranks).

Going to completely swap coils and plugs from bank 1 to bank 2 this time (just swapped within bank 1 before... easier to get to).

Ape Factory
10-13-2018, 07:55 AM
This is nuts. Do you have any hair left? I'm literally praying you find an easy solution and I'm not religious.

volvofan
10-19-2018, 12:59 PM
Okay, here are some datalogging results. Lambda seems to indicate a problem with fuel, even though I logged fuel PSI (both low pressure and rail pressure) and saw nothing out of the ordinary. All three graphs are data from the same run (notice the RPMs and speed are the same). Misfire counts cleared when I reset the computer to get it out of limp mode.

Very strange how the lambda correction goes in OPPOSITE DIRECTIONS on the two cylinder banks... like one is going too lean and the other is going too rich. That could mean one side is getting too much air and the other is not getting enough. OR, it could mean the same thing... but with fuel. Augh!

At first, I was thinking this is definitely a fueling issue. More research into the fuel system shows that on the APR Stage III kit (and maybe on the stock system, too), the distribution block combines the feeds from the two HPFPs together before sending the output down a SINGLE line, which then splits (under the supercharger, of course) to feed both banks. Similarly, return lines (after the pressure relief valve) combine together and come back up from under the supercharger as a single return, which then gets redistributed back to the HPFPs.

Perhaps some sort of contaminant worked its way down into the line, then went down only ONE side to foul up the injectors on that side? If it didn't completely block things up then there would still be enough fuel to idle, but anything more than that and it would start to die out. IF the contaminants reached the actual injectors themselves (not just in the line/rail), then this wouldn't show up in datalogging the high pressure fuel sensor's reading, because the sensor is actually attached to the bank 1 rail.

If it's an air supply issue, then I'll probably be able to tell before completely removing the supercharger. I'll pull the intercooler and have a look to see if anything is blocking one side or the other. If nothing obvious shows up, then I'm afraid the supercharger's going to have to come off.......

Thoughts?

-Jon

Lambda (https://imgur.com/P7eZeNC)

Misfires (https://imgur.com/IAnudiR)

Timing Retard (https://imgur.com/PKV581Q)

Ape Factory
10-19-2018, 06:44 PM
Well that sucks. I'm sure it's not fun to remove the SC. I can't see the attachments btw...hoping you're on the right track.

volvofan
10-20-2018, 05:41 AM
Fixed links (hopefully) and thought some more about what the problem could be, updating previous post. Off to do #4 on the list (inspect supercharger intake tract). If no luck, then I'll have to pull the supercharger to inspect lines, rails, and injectors.


-Jon

Ape Factory
10-20-2018, 07:14 AM
What did you use to log, the VCDS? If so, could I see the raw, un-graphed files? Just shoot them to me in an email.

volvofan
10-20-2018, 07:30 AM
Yes, thatís what I used. Iíll shoot them over to you.

Thanks again!

BTW, just got the bank one intercooler core out of the SC housing. Nothing obvious I can see yet, but I just ran inside to grab my borescope.

-Jon

volvofan
10-23-2018, 08:28 AM
Okay, supercharger is off. No obvious clogs/blockages in the supply or return lines, or the fuel rails themselves.

I was going to try to DIY clean the injectors or maybe send them out, but given what a PITA it is to get to them and the mileage on the vehicle, I decided it best to invest in a new set in order to ensure a potential problem with them is eliminated. They were definitely "yucky" and had some particulates in their little integrated filter basket but no obvious clogs around the nozzles themselves. Fun fact... those tiny micron-sized screen filters built into injectors A) do NOT come out via screwdriver/pliers/small spanner wrenches (as far as I could tell) and B) the filter media is VERY easy to tear/break through. Ask me how I know. Anyways, injectors are often changed by folks every 2-3 carbon cleanings (40-80k miles) and I bet this is the original set, so... not a terrible idea to replace them anyhow. At least, that's how I'm rationalizing this dive into the engine as not a "wasted trip." It's just... preventative maintenance :) New set arrives tomorrow.

I'm not going to replace the pressure relief valves, because they cost even more than the injectors do and they're hard to source. I will swap them from one bank to the other so that in the unlikely event that they are the problem, I'll at least be able to identify that.

I did use carb/choke cleaner to blow out the distribution block and all the lines.

I'm also going to replace the fuel filter just in case something was plugged up or somehow shed contaminants downstream. Will then blow out that line from the fuel tank to the distribution block before reassembly just to ensure any possible residual contaminants are purged.

Finally, I'll be sure to carefully inspect all wiring and connectors to make sure there isn't a break or short somewhere.

Fun fun fun......

-Jon

volvofan
10-30-2018, 07:32 PM
Well, that's almost a thousand bucks and several hours worth of labor down the drain. New injectors installed, pressure relief valve swapped over, fuel rails/lines/distribution block checked for obstructions, everything in the engine valley all cleaned up, injector harnesses inspected... no effect. Same problem, same cylinder bank.

Probably time to delve into the wiring diagrams at this point. One of my buddies mentioned that if this problem showed up overnight (in Vermont) then maybe a small animal crawled inside the engine bay somewhere to get warm and snacked on a wire?

Also possible that something went awry in the ECU?

-Jon

Ape Factory
10-30-2018, 09:01 PM
It isn't unheard of for a rodent to chew on wires. I had a Maserati Merak way back when that I was attempting to restore to running order and rodents had destroyed about half of the wiring harness all over the car. What a nightmare. Like double just owning an early 80's Italian car sort of nightmare.

It's very possible for the ECU to go bad, just a portion, like the injector drivers. But honestly it does sound like some sort of wiring/ground thing at this point.

Ape Factory
10-31-2018, 07:43 PM
Hey just came across a thread on the Audirevolution forum where an RS4 (B7) was having misfires on two cylinders due to low voltage at the fuel injector. Now his car threw a code but after trying to chase things down, it turns out the injector driver in the ECU went bad. He tried to chase down a possible short and eventually took it to an Indy shop. They found no short but diagnosed the bad injector drivers.

The poster also mentions doing some research and that most with this issue suffered from a bad relay.

volvofan
11-01-2018, 05:22 AM
This is obviously a heavy dose of confirmation bias at play, but the more I look into "Audi S5 ECU problems" the more threads I find showing bad ECUs on Audi and VW platforms. I have NEVER had an ECU "go bad" in any of the other cars I've owned, driven, or helped friends work on. It always seems like an expensive part that the dealer chucks at a difficult-to-diagnose problem, not something that ACTUALLY fails. Well, maybe not the case here? It also seems to be reported with greater frequency on tuned vehicles, especially when they've had to crack the ECU open to tune it. Not sure how APR does their tuning, but I'll be finding out today, I guess. I may pull the ECU today or tomorrow and see if I can have a peek inside to identify any obvious bad components. That would certainly be easier than chasing the entire wiring harness through the dashboard / engine compartment looking for the problem.

