It was a beautiful spring day at Opus IVS. The sun was shining, the birds were singing, nothing could go wrong. Or so I thought…
A client with a 2008 Audi A8 called in for support. The vehicle ran horribly and the client had questions about timing codes. Among the myriad of codes in the vehicle was one that stuck out: P0014 “Bank 1 exhaust camshaft retard, SPECIFICATION NOT REACHED (Figure 1).”
Before I get into the diagnostics of this code, let me tell you a bit about this engine. In short, it’s not fun. This is a V-10 that Audi designed to put into the Lamborghini Gallardo. In 2006, Audi decided to install this engine in the A8 – in a space barely large enough for a V-6. As this engine was designed for an exotic sports car, there was little knowledge of these vehicles operating at higher mileage. Known-good data is few and far between. In my experience, these engines were typically replaced before any timing diagnostics had to be performed. With that being said, let’s get to the code diagnostics.
How is the failure determined? According to the service information, this code sets when the computer detects a difference of more than 10 degrees of crankshaft variation from that which is specified (Figure 2). This is monitored continuously while the engine is idling for more than 3.5 seconds. The MIL will illuminate with the second key cycle.
The client provided a little back story on this vehicle. This was not a regular customer for this shop. This vehicle has been around. Its latest trip was from the dealership, where they advised the customer to replace the timing chain. On top of this, this vehicle was a newer purchase for the owner and has never run right. The owner naturally wanted a second opinion.
Enter our client. He has a reputation for being the “go-to guy” for European vehicle diagnostics and repair. He is a pleasure to speak to and is always willing to go the extra mile to get the vehicle properly diagnosed. He made the first call in for support and my co-worker took the ticket. After a discussion, they decided the best path to start with (due to the vehicle’s mileage) should be a mechanical check of the timing.
When he called in again, he asked for clarification of instruction on locking the cam phasers after a timing chain replacement. He was confused as to why there was no locating pin on the cam adjuster. I explained to him that the cams get locked with the timing tools and the phasers get locked down after the pins are pulled on the tensioners. I asked him how far the chain was off. He told me that the timing tools slipped right in (that sounds to me like the cams are indeed in time), but he wanted to install chains anyway because of the mileage and age of all the components. For those of you who are unaware, this timing chain setup is in the back of the engine. Chain replacement involves removing the engine and taking the rear timing cover off and replacing the four (yes, four!) timing chains. This is not a small job!! When he called back again, he stated the code was still there (not surprisingly). This is when I took over and put my diag hat on. This was going to be a fun one.
A shot at redemption
One of the most overlooked items on these engine types is the camshaft adjuster (or phaser). The adjuster has oil passages that fill with pressurized oil on one side or another. This in turn advances or retards the camshaft. There is no black magic here, it is that simple. One of the biggest issues with these adjusters is the locking pin. This pin locks the adjuster in its fully-retarded position when there is less than 7 psi of oil pressure. Over time, the locking pin or recess where the pin locks can wear out and will cause the adjuster to advance before the vehicle requests it to, and cause a timing-advanced code. This is the first thing I wanted to have him check.
I also wanted to look at the codes that came back to see if there were any clues that would help get us going in the right direction. This is odd. The code number is the same, but the description is completely different (see Figures 1 and 2 for comparison). I checked OEM service info for this to be sure and the definition is for an over-advanced code. I guess technically – if the timing can’t reach the retard point – then by definition, it’s over advanced. Let’s add this to the pile of data we are collecting.
The customer checked the cam adjuster, and it was tight/locked. I also asked him to pull the cover off of the adjuster to see if there was anything stuck inside it that could impede oil flow. He pulled the cover off and it was clean.
As we are dealing with an adjustable timing code, I needed to adapt my test plan. One of the easiest things to do in a situation like this is a little "swaptronics." It is a quick thing to swap left and right bank adjuster solenoids to see if the code moves. It did not.
Looking at a wiring diagram, I can see that the intake adjuster valve, exhaust adjuster valve, and torque valve are all on the same circuit. These components are all capable of setting circuit codes if unplugged or if a fuse is blown. There are no circuit codes, so I know the wiring and fuse are fine. I also had the client swap components (left to right) and the code stayed on the same side. This tells me the problem is something that is on that side.
