Hi gents,

I have a saga and I'm out of ideas. Hoping someone with more knowledge of these engines has some input.

The story so far. I bought an ML63 (2007) with what was thought to be a blown head gasket. Coolant in the oil and oil in the coolant. I ran it up with some fresh oil to see what the situation was, and it ran fine but blew copious amounts of white smoke from BOTH tail pipes, which I thought was weird (twin system, totally independent exhausts from each bank). Then I compression tested it, and found that cylinder 6 was low. All the others had 175-180 psi, cylinder 6 had maybe 140 psi. Pulled that head off and one of the adjacent head bolts was broken around an inch from the bottom, so not the common failure mode I have seen, but at least it would explain a blown head gasket.

Took the head to a head specialist friend, and he thought he could see evidence of a breach near that pot, checked the head for straightness (good), hardness (good), and vacuum tested it after cleaning the valves. All good. Bought an OEM Mercedes head gasket, new bolts, and re-fitted the head, ran the car up and it wouldn't rev. Took it for a drive, still no good, but no engine codes, no check engine light, nothing. When I returned it was blowing coolant from the overflow. Checked and found that there was oil in the coolant again, and lots of water in the oil too. Compression tested the engine and found all four pots on the side I had done the head gasket on were fine (175-180 psi). But the driver's side (RHS in Australia) had low compression on all four (around 125 psi). Pulled the rocker cover and found that the inlet cam phaser had slipped around slightly, and that was strange, given that I used new bolts and torqued them to spec, but I am prepared to assume that somehow I screwed that up. Can't see how, but leave that aside. I still had coolant in the oil and vice versa, and no obvious reason.

Also, I disassembled the intake manifold and cleaned it thoroughly before re-fitting it. After I removed it the second time I found oil with water contamination in the exit of the new PCV valve I fitted. I think that's odd too, and assume that the crankcase is seeing too much pressure?

Pulled that head off and there's no sign of a breach in the gasket, and the head is straight. The deck is straight too. Took this head to the same head specialist friend, checked the head for straightness (good), hardness (good), and vacuum tested it after cleaning the valves. All good again. Again bought an OEM Mercedes head gasket, new bolts, and re-fitted the head, ran the car up, and within a couple of minutes had water in my oil and oil in my coolant.

I'm thinking a crack somewhere in the block, but I haven't been able to find one.

Also, when I got the car I cleared the fault codes - flat battery, (low voltage so dozens or even hundreds of codes). There are no fault codes now,even after all of the above.

I intend to compression test again, and also do a sniff test on the coolant to see if I have CO2 in it, but I am pretty confident it's not a blown head or head gasket.

Anyone have any thoughts?

Just an FYI, I also dismantled and cleaned the intake plenum, and rebuilt the cam adjusters (phasers), and polished the cams, which are in good nick.Also replaced all lifters (buckets) with new. Loats of money has already gone into this beast, and even more time. Huge amounts of time.

 

If you put pressure on the cooling system, you can send a boroscope into the cylinders and see if you can see water dripping down the cylinder wall. Otherwise, have you checked al other areas where water and oil can mix such as an oil cooler/seal? Im not too familiar with the m156 engine, but these are some of the first things to do when suspecting a failed gasket in the head. Since we cant see all of the pistons in your pic, are any of them "washed/steam cleaned" from water intrusion?

 

God idea on the cooling system pressure and camera.

There's no oil/water cooler, these have an air/oil cooler.

There were no differences between the condition of the tops of the pistons.

I was hoping someone with detailed knowledge of these engines could suggest where the fault might be. I'm wondering for example if there could be an issue in the timing cover area, although the timing chains are not getting particularly contaminated with water.

 

As an owner of 2 M156 engines, a cracked block would be a new one for me. The low compression on the one cylinder could be a number of things but unlikely the cause of the issue here - you would be seeing just oil and not a mix

I've done head bolts on both my engines (both pre-2012 production), and I cannot stress enough to blow out the holes in the block before putting in the new bolts. Even a small amount of fluid in the wholes could skew the torque readings and the heads won't be properly seated.
these engines are pretty sturdy if maintenance properly (catch-can is a must, along with revised injectors, and re-building the intake manifold with a new plate).

the only other possibility is that someone replaced the water pump and/or thermostat at some point and mixed up the bolts. they have different sized bolts on the same part, so if one was to put a longer bolt in the spot where a shorter bolt would have gone, they could theoretically crack the block and allow coolant to enter the oil

I think your best bet is what Baltistyle recommended. Pressurize the coolant system and see where the leak is

Keep us updated!

 

Agreed on the head bolt holes. Cleaned them thoroughly, including using a thin tube fitted to a shop vacuum, then checking repeatedly with a camera, to ensure all holes were dry before fitting.

We did a CO2 sniff test last night and there's no sign of CO2 in the coolant.

I can't see that pressurising the cooling system is going to get us anywhere at this stage. It can't be a crack in a bore or head, or we'd have combustion gas escaping, and we don't have that. So where could we look with a camera to find the leak, other than in the bores?

The water pump bolt suggestion is exactly what I was hoping someone would come up with. We might pull the pump and thermostat housing and take a look. Wherever the leak is, it's substantial. This was fresh oil run for only a few minutes. The white colour is correct in real life. It looks like milk.

 

Water pump and thermostat housing off. No sign of any issues.

 

Only other thing I can think of that's External to the block is the oil filter housing. Coolant flows through there as well but the gasket has to have a major fail or was not included to get the same results you're getting

If the leak was at any of the attachment points externally a preasure test and some soapy water would highlight it pretty quickly. But if the engine is already apart for the thermostat and water pump, getting the filter housing off would be pretty easy

 

Any update here?

 

Given there was nothing found behind the water pump or thermostat housings, I was advised by a local M156 builder that it was either engine out and disassemble, or just use a sealant goop in the coolant. We decided to do the latter and it worked. Now I'll possibly never know where the issue was, but I can live with that.

Next step is to replace the radiator and try and clean out the remaining oil contamination in the cooling system, and sort out some fault codes that remain (MAF sensor open circuit, one injector open circuit, and fuel pressure sensor).

 

Well that was a false hope. The product that we used stopped water getting into the oil, but oil is still getting into the water (coolant).

 

Is this correct? If so, why? Seems like an odd design choice!

 

Here's the oil filter housing. There is no coolant gallery. Only oil.