Also, if the injectors on one side weren't firing properly, that would cause the issue we were seeing with the lambda - the cylinders with properly functioning injectors would tend to go rich as they get more fuel than expected, while the cylinders with malfunctioning injectors would go lean... right?

-Jon

ips_cou
11-01-2018, 11:58 AM
Been trying to follow this, just super busy.

I haven't messed with some of this in a while but a few things:

Bank 1 Vs Bank 2; I thought Bank 1 was driver side. Some ways I think you could verify is, unplug front O2 sensor or possibly one COP and start the car and see if that gives a 'Bank' in the fault message.

You said you ruled out injectors? Swapping all from one side to the other?

Their is not to much to the engine harness(especially for the injection system/COP), runs from the ECU, out the top front corner of the ECU box into one big harness. From there it splits off near the components with a few, if any connectors in-between. Doubtful their is any damage to the wires inside the protective covering, which is over most of the wires. One of the things to look for is where the wire come out of the protective coverings is the wire jacket it self failing off/cracking. I experienced this, I don't know of it specifically causing a issue, but I sealed it up with liquid electrical tape(something like ~10 spots on the engine harness, typically near sensors).

Also, When removing/installing the transmission for the tenth time(was dealing with a clutch issue, thank you OEM and Aftermarket Mfgrs!), The harnesses for the injectors are right above the trans/clutch bell housing(below and behind the throttle body/PCV oil separator), and it caught the driver side harness and broke the base of the male connector, which partially pulled out from the rest of the connector(unbeknownst to me at the time as everything is so crammed in) causing intermittent low oil pressure warnings(our cars only have a dummy switch, installed a real pressure gauge now). So this may be a area to look at.

Some food for thought on the ECU possibly failing: I currently have a issue with my Advanced key and the fault message was ~'fault to ground', even though I checked all the wiring and found nothing but pristine wiring. I have read of the BCM(Body Control Module) having components on the PCB fail. As of right now this is a open issue for me, that I would like to remedy, I had other issues come up with higher priority. I also recall someone else having a no start issue being remedied, at least for a short time, by replacing the ECU.

On the tuning method: I believe with these cars at least, APR does it like most others, a OBD2 port flash. My none APR tune was OBD2 flashed.

volvofan
11-01-2018, 12:23 PM
Been trying to follow this, just super busy.

I haven't messed with some of this in a while but a few things:

Bank 1 Vs Bank 2; I thought Bank 1 was driver side. Some ways I think you could verify is, unplug front O2 sensor or possibly one COP and start the car and see if that gives a 'Bank' in the fault message.

You said you ruled out injectors? Swapping all from one side to the other?

Their is not to much to the engine harness(especially for the injection system/COP), runs from the ECU, out the top front corner of the ECU box into one big harness. From there it splits off near the components with a few, if any connectors in-between. Doubtful their is any damage to the wires inside the protective covering, which is over most of the wires. One of the things to look for is where the wire come out of the protective coverings is the wire jacket it self failing off/cracking. I experienced this, I don't know of it specifically causing a issue, but I sealed it up with liquid electrical tape(something like ~10 spots on the engine harness, typically near sensors).

Also, When removing/installing the transmission for the tenth time(was dealing with a clutch issue, thank you OEM and Aftermarket Mfgrs!), The harnesses for the injectors are right above the trans/clutch bell housing(below and behind the throttle body/PCV oil separator), and it caught the driver side harness and broke the base of the male connector, which partially pulled out from the rest of the connector(unbeknownst to me at the time as everything is so crammed in) causing intermittent low oil pressure warnings(our cars only have a dummy switch, installed a real pressure gauge now). So this may be a area to look at.

Some food for thought on the ECU possibly failing: I currently have a issue with my Advanced key and the fault message was ~'fault to ground', even though I checked all the wiring and found nothing but pristine wiring. I have read of the BCM(Body Control Module) having components on the PCB fail. As of right now this is a open issue for me, that I would like to remedy, I had other issues come up with higher priority. I also recall someone else having a no start issue being remedied, at least for a short time, by replacing the ECU.

On the tuning method: I believe with these cars at least, APR does it like most others, a OBD2 port flash. My none APR tune was OBD2 flashed.

Dumb question - what is COP?

A healthy amount of internet research indicates that as you stare down at the engine compartment while standing in front of the car, bank 1 is on your left (passenger side). Could still be wrong, I suppose.

I ruled out injectors by replacing all eight as long as I had the supercharger off.
I swapped the pressure relief valves from one rail to the other while I had the supercharger off.
I swapped intake & exhaust cam position sensors and positioning valves as well, from one side to the other... I did not swap the exhaust cam positioning valve or sensor on the driver's side.
I swapped HPFPs and the regulating valves attached to them (this was before I realized they fed into a single line via the distribution block before splitting back out to each rail and that was unlikely to be the problem)
I replaced the passenger side front O2 sensor. I did not swap/replace the driver side front O2 sensor. I did not mess with the rear O2 sensors as I believe they only are used to assess catalytic converter performance.
I swapped coils and plugs from one side to the other (at the same time I did the compression test).

Have never had to separate the engine from the transmission, thank goodness, but I checked that area of the harness over pretty thoroughly while I had the SC off, as it was quite visible and accessible without the TB and oil separator and bits in the way. No apparent issues I could see.

I just got off the phone with APR, and they confirmed that their tune is done via an OBD2 port flash.

Thank you all for the brainpower expended on my behalf!

-Jon

ips_cou
11-01-2018, 01:08 PM
Dumb question - what is COP?

A healthy amount of internet research indicates that as you stare down at the engine compartment while standing in front of the car, bank 1 is on your left (passenger side). Could still be wrong, I suppose.

I ruled out injectors by replacing all eight as long as I had the supercharger off.
I swapped the pressure relief valves from one rail to the other while I had the supercharger off.
I swapped intake & exhaust cam position sensors and positioning valves as well, from one side to the other... I did not swap the exhaust cam positioning valve or sensor on the driver's side.
I swapped HPFPs and the regulating valves attached to them (this was before I realized they fed into a single line via the distribution block before splitting back out to each rail and that was unlikely to be the problem)
I replaced the passenger side front O2 sensor. I did not swap/replace the driver side front O2 sensor. I did not mess with the rear O2 sensors as I believe they only are used to assess catalytic converter performance.
I swapped coils and plugs from one side to the other (at the same time I did the compression test).

Have never had to separate the engine from the transmission, thank goodness, but I checked that area of the harness over pretty thoroughly while I had the SC off, as it was quite visible and accessible without the TB and oil separator and bits in the way. No apparent issues I could see.