Employing deductive reasoning
As diagnosticians, our job is to remove as much as we can from the board as efficiently as possible. I know the cam adjuster is capable of moving the cam. I know the solenoid is capable of doing its job. I also know that the circuitry can do its jobs. We are left with only the camshaft, engine computer, and engine oil (hydraulic control).
The next thing I wanted to check was if the computer was telling the adjuster solenoid to work. This customer owns a PicoScope and is willing to use it. I have diagnosed many cars in the past with him using his scope; he expects me to ask him to use the scope at some point during our conversations. I had him hooked to the control wires for the bank 1 and bank 2 exhaust cam adjuster solenoids. I remotely logged into his laptop and watched his scope screen. With the vehicle running, I could see bank 2 being controlled but bank 1 had no activity.
The computer is not even trying to control bank 1. Thinking about it, this makes sense because the computer thinks the engine is out of time. It wants to avoid any valvetrain damage that could be inflicted by advancing the timing any more than maximum spec. The next step is to see what the computer sees. Even though the customer used the correct timing tools on the engine, the engine computer doesn’t see this; it uses the cam sensors and crank sensor to determine engine timing.
I had the customer hook his scope to the crank sensor, bank 1 intake, and bank 1 exhaust. I had him also repeat this for bank 2 (Figure 3). At first glance, this seems off. In my experience, bank 1 and bank 2 should mirror one another. I have known good captures for a V8 – that is in the same engine family as this engine, but I can’t tell my customer to tear this engine down for another timing chain job based on an educated guess. Also, the V-8 engine has 2 adjusting cams, the V-10 has 4. There were just too many differences for me to trust.
I needed to find a known-good pattern. This turned out to be a problem. As I mentioned in the beginning, this engine is a rare bird. I called some techs with dealership experience with these engines (both Lamborghini and Audi), and they all said the same thing; “we only ever put chains on them.” Not much help there. I asked if there were any available cam and crank scope captures; that was another no.
I have a friend who sometimes works on exotics. He did not have the captures but pointed me in the direction of someone who might. I messaged him through Facebook, and not only did he have the same vehicle at his shop and was willing to get me the much-needed information, but he was also a client of ours! Later that evening, he emailed me the known-good cam and crank signals so I could compare with what I had acquired (Figure 4).
Now I can tell we have a problem. One of the pulses from one of the cams should line up nicely with the beginning of the sync notch of the crankshaft. This should also be mirrored on the other side. This is nothing like the capture on the customer’s vehicle. If anything, the code should be triggering on bank 2! Something is wrong here. If the timing was this off on both banks, the vehicle should not run. I needed the customer to recheck his timing with the mechanical tools. I needed to be sure we weren’t missing anything before going forward.
I am now wondering if there was anything wrong with this camshaft or anything else. If the timing tools slide right in and the proper procedures are followed, this engine should purr. I can't trust what any of my testing has told me because the evidence contradicts itself. Being that this engine is special, I can’t have the customer look at a picture of a camshaft to see if the lobes look correct. He can't order one to compare side to side either. Every component on this engine is of special-order policy.
Digging deep for the answers
I needed to see if this engine was properly timed (for sure, this time). The customer and I decided that going in-cylinder with a pressure transducer would tell us if the valve openings were happening at the right time (meaning, the engine is in time). The customer did not have a pressure transducer and there wasn’t a shop locally that he could borrow one from. I decided to overnight him my personal WPS 500 kit from Pico. I needed to get this thing figured out, and I am not above lending out my stuff to make it happen.
He called me when he got the WPS. I had him take out a plug from each bank and install the hose and transducer. I also had him disconnect all the coils so the vehicle would not run. I remotely logged back into his laptop and took control of his scope. He went into bank 2 (the driver side) first and followed up with bank 1 (the passenger side) (Figure 5).
I am not here to nitpick these captures and go over every little nuance. True, I could’ve zoomed in for some better detail, but I just wanted to see if the valves were opening and closing at the right times. Looking at these, there was no appreciable difference side to side, and there were no weird valve timing events. This tells me that the chains had to be in time. I told the client that I needed to think about this and to call me back a little later.
The customer called in the next day. He got to thinking and remembered that this vehicle was at the dealership before it got to him. He decided to do a little looking around and saw that the bank 2 intake and exhaust sensors appeared to be swapped around. He swapped them back and the engine purred like a kitten. I wanted to hook the scope back up to the bank 2 cam sensors and crank sensor to see what difference was made. Apologies for the wobbly trace, I believe the customer had a poor ground connection. But you can see that we were much more in line without known good now (Figure 6). The customer decided to button everything up and drive the vehicle. He called back two days later.
