Admittedly, the diagnosis part of my job usually doesn’t take more than an hour for any given car or problem. But every once in a while a car comes in with a problem that defies logic and baffles everyone in the shop. This is the story of one of those cars.
The customer purchased a non-running C32 AMG for about 30% of what they sell for in good running condition. Previous owner had already tried getting it to run but was obviously unsuccessful. Previous visit to a dealership was three months prior with complaint of an “engine noise” and the mileage was basically the same since.
For those who don’t know or don’t care to look it up, the C32 has an 18 valve, SOHC, supercharged V6 with two spark plugs per cylinder.
Engine cranks over beautifully, sounds totally normal, except for when it occasionally seems to backfire slightly, which tended to happen after sitting for a few hours but not after 10-15 seconds of cranking. But it just won’t fire and run. Now is the part where you go through and check all the basics; spark, fuel pressure and volume, injector pulse. Did that, all tested fine, even checked spark and injector pulse on every cylinder instead of just one like people usually do. Checked all the fluids for level and condition, fine. Checked the oil cap for the foamy white stuff you get from bad had gaskets and such, totally clean. Hmm, that’s weird, why won’t it start?
Next step is to check compression, so I pulled one plug from every cylinder and found them all to look normal but we’re soaking wet with a strong fuel odor. Makes sense considering it won’t run. Checked compression and found it to be 9-9.5 bar, totally fine for a stone cold engine with fuel washed cylinders. While the plugs were out I decided to have a look at the cylinders with the ol’ trusty colonoscopy cam, or bore-scope as it’s more commonly referred. Found the cylinders to be damp but no signs of damage and no standing fluid in the pistons.
Ok, clean and dry all the spark plugs, disconnect the injectors and crank the engine over a few times to dry out the cylinders just in case it is too wet to fire. Slapped it back together quick and cranked it over. And BAM! It fired and ran for all of 3-4 seconds before stalling! Must have been too wet, but why won’t it stay running and how did it get so wet?
Next I tried running it on an external fuel source introduced straight to the fuel rail to rule out fuel quality, pressure and volume even further, no change. Then I used a very fancy tool (read: screwdriver stuck in the plug wires and held varying distances from metal surfaces) to check the strength of the spark/coils, and found the spark to be very strong and easily jump 1" gaps.
Hmm, maybe we’ve got poor spray patterns from the injectors, but all at once would be really weird. Another possibility is incorrect cam timing. So I pull the right valve cover off, set the crank where it needs to be and try to install the locating jig on the cam, no go, it won’t fit because timing is off! I’ve seen stretched chains on these engines before but never cause a no start and they’re really rare. So then I turned the crank until the locating jig would drop into place, then checked how fast the crank had rotated, 100°! The chain appears to have stretched 100°! That’s insane! So I used to colon cam to look inside the timing case cover for damaged sprockets and chain guides, but everything looked perfect. Rotated the engine over slowly by hand a few times to listen for abnormal noises, but heard nothing.
So then I rolled a new chain into it. For those who don’t know, on most Mercedes, you separate the old chain then use a temporary link to connect it to the new chain, just rotate the engine until the new chain is in and the old chain is out, then install the permanent master link on the new chain. Easy peasy. After installing the new chain I measured it all out and found timing only improved by about 15°. Well maybe the chain jumped, so I completely correct timing on both banks so I had to pull the other valve cover off as well. After doing that the engine would only rotate about 90° in either direction. Very strange.
Wait a second... What if the timing marks on the crank pulley are wrong? So I pull the #1 plug and manually find TDC. Low and behold, it’s exactly the same distance off the marks as the cam timing. Crap, the pulley rotated on itself, which is so incredibly rare on Mercedes that most (all but one actually) techs I talked to about this never even knew this was a possibility. You see, the pulley consists of the parts: the inner which mounts to the crank, a rubber damping section, then the outer pulley section. Turns out the rubber section became unbonded enough to allow the outer to rotate independently of the inner. The timing marks are on the outer.
Okay, replace the crank pulley and set timing up all over again, reassemble the engine completely and go to fire it up. POP POP POP! Backfiring some more and then completely back to how it was originally. Go back through every diagnostic step all over again to make sure I haven’t missed something. But this time while the plugs were out I measured piston height on every cylinder to rule out bent connecting rods, all equal down to the millimeter. Fine, be that way.
Let’s smoke the intake for leaks, just for fun. Found a few little leaks and fixed them, but nothing bad enough to cause a no start. So now I’m thinking, stretching really, that maybe this car had a whole bunch of little problems all adding up to a no start...
At this point, as I’m sure you can imagine, I’m frustrated and starting to feel defeated, but I’m also intrigued and not about to give up.
On to the only other thing I can think of, the injectors. So I remove the throttle body and charge/boost air manifolds to get to the fuel rail. As soon as I get those off I find they’re disgustingly coated in that milk-shake-looking-crap that you get when oil and coolant mix. That’s really strange, especially since all the fluids are perfect and the air boxes/intake prior to the supercharger are all clean. Anyways, with the charge manifolds off I can clearly see the injectors so I have another guy crank the engine while I watch.
BANG! Engine backfired, my face melted off, and I died.
Ok maybe not, but if you do this, protect yourself, I had full face protection and more helpers nearby just in case, this IS dangerous.
Ok back to the story. Injectors all sprayed perfectly and evenly. So let’s see if we can get to the bottom of this milk shake thing going on in here. Have a look down into the water-air charge air cooler situated below the supercharger and find that it is half full of coolant!
So when the engine cranks it pulls just enough coolant into the cylinders to extinguish the spark, which is why it would try to fire after sitting and ran briefly after drying the cylinders out. Couldn’t smell coolant on the plugs because fuel smells so much stronger. So the charge air cooler is leaking coolant internally, causing the no start condition. Looks like the veritable smoking gun to me. It’s about damn time.