I just got off the phone with APR, and they confirmed that their tune is done via an OBD2 port flash.

Thank you all for the brainpower expended on my behalf!

-Jon



COP = Coil on Plug. The big red things above the spark plugs [:D]

When in doubt, best to test. Or even just double check before opening any more doors. Also, I've noticed a confliction info. And just flat out wrong info, even from Audi on multiple occasions. The ECU will even give wrong fault messages at times. Their Quality Control leaves plenty to be desired.

Another issue I had, (which doesn't seem like yours, but more food for thought as these cars do weird things with out much reason at times, like bad/intermittent ABS pump not throwing a fault code, causing MMI to shut off at random times when driving steady state)
.. ugh it doesn't end with these cars I'm learning.., was one of my HPFP was on the edge of going out and would randomly not meet the requested fuel pressure under ~30%+ load ..again intermittently.. tracked it down when I found VCDS can log High pressure fuel demand and/VS actual, saw dips correspond with power drops and took the gamble and replaced both being they were original with ~100K on them. Fixed that problem.

May want to try seeing what 'Demand vs actual' params you can find/log in VCDS before spending any more money on components being they add up so quick. Also, although the chances are low, it always possible and does happen, you may get a new part that is defective and it will just add to the problem and possibly hide the actual problem if you fix the original problem latter along in the R&R chain.

Ah also, You said you did a compression test, and I doubt this will prove much being it came out of no where, but a leakdown test may be worth while if you are running out of ideas, benifit is it helps verify the compression test and the condition/life of the cylinders.

volvofan
11-01-2018, 04:29 PM
Coil on plug... aha! Yes, switched those.

I did check several ďdemand vs actualĒ camshaft and fuel pressure values... all spot on.

Thanks,

-Jon

ips_cou
11-02-2018, 12:17 PM
Heres the ~CAUA engine harness. may be of some use.. better visual than the typical graph:
https://audi.7zap.com/en/usa/audi+a5+s5+coupe+sportback/a5co/2009-620/9/971-971062/

Just remember what I said about not always be 100% accurate.

Another site which nav's better is realoem.me but seems to be down at the moment.. happens often.

volvofan
11-02-2018, 12:28 PM
Heres the ~CAUA engine harness. may be of some use.. better visual than the typical graph:
https://audi.7zap.com/en/usa/audi+a5+s5+coupe+sportback/a5co/2009-620/9/971-971062/

Just remember what I said about not always be 100% accurate.

Another site which nav's better is realoem.me but seems to be down at the moment.. happens often.

Hmmmm.... so, the coils/plugs are part of the main harness, but there is actually a connector between the main harness and the two injector harnesses. Will be opening up ECU this afternoon... if nothing obvious, then maybe those connectors are my next stop.

-Jon

ips_cou
11-02-2018, 12:31 PM
Hmmmm.... so, the coils/plugs are part of the main harness, but there is actually a connector between the main harness and the two injector harnesses. Will be opening up ECU this afternoon... if nothing obvious, then maybe those connectors are my next stop.

-Jon

If you dont mind, take some pics. Would be interesting to see inside.

volvofan
11-04-2018, 04:18 AM
Some ECU photos to follow, but here are the results:

It was a bitch to open. Strong adhesive around the perimeter that required a great deal of prying at the corners and sides after generous application of heat via a hot air solder rework gun just to soften... wasnít brave enough to get adhesive hot enough to melt. Once pried a bit to access, use of a razor blade to cut the adhesive bond proved the quickest and most effective means of access. To make it even harder, three heat sink strips (gray) also adhered the case to the board. Took over an hour and an extra pair of hands to force my way in. Not an effort for the faint of heart. ZERO percent chance you could open one of these puppies up without it being obvious to a casual visual inspection down the road. You will have lid/housing deformation and tool marks around the whole perimeter. I sealed everything back up (after reapplying NON-adhesive heat sink paste) using 3M heavy duty double-sided tape.

No signs of burned out components, so the search continues. On to the injector wiring harness connector; will bust out the multimeter first and the scope next if necessary. Just because there was no visual evidence of damage doesnít mean it couldnít still be a bad ecu.

-Jon

volvofan
11-04-2018, 04:21 AM
Photo of exterior w/part number
https://i.imgur.com/MX1i2GS.jpg

Interior under cover
https://i.imgur.com/yN5RprW.jpg

Interior underside
https://i.imgur.com/eLWwRX2.jpg

volvofan
11-04-2018, 12:28 PM
Tested wiring harness:

1) Disconnected the injector harness from the main harness and checked continuity back through each injector... 1.7 - 2.1 ohms on all four.
2) Disconnected ECU and checked continuity from the injector harness connection back to the ECU plug; all eight wires showed no issues.
3) Probed the positive side (assuming injectors are ground-switched) of all four injectors and only showed .7v on each wire. I was able to reach the #5 injector (front right, driver's side) and probed it to compare, and also saw .7v. Very strange... I would have thought I'd find +12V there.

Anyone know if the ECU later revisions are still compatible? There are several used ECUs for sale right now with the proper part number (8T0 907 560) but some are AD, some are AE, and some are AF. I would think that the later revision, the better, but who knows? Maybe it relates to coupe vs. convertible, auto vs. manual, or sport differential vs. non-sport differential?

Thanks,

-Jon

Ape Factory
11-06-2018, 07:06 AM
I would consider sending the ECU out to be tested as the damage may not be visible. Probably pretty cheap to do and I saw one place that had 1-2 day turn around for repairs. Maybe check with APR themselves to see if they can test it since it's their software.

Edit: Things like capacitors (those big orange things) are obvious when they blow but the chips or corrupted ROM, not so much.

volvofan
11-06-2018, 07:22 AM
I would consider sending the ECU out to be tested as the damage may not be visible. Probably pretty cheap to do and I saw one place that had 1-2 day turn around for repairs. Maybe check with APR themselves to see if they can test it since it's their software.

Edit: Things like capacitors (those big orange things) are obvious when they blow but the chips or corrupted ROM, not so much.

APR won't touch it... I asked.

Yeah, I was hoping to see a blown cap or a torched MOSFET chip or something obvious.

I thought about sending it out, but the cost of sending it out (plus being without the car for 3-4 days) for a repair ($100-$300) approaches the cost of getting a used ECU and having it set up "bone stock" by the dealer via programming ($400 or so). With a scope, I may be able to diagnose a failed ECU myself, and just skip straight to replacement.

When they "reprogram" a car for a new ECU, I wonder if they're programming the ECU itself, or rather, programming the immobilizer system and other control boxes in the car to accept the new ECU? I would hope the former (then I could have more than one ECU coded to the same VIN, a stock one and an APR one), but would suspect the latter is more likely.