It’s just physics The code came back. The customer mentioned that the plug for the exhaust cam adjuster solenoid was broken, and he was unsure of where the wires went. When he put them one way, he got the timing advanced code. When he swapped them, he got a cam/crank correlation code. I went over the wiring diagram with him, and we found that we had the wires the correct way and the timing advanced code is the true one.
He also mentioned that he smelled coolant up by the cowl area where the ECM lives. He took some covers off and saw that ECM1 (yes, this vehicle also uses two engine computers) was wet with coolant at one time. This may cause an issue. These ECMs are “swappable” if you have ODIS (OE software). This customer has his ODIS, and we swapped the computers, matched the immobilizer, and rechecked. The code did not move with the ECM.
I now know that this vehicle is mechanically timed correctly. I also know that electrically, this vehicle is fine, too. Going through the ODIS that was still hooked up to the vehicle, I saw a test for the cam phasers. Now that I knew the engine was in time, I saw no harm in running it. I saw movement in the data for all the adjusters except for bank 1 exhaust cam. There was no reason that that computer should hold back timing adjustment. I cleared the codes and tried moving the adjuster again. Nothing. I had a customer hook up a noid light to the adjuster solenoid and ran the test again – it lit up!
I know the customer swapped these solenoids bank to bank in the beginning, but we needed to be 1000 percent about everything now. I had him remove the solenoid and bench test it. It worked just fine. So we have control of the solenoid and no movement from the adjuster. How does this adjuster move? By oil pressure.
I then told the customer to leave the solenoid out of the vehicle and have someone stand by the key. If everything is correct, he should have a geyser of oil shooting out of this hole when the vehicle was started. I had him start the vehicle, and there was nothing. All he got out of the oil passage was a gurgle. Of course the cam won't adjust; it was not getting oil pressure!
I then pulled up the Self Study Program for this engine and looked at the oil supply diagram (Figure 7).
The GREEN is pressurized oil coming from the oil pump. The oil makes its way to a passage in the crankcase breather and flows up through the block and cylinder head, into oil galleys for the adjuster solenoids.
If the passage was blocked, restricted, or leaking in any of those spots, there would be a problem on the entire bank. This problem was localized. The customer looked in the empty solenoid hole and found nothing blocking any passages. This made no sense.
It was the end of the day, and I told the customer I would need to think about this. Later that evening, I was catching up with my friend Brandon Steckler. As mechanics do, we eventually got to talking about cars. I started telling him about this vehicle. I even had him log into my computer and I went over all the data I had gathered.
I got to the point with the oil flow diagram and was explaining how it makes no sense that there can be no oil available to only one part of an internally shared system. Then, in mid-sentence, I stopped. I told him that I figured it out. The diagram shows the correct path of oil flow. What it does not show are all the parts that the oil flows through.
The pressurized oil comes up from the crankcase breather and travels through oil passages in the cylinder head. It then makes a 90-degree turn and heads to the timing chain tensioner (not pictured), through the oil galleys in the tensioner, and another 90-degree turn back into the cylinder head. From there the oil then makes its way back to the exhaust cam adjuster solenoid. There is pressurized oil to the intake cam adjuster. This means the only place we are losing oil pressure is in the timing chain tensioner.
Remember, this vehicle has been through a bunch of hands. My customer has never seen it run correctly. If someone had this chain tensioner apart and botched it up, or the gasket shifted, this could cause the problem.
I emailed my findings to the customer that evening and called the customer as soon as I got in the next day. I explained how this must be the issue and we could finally get this vehicle back on the road.
He was happy that we had an answer, but he then informed me that the customer took the vehicle back because he was not happy with how long this process was taking. We both felt defeated in the fact that we were not going to see a happy ending.
But, there is some good news. I did not have the customer replace any parts that were not needed. We diagnosed an issue for which service information was sorely lacking. We also used what we did know to figure out what we didn’t know. Not all stories have a happy fairytale ending. Sometimes it’s more about the journey than it is the destination. I am still calling this a win. I know what it will take to fix this vehicle, and sometimes in diagnostic support, that is good enough.