-Jon

Ape Factory
11-06-2018, 07:28 AM
That's a good question...It'd be easier to flash just one ECU but it is a German car. Still, would be awesome to have one with OEM code and one with a tune! I wonder if APR turned off some of the "protections" when they tuned your ECU. Maybe that's why you're not seeing any codes. I had a tuned car prior to the RS5 and that was the case with it. He was able to turn off certain sensors or parameters that would throw a code and cause a check engine light.

volvofan
11-06-2018, 07:36 AM
That's a good question...It'd be easier to flash just one ECU but it is a German car. Still, would be awesome to have one with OEM code and one with a tune! I wonder if APR turned off some of the "protections" when they tuned your ECU. Maybe that's why you're not seeing any codes. I had a tuned car prior to the RS5 and that was the case with it. He was able to turn off certain sensors or parameters that would throw a code and cause a check engine light.

They definitely code some stuff out. The supercharged version eliminates the intake manifold flaps and that whole system, for one. Not sure what else might be bypassed, but I suppose the tune could be suppressing some codes that would otherwise be showing up (and that might indicate the source of the problem)?

-Jon

ips_cou
11-06-2018, 07:58 AM
APR won't touch it... I asked.

...

When they "reprogram" a car for a new ECU, I wonder if they're programming the ECU itself, or rather, programming the immobilizer system and other control boxes in the car to accept the new ECU? I would hope the former (then I could have more than one ECU coded to the same VIN, a stock one and an APR one), but would suspect the latter is more likely.

-Jon

Yes I believe the ECU and Kessy/BCM(Body Control Model) is reprogrammed to the new key and the old one will not work if the dealer does the install/programming. Requires calling back to the mother ship(AUDI HQ in Europe) to do the process. Supposedly their are hacks/devices that can get around this, but I believe the cost is more than what Audi will charge. Also, if you do have them do the reprogramming of the key, their is a special code that can only be gotten during the process which you can save and use in the future to reprogram keys(Name is escaping me at the moment). Look around on VCDS/Ross-tech site for the name of this/procedure.

My understanding is, the ECU has a area of memory dedicated to the immobilizer, which if you want to keep your keys, would need to be copied, and can only be done by opening the ECU case as you did and doing a so called BDM read(bit like JTAG if thats familiar to you). Basically you need to clone the entire ECU's memory which is not accessible through OBD2. I believe a similar process is required for 'immo off', which would require no specific key to start the car as the encryption/key unique-ness is ignored. You may want to persue the 'immo off' option on the new/used ECU at least initial to save you the dealer cost of programming new dealer keys.

Nice explanation of BDM:
src:https://www.element14.com/community/thread/19451/l/jtag-vs-bdm?displayFullThread=true


As usual, you can find out more about both topics at Wikipedia: http://en.wikipedia.org/wiki/Background_Debug_Mode_interface and http://en.wikipedia.org/wiki/Joint_Test_Action_Group.



Here's my understanding: BDM is a Freescale (formerly Motorola) technology implemented in some of their chips to allow an external debugger to control the chip over a small number of serial lines. BDM provides functions like memory read and write, setting breakpoints, single step, and start/halt execution.



JTAG is a standard 4- or 5-wire serial interface used by many manufacturers for many purposes. It was originally designed for testing chips and PC boards at the IC pin level to diagnose bad chips and bad connections between chips. Basically, all the I/O pins for all the JTAG-capable chips are connected into a shift register and you can shift test patterns in, single step, and shift the results out. It's a serial replacement for "bed of nails" testers, which became impractical with high-density surface-mount components.



The JTAG standard is flexible enough that it can also be used to access internal debugging capabilities (if the chip has them) which are functionally equivalent to BDM. I believe Freescale has pins that can be used either as a BDM interface or JTAG with BDM functionality.



So you can think of JTAG as a superset of BDM, since JTAG can also be used for testing I/O pins. BDM is probably standard across all the Freescale chips that use it. The BDM-like capabilities for other chips vary among processor architecture, and documentation for a particular architecture may or may not be available. ARM has a nice set of BDM-like debug capabilities which are documented well.

volvofan
11-07-2018, 06:32 AM
Yes I believe the ECU and Kessy/BCM(Body Control Model) is reprogrammed to the new key and the old one will not work if the dealer does the install/programming. Requires calling back to the mother ship(AUDI HQ in Europe) to do the process. Supposedly their are hacks/devices that can get around this, but I believe the cost is more than what Audi will charge. Also, if you do have them do the reprogramming of the key, their is a special code that can only be gotten during the process which you can save and use in the future to reprogram keys(Name is escaping me at the moment). Look around on VCDS/Ross-tech site for the name of this/procedure.

My understanding is, the ECU has a area of memory dedicated to the immobilizer, which if you want to keep your keys, would need to be copied, and can only be done by opening the ECU case as you did and doing a so called BDM read(bit like JTAG if thats familiar to you). Basically you need to clone the entire ECU's memory which is not accessible through OBD2. I believe a similar process is required for 'immo off', which would require no specific key to start the car as the encryption/key unique-ness is ignored. You may want to persue the 'immo off' option on the new/used ECU at least initial to save you the dealer cost of programming new dealer keys.

Nice explanation of BDM:
src:https://www.element14.com/community/thread/19451/l/jtag-vs-bdm?displayFullThread=true

Wow, that's a lot of great info!

I wonder if it's possible, by looking at the photos of our ECU, to determine where the programmable portions of the control software are stored (i.e. the parts that would vary... immobilizer codes, adaptations, VIN, custom tunes, etc)? If everything else on the PCB is "standard" from vehicle to vehicle (all the ignition / fuel injector drivers, power circuits, etc) then IF the ECU is bad, the odds are that it'll be one of those components and not a suddenly-corrupted memory. If I could just swap over that memory chip (or chips) to a new ECU, then I wouldn't need to get reprogrammed, ya?

-Jon

ips_cou
11-07-2018, 09:16 AM
Wow, that's a lot of great info!

I wonder if it's possible, by looking at the photos of our ECU, to determine where the programmable portions of the control software are stored (i.e. the parts that would vary... immobilizer codes, adaptations, VIN, custom tunes, etc)? If everything else on the PCB is "standard" from vehicle to vehicle (all the ignition / fuel injector drivers, power circuits, etc) then IF the ECU is bad, the odds are that it'll be one of those components and not a suddenly-corrupted memory. If I could just swap over that memory chip (or chips) to a new ECU, then I wouldn't need to get reprogrammed, ya?

-Jon

Unfortunately my previous post is most of my knowledge in that area. The first ECU PCB pic., the bottom edge right center, those 14(7 by 7 parallel to the edge, contact points) are where the BDM100 tool interfaces. Two chips above that (Part# MPC564MZS6, which looks to be a BGA style chip) that is the main processor. Also I haven't seen it mentioned in this thread, but this ECU is called MED9.1.1, Most info for MED9.1 will apply and you are most likely to find info referring to the MED9.1 ECU(used in the turbo 2.0/ TFSI motors).

If you are not familiar with nefariousmotorsports.com forum you may want to search around there, be forewarned, they are not fans of newbies. Do plenty of searching before asking a question, if not, you will be lucky to get any reply. It's a bit like may other Audi outfits, their shit dont stink and you are not worthy(would really love to ring many of the peoples necks, it kills the progress and community of the whole hobby, I'm talking many places too, not just that site).

If you can get somewhere while keeping some sanity and sleep, I would suggest doing 'immo off' on a new/used ECU using what ever data(tune/OS) that is already on the new ECU. So steps should be, 1.)BDM100 read new ECU, 2.)find info/someone without a stick up their ass to 'immo off' the binary file read with BDM100, 3.)write binary back onto new ECU. 4.) Running car! ...hopefully[;)] I've not done this myself so I may be missing something.

Hope that helps. If it does, It would be awesome if you keep us updated.

volvofan
11-07-2018, 09:41 AM
Unfortunately my previous post is most of my knowledge in that area. The first ECU PCB pic., the bottom edge right center, those 14(7 by 7 parallel to the edge, contact points) are where the BDM100 tool interfaces. Two chips above that (Part# MPC564MZS6, which looks to be a BGA style chip) that is the main processor. Also I haven't seen it mentioned in this thread, but this ECU is called MED9.1.1, Most info for MED9.1 will apply and you are most likely to find info referring to the MED9.1 ECU(used in the turbo 2.0/ TFSI motors).

If you are not familiar with nefariousmotorsports.com forum you may want to search around there, be forewarned, they are not fans of newbies. Do plenty of searching before asking a question, if not, you will be lucky to get any reply. It's a bit like may other Audi outfits, their shit dont stink and you are not worthy(would really love to ring many of the peoples necks, it kills the progress and community of the whole hobby, I'm talking many places too, not just that site).

If you can get somewhere while keeping some sanity and sleep, I would suggest doing 'immo off' on a new/used ECU using what ever data(tune/OS) that is already on the new ECU. So steps should be, 1.)BDM100 read new ECU, 2.)find info/someone without a stick up their ass to 'immo off' the binary file read with BDM100, 3.)write binary back onto new ECU. 4.) Running car! ...hopefully[;)] I've not done this myself so I may be missing something.

Hope that helps. If it does, It would be awesome if you keep us updated.

Thanks for all the help! I did already find that forum and did already post a newb question (as someone had a thread going re: MED 9.1.1 from an S5) so I probably won't get a reply. However, I also did research some more about BDM/JTAG and independently concluded that the steps 1-4 you describe above are probably the most cost effective solution at this point (IF AND ONLY IF MY ECU DATA IS, IN FACT, NOT CURRENTLY CORRUPTED). I also searched the main processor's part number (as you did) and it seems that it has about 512k of flash memory on board. That likely rules out the simpler "unsolder this chip over here, move it to that board, and you're good" option unless the immobilizer data is stored elsewhere on the PCB. Possibly that little 8-pin IC in the upper right corner of the board in the first pic (same side as the processor)?

-Jon

ips_cou
11-07-2018, 10:00 AM
Thanks for all the help! I did already find that forum and did already post a newb question (as someone had a thread going re: MED 9.1.1 from an S5) so I probably won't get a reply. However, I also did research some more about BDM/JTAG and independently concluded that the steps 1-4 you describe above are probably the most cost effective solution at this point (IF AND ONLY IF MY ECU DATA IS, IN FACT, NOT CURRENTLY CORRUPTED). I also searched the main processor's part number (as you did) and it seems that it has about 512k of flash memory on board. That likely rules out the simpler "unsolder this chip over here, move it to that board, and you're good" option unless the immobilizer data is stored elsewhere on the PCB. Possibly that little 8-pin IC in the upper right corner of the board in the first pic (same side as the processor)?

-Jon

If I read that correctly, you want to pull the tune/entire memory from you current ECU and flash it to a new/used ECU? If so, I would recommend, if you can get it working anyway, just doing 'immo off' on the new ECU and see if the misfire still happens, especially being it happens at low load, your not going to hurt anything. Just keep the RPMs and load down you will be fine. And for extra insurance you can pour some Torco octane booster(awesome stuff, just need a $20 bottle)

Not sure I saw it mentioned, had a tickle in the brain, think I've heard a bad motor mount can cause this as they are active. Should be easy to check, I think all you need is a resistor hooked to the hardness to simulate a good mount.

One problem with desoldering IC's is the possibility of damage during.

No idea on the 8 pin IC, never got that far into it.

volvofan
11-07-2018, 11:51 AM
If I read that correctly, you want to pull the tune/entire memory from you current ECU and flash it to a new/used ECU? If so, I would recommend, if you can get it working anyway, just doing 'immo off' on the new ECU and see if the misfire still happens, especially being it happens at low load, your not going to hurt anything. Just keep the RPMs and load down you will be fine. And for extra insurance you can pour some Torco octane booster(awesome stuff, just need a $20 bottle)

Not sure I saw it mentioned, had a tickle in the brain, think I've heard a bad motor mount can cause this as they are active. Should be easy to check, I think all you need is a resistor hooked to the hardness to simulate a good mount.

One problem with desoldering IC's is the possibility of damage during.

No idea on the 8 pin IC, never got that far into it.

Good idea on removing the immobilization from the ECU to see if it fixes the problem. Cheaper than dealer reprogramming, and quicker than trying to learn how to do JTAG voodoo myself a la steps 1-4 as we hit on in posts 53 & 54.

If that does the trick, I can put the bad ECU back in, drive to the APR dealer, and have them flash the "Immo off" ECU.

I also considered the motor mount, but I believe that a failed motor mount (one that's bad enough to cause misfiring) generates a code?

-Jon

ips_cou
11-07-2018, 01:42 PM
Good idea on removing the immobilization from the ECU to see if it fixes the problem. Cheaper than dealer reprogramming, and quicker than trying to learn how to do JTAG voodoo myself a la steps 1-4 as we hit on in posts 53 & 54.

If that does the trick, I can put the bad ECU back in, drive to the APR dealer, and have them flash the "Immo off" ECU.


I believe Immo off still requires the BDM read->edit->write process. Just using the binary already on the new ECU.






I also considered the motor mount, but I believe that a failed motor mount (one that's bad enough to cause misfiring) generates a code?

-Jon

I think that was the rub, it wouldn't throw a code, leading to confusion.

IIRC, the connectors at the mounts can be gotten to relatively easy.

I think I may know where my old ones are, Ill try looking tonight, I'll see what they ohm out at.

FYI, I may be off on the mount thing being a electrical issue when they go bad. I never had the issue, i just vaguely remember others talking about it. Just trying to throw anything in simple that may have been missed.

Edit:
quick search for N144(driver side mount) says 19 or 22 ohm resistor fools the ECU.

https://www.audizine.com/forum/showthread.php/590025-Motor-mount-questions-homemade-DIY-B7-resistor-soft-codes

Edit2:
another link that may help(immo off)
http://nefariousmotorsports.com/forum/index.php?topic=5418.0

Ape Factory
11-07-2018, 09:19 PM
Man if this turns out to be a motor mount I may just start writing letters. I never write letters (ok once, I wrote a letter once) but that would put me over the edge.

As far as desoldering chips, you'd need special gear for that. Would not be technically easy let alone the likelihood of frying chips with too much heat. Best left to the experts. I do quite a bit of soldering on far less complicated circuits and there's no way in hell I'd attempt to desolder a multi-prong chip from a board like that unless I had the proper equipment. Would probably cost you more in equipment than replacing with a brand new ECU.

Developing carpel tunnel from writing many, many angry letters in broke German aside, I hope it's the motor mount. Easy enough to replace and get the car running again.

volvofan
11-08-2018, 05:34 AM
Man if this turns out to be a motor mount I may just start writing letters. I never write letters (ok once, I wrote a letter once) but that would put me over the edge.

As far as desoldering chips, you'd need special gear for that. Would not be technically easy let alone the likelihood of frying chips with too much heat. Best left to the experts. I do quite a bit of soldering on far less complicated circuits and there's no way in hell I'd attempt to desolder a multi-prong chip from a board like that unless I had the proper equipment. Would probably cost you more in equipment than replacing with a brand new ECU.

Developing carpel tunnel from writing many, many angry letters in broke German aside, I hope it's the motor mount. Easy enough to replace and get the car running again.

I have the special gear (re-work station) but am disinclined to use it unless I have a high degree of confidence in the fact that it'll fix my problem. Like you said, too easy to screw up.

I'll take another look at the motor mount... I wonder what triggers the code? Is it just something like a brake pad wear sensor where as soon as it makes contact from wearing to a certain point, the "idiot light" comes on? Or, is it a transducer-type setup where it's actually measuring movement to compare actual to what is desired? Hmmm.....

volvofan
11-08-2018, 05:40 AM
I believe Immo off still requires the BDM read->edit->write process. Just using the binary already on the new ECU.





I think that was the rub, it wouldn't throw a code, leading to confusion.

IIRC, the connectors at the mounts can be gotten to relatively easy.

I think I may know where my old ones are, Ill try looking tonight, I'll see what they ohm out at.

FYI, I may be off on the mount thing being a electrical issue when they go bad. I never had the issue, i just vaguely remember others talking about it. Just trying to throw anything in simple that may have been missed.

Edit:
quick search for N144(driver side mount) says 19 or 22 ohm resistor fools the ECU.

https://www.audizine.com/forum/showthread.php/590025-Motor-mount-questions-homemade-DIY-B7-resistor-soft-codes

Edit2:
another link that may help(immo off)
http://nefariousmotorsports.com/forum/index.php?topic=5418.0

Boom. Will look into the motor mount and try to rule that out before I finish going down the ECU road... First thing I need to figure out: I don't doubt that "misfire from 'bad motor mount'" is a possible issue, what I am trying to sort is-

Is the motor mount actually physically BROKEN even though the ECU doesn't detect that and indicate as such, and somehow the resulting vibrations in the motor are causing the misfires? In other words, it's a MECHANICAL problem and I must replace the mount? OR, is the ECU relying on some sort of signal from the motor mount that it's no longer getting (even though the motor mount may be mechanically just fine), and it's not getting that signal so it screws up ignition or fuel or something? In other words, an ELECTRICAL problem and I must fool the computer?


-Jon

ips_cou
11-08-2018, 08:27 AM
Boom. Will look into the motor mount and try to rule that out before I finish going down the ECU road... First thing I need to figure out: I don't doubt that "misfire from 'bad motor mount'" is a possible issue, what I am trying to sort is-

Is the motor mount actually physically BROKEN even though the ECU doesn't detect that and indicate as such, and somehow the resulting vibrations in the motor are causing the misfires? In other words, it's a MECHANICAL problem and I must replace the mount? OR, is the ECU relying on some sort of signal from the motor mount that it's no longer getting (even though the motor mount may be mechanically just fine), and it's not getting that signal so it screws up ignition or fuel or something? In other words, an ELECTRICAL problem and I must fool the computer?


-Jon


Apparently it's a relatively dumb system, inside the mount is a solenoid valve which is normally open below 1100 rpm, it closes above 1100 rpm, and thats it. It has a fluid and air chamber, separated by a membrane, the solenoid controls air flow in and out of the air side. In effect, the mount is soft below 1100 rpm, hard above that.(see link in above post for more detail if interested) An yes, the possible mechanical(over deflection..metal to metal contact) ramifications is not something to ignore.

I ohm'ed my old mounts, which I think leaked their fluid, but had intact rubber. 29 Ohm each(my car is a 2009 btw). I didn't have any issues I knew of, was just doing other work and it was the best time being they basically require the engine removed...one step away anyway. FYI, I went with the 034 Motorsports mount and have worked good so far. Not a lot of miles yet, but quite a few launches.

... coming back to me, I think the passenger side mount had at least partially collapsed and may have been sitting on the hard stop, also I think the excessive movement led to damaging the power steering pump feed line(was leaking right where the hose connects to the pump, uses a hose clamp). Also the fact that it's inches from the header and in a enclosed environment(oven like). I replaced that hose, covered it in high temp shielding(1000F+ rated) and used a constant tension clamp.. no leaks yet!

EDIT:
If it's electrical and the ECU is not reporting on it properly or at all(may be intermittent(as in milli seconds or less of faulting and may fall outside a hysteresis), not enough to trigger a fault code or even improper algorithm logic or even electrical noise in the circuit(intermittent connection could cause some nasty voltage spikes)), easiest way to null all that out would be just first, disconnecting the mount solenoid connector, see what happens. Next, and/or putting a ~20 Ohm resistor on each one.

volvofan
11-08-2018, 09:04 AM
Apparently it's a relatively dumb system, inside the mount is a solenoid valve which is normally open below 1100 rpm, it closes above 1100 rpm, and thats it. It has a fluid and air chamber, separated by a membrane, the solenoid controls air flow in and out of the air side. In effect, the mount is soft below 1100 rpm, hard above that.(see link in above post for more detail if interested) An yes, the possible mechanical(over deflection..metal to metal contact) ramifications is not something to ignore.

I ohm'ed my old mounts, which I think leaked their fluid, but had intact rubber. 29 Ohm each(my car is a 2009 btw). I didn't have any issues I knew of, was just doing other work and it was the best time being they basically require the engine removed...one step away anyway. FYI, I went with the 034 Motorsports mount and have worked good so far. Not a lot of miles yet, but quite a few launches.

... coming back to me, I think the passenger side mount had at least partially collapsed and may have been sitting on the hard stop, also I think the excessive movement led to damaging the power steering pump feed line(was leaking right where the hose connects to the pump, uses a hose clamp). Also the fact that it's inches from the header and in a enclosed environment(oven like). I replaced that hose, covered it in high temp shielding(1000F+ rated) and used a constant tension clamp.. no leaks yet!

EDIT:
If it's electrical and the ECU is not reporting on it properly or at all(may be intermittent(as in milli seconds or less of faulting and may fall outside a hysteresis), not enough to trigger a fault code or even improper algorithm logic or even electrical noise in the circuit(intermittent connection could cause some nasty voltage spikes)), easiest way to null all that out would be just first, disconnecting the mount solenoid connector, see what happens. Next, and/or putting a ~20 Ohm resistor on each one.

Any interest in selling those old mounts? [wrench]

This is the most telling question/reply in the whole thread (assuming our mounts still work this same way):

(Question):
So is there any disadvantage of not having a proper sensor for the motor mount because you need a resistor to trick the ECU?
Does the sensor check for a damaged/worn motor mount or a loose or uninstalled mount?

(Reply):
The stern mounts feel very much like stock around idle with only a slight increase in vibration, most people will not even notice it.
The sensor actuates the mount just so you do not feel any vibrations at idle. It does not sense when a mount is broken or loose.

ips_cou
11-08-2018, 09:59 AM
Any interest in selling those old mounts? [wrench]

This is the most telling question/reply in the whole thread (assuming our mounts still work this same way):

(Question):
So is there any disadvantage of not having a proper sensor for the motor mount because you need a resistor to trick the ECU?
Does the sensor check for a damaged/worn motor mount or a loose or uninstalled mount?

(Reply):
The stern mounts feel very much like stock around idle with only a slight increase in vibration, most people will not even notice it.
The sensor actuates the mount just so you do not feel any vibrations at idle. It does not sense when a mount is broken or loose.

Sorry saving them for potential mock up of a more compact mount..potential area for some tubing...[evilsmile]

Not sure of the exact question/statement. One point though, there is no sensor. Only a solenoid(electro magnet, actuating a valve). Which is closed(energized) or open, depending on the RPM the ECU is measuring. RPM less than 1100RPM = not energized = open valve. RPM greater than 1100 = energized = valve closed.

Disclaimer I'm not a Electrical Engineer(Just know enough to be dangerous[:d]) Most likely, the ECU only expects current when applying power to the solenoid, if it does not see the current, fault code is thrown(amusing a perfect world). (They may also run a small amount of current when the solenoid is suppose to be open to detect if the sensor is unplugged/coil in solenoid failed(could end up with floating voltage, a small variance)(and problem with variances are they are hard to measure and susceptible to noise), or a short to ground, both of these are the most likely intended designs that come to mind for this application). I don't see how it would be much more complicated than that as that is the only hardware inplace.

Now if the mount is a issue and is not faulting, that is where bugs come in. Very easy to shoot yourself in the foot and not know it in EE and Programming. And if my anecdotal experience is remotely accurate, and what others report is as well, it's safe to say any number of the already mentioned possibilities could be manifesting as Audi's QA is clearly not perfect, if not far from it.

(Assuming a stock mount driven over a long period)So by putting the resistors in place, best case you feel the motor vibrations slightly more, if at all. Worst case, the mounts go out quicker due to increased range of motion due to the valve being open all the time and the ECU detects false ignition knock due to the mounts over extending and having metal to metal contact.

I would bet though, your mounts are due for replacement anyway if they are stock, especially with the extra power/ low end torque and 90k+ Mine were toast well before that.

That last bit '..The sensor actuates..' is either someone not knowing what they are talking about/and or market fluff for joe schmo.

volvofan
11-08-2018, 07:47 PM
OK, so I was unclear regarding what I was referring to when I said "This is the most telling question/reply in the whole thread". I was talking about the thread you linked to with info about motor mounts and misfires (https://www.audizine.com/forum/showt...tor-soft-codes), not this current thread. My point being, the mount itself failing (rubber tearing, fluid coming out) will NOT necessarily be detected by the ECU and flagged with a code. Second point - simply installing a resistor to "spoof" the ECU into thinking the motor mount solenoid is A-OK probably isn't going to help me out, because the ECU is already NOT getting a code indicating a solenoid failure.

Confirmation bias at work once again, but the more I think about it, the more this is a plausible reason for the fault. I stated that the car seemed to go bad "overnight", but the reality is, it became progressively more undrivable over a very short period of time (a day or two). If the motor mount finally tore at the end of the day before the problem manifested, then it would've had the night to have some fluid leak out. The next morning, it would've started running rough, but might have still had some fluid in the mount, so I would've started seeing misfires and drivability issues. By a day or two later, with all the fluid squished/leaked out, the problem got to its worst (and current) state. The solenoids are probably still just fine, clicking away to stop the flow of fluid that no longer exists inside the mount. Hence, no code. This also explains why it runs relatively well at idle, and under very light load/throttle (ie low RPMs). It's not until the engine really starts revving that the mount starts bottoming out, as there is now only an air pocket where a fluid-filled cavity was before. SOMEHOW (and this is the only place I have reservations), this bottoming out is causing the misfires. I'm not seeing any knock counts or timing retard show up, which is what I would have figured on seeing if a motor mount was clanking away.

Given the age/mileage on the car and the availability of an upgraded part (the 034 Motorsport street density mounts), I can easily justify the installation of replacements even if this does not ultimately wind up being the cause of the issue. Just placed my order and will report back once installed.

-Jon

ips_cou
11-09-2018, 09:00 AM
OK, so I was unclear regarding what I was referring to when I said "This is the most telling question/reply in the whole thread". I was talking about the thread you linked to with info about motor mounts and misfires (https://www.audizine.com/forum/showt...tor-soft-codes), not this current thread. My point being, the mount itself failing (rubber tearing, fluid coming out) will NOT necessarily be detected by the ECU and flagged with a code. Second point - simply installing a resistor to "spoof" the ECU into thinking the motor mount solenoid is A-OK probably isn't going to help me out, because the ECU is already NOT getting a code indicating a solenoid failure.

Confirmation bias at work once again, but the more I think about it, the more this is a plausible reason for the fault. I stated that the car seemed to go bad "overnight", but the reality is, it became progressively more undrivable over a very short period of time (a day or two). If the motor mount finally tore at the end of the day before the problem manifested, then it would've had the night to have some fluid leak out. The next morning, it would've started running rough, but might have still had some fluid in the mount, so I would've started seeing misfires and drivability issues. By a day or two later, with all the fluid squished/leaked out, the problem got to its worst (and current) state. The solenoids are probably still just fine, clicking away to stop the flow of fluid that no longer exists inside the mount. Hence, no code. This also explains why it runs relatively well at idle, and under very light load/throttle (ie low RPMs). It's not until the engine really starts revving that the mount starts bottoming out, as there is now only an air pocket where a fluid-filled cavity was before. SOMEHOW (and this is the only place I have reservations), this bottoming out is causing the misfires. I'm not seeing any knock counts or timing retard show up, which is what I would have figured on seeing if a motor mount was clanking away.

Given the age/mileage on the car and the availability of an upgraded part (the 034 Motorsport street density mounts), I can easily justify the installation of replacements even if this does not ultimately wind up being the cause of the issue. Just placed my order and will report back once installed.

-Jon

I agree will all above but one part. The imperfections of the vehicle in..well may aspects.. leave me suspicious as to the ECU code/hardware having a slight issue with properly reporting its believed state of the motor mounts(I've encountered many of these types of problems with this car now). I would be inclined to try out the resistor spoof in the mean time just to see if it has any effect.

But I hope thats all it is(annoying as that could be). I'll keep a eye out for a update, good luck!

volvofan
11-12-2018, 06:48 AM
I agree will all above but one part. The imperfections of the vehicle in..well may aspects.. leave me suspicious as to the ECU code/hardware having a slight issue with properly reporting its believed state of the motor mounts(I've encountered many of these types of problems with this car now). I would be inclined to try out the resistor spoof in the mean time just to see if it has any effect.

But I hope thats all it is(annoying as that could be). I'll keep a eye out for a update, good luck!

Ok I see what youíre saying... the motor mount solenoid COULD still be malfunctioning and the ECU might not properly be detecting/reporting it.

I guess since I have committed to the ECS solid mounts, Iíll get a good look at the old ones in ďpost-mortemĒ.

volvofan
11-17-2018, 02:37 PM
Okay... WOW.

The stock mounts were completely shot on both sides. This shouldnít have been a surprise to me, given that they were original (AFAIK) and that this car is putting out a couple more HP than normal. Passenger side (same side as the misfires) was the worst, cracked all the way through both sides of the rubber. Driverís side was also cracked through one side but not leaking fluid yet. I got both replaced and started buttoning things up, but I ran out of time at the shop and didnít get it all put back together to test for sure that it solved the misfires. Given the state of the mounts, though, Iím feeling increasingly confident that this will fix the problem. Even if it doesnít, itís obviously something that needed to be done regardless.

More to follow next week when I get it all back together.

-Jon

https://i.imgur.com/U3w2wWJ.jpg

https://i.imgur.com/a4qz2Hx.jpg

volvofan
11-21-2018, 06:57 AM
Definitely feels better with the motor mounts in there, but unfortunately it's still misfiring, just like before. Bank one only. Time to bust out the scope, I guess... :(

Happy Thanksgiving, all!

-Jon

Ape Factory
11-21-2018, 06:28 PM
Definitely feels better with the motor mounts in there, but unfortunately it's still misfiring, just like before. Bank one only. Time to bust out the scope, I guess... :(

Happy Thanksgiving, all!

-Jon

I was seriously hoping that would solve it. Sorry to hear!

Happy Thanksgiving to you as well.

volvofan
12-05-2018, 01:35 PM
I believe we have a diagnosis!

Before I resorted to scoping the ECUís inputs and outputs, I decided to head over to the Ross-Tech forums to see if they had any ideas. Some good guidance on things to test via VCDS and some neat tricks I wasnít aware of. Very helpful and very responsive. If I had any lingering doubts about springing for the ďrealĒ VCDS hardware, I have no doubt now that it was worth it... that thread where I got all the great help is for registered users only.

Anyhow, in chatting back and forth about what Iíve tested and observed, I decided to check ďdownstreamĒ before probing further into the ECU and wiring harness. Today, I pulled the front O2 sensor for bank 1 and stuck my borescope in there... it was a horror show. Disintegrated chunks of catalytic converter were lying loose in the housing. What honeycomb remained intact looked like someone had spackled the holes shut. The cell that sits upstream of the rear O2 sensor was almost completely gone... I could see the rear sensor from the front sensorís hole... that ainít good.

I think one of two things happened:
1) The cat failed (as they are known to do at a certain age, especially on a forced induction vehicle) more or less suddenly, and the motor mounts were already shot (or nearly so) and unrelated to the misfires.
2) A motor mount failed, which caused the misfires. The misfires caused the cat to fail and clog, so even when I replaced the motor mounts, the symptom remained.

In the end, it doesnít matter which is the case. For now, my focus is on replacing (or eliminating) the cat to ensure that addresses the misfire issue, then Iíll go from there.

Thanks to everyone on the thread for all the help along the way!

-Jon

volvofan
12-05-2018, 01:40 PM
Some pics of the cat:

https://i.imgur.com/1quLmM6.jpg

https://i.imgur.com/DHGa0Ke.jpg

https://i.imgur.com/bQMS5TA.jpg

https://i.imgur.com/WZRphll.jpg

https://i.imgur.com/yrEMdxf.jpg

Ape Factory
12-05-2018, 01:40 PM
Yes! Great information for future troubleshooting too. And I can vouch for the usefulness of the VCDS. Very powerful tool. Really glad you got it figured out.

volvofan
12-05-2018, 01:53 PM
Yes! Great information for future troubleshooting too. And I can vouch for the usefulness of the VCDS. Very powerful tool. Really glad you got it figured out.

Yup! The only thing that brings some satisfaction and closure to a problem like this is writing it up to help others avoid the same pitfalls (you know, like $800 worth of injectors I probably didnít need).

-Jon

Ape Factory
12-05-2018, 01:55 PM
Look at the bright side, It could be worse. You could have an RS5 with two more injectors and one-piece headers/cats that necessitate engine removal to change out.

volvofan
12-05-2018, 02:16 PM
Look at the bright side, It could be worse. You could have an RS5 with two more injectors and one-piece headers/cats that necessitate engine removal to change out.

Yeah, that would suck!

What are the other two injectors for, and where do they go???

-Jon

Ape Factory
12-05-2018, 07:47 PM
Oh wait...you have the 4.2L. I keep thinking 3.0T due to the supercharger, LOL. Not enough sleep this week.

volvofan
01-16-2019, 07:46 PM
Well, thatís running a little better :)

https://www.audizine.com/forum/showthread.php/841622-Point-me-towards-a-cat-removal-how-to?p=13484948#post13484948

Driverís side cat was on the way out, too... not clogged yet, but partially blown. Already hollowed out between the front and rear O2 sensors.