M20 Oil Leak Head Gasket

[88Sedan]

My m20 has a significant oil leak on the PS corner of the cylinder head gasket. I just purchased my 88 325i and PO said motor was a used replacement after the timing belt snapped on original motor, 120,000 miles on motor is what I was told. Other than oil everywhere it runs great, passes smog, sounds like a normal m20.

When the replacement motor was installed, all the gaskets ABOVE the cylinder head were replaced. I think they were hoping something there was the source of the oil leak. In the parts boxes I was given, was a head gasket kit with only the head gasket remaining. Everything seems to point to the head gasket including the location of the oily mess.

Has anyone ever had success with new head gasket/bolts WITHOUT having to have their head surfaced? Its impossible to know if one of the PO's overheated it, but its running well as it sits now and the entire cooling system has been maintained well at the point of sale.

Obviously I want to avoid surfacing the head IF I can. I wasn't planning on full head disassembly, but was wondering how foolish a new HG would be without having the head surfaced.

I really love this ride and want to take care of it!

[packratbimmer]

I have changed out many head gaskets without a resurface and driven them for years. Key is to clean the surface without "polishing" it in any way. Find an aluminum safe detergent and scrub with a plastic brush. The bad news is that after 30-35 years, there are likely to be imperfections in the surface that will require resurfacing anyway. Obviously, check for warping and free movement of the cam.

[88Sedan]

Thank you for your input. While I am going to try and definitively determine it is the HG first, I will certainly update as steps are taken. For cam movement, fore/aft or side to side/lateral? I assume you mean lateral. I am going to try to find the time to do a heavy duty degrease project, hoping to pinpoint the issue. For your HG projects, was it due to oil/coolant mixing or burning, or external oil leak? Curious.

If the head is coming off, I too would be surprised if the head surface looks good. In removal scenario, what is your opinion on:
Head removal, disassembly (marking location of all valves, guides, add NEW valve seals afterwards), surfacing, then reinstall everything? I figure if everything looks good, why not save some dough and just use new seals. I have no idea if m20s are known for valve guide wear. Engines I am used to don't really have an issue there. It passed smog very cleanly, so I would assume things overall are looking pretty good. I am aware to replace the fragile timing belt after removal, despite only having 4k miles on it.

[packratbimmer]

I would do a compression test and a leak down with and without oil added to the cylinders before anything and repeat if necessary. If the head is weak, it might be better to pop a better one on there, even if temporary.

The other quick test is to closely examine the cam lobes for pitting and then do a thorough valve adjustment. Look at the eccentrics for pitting as well and determine that the eccentrics and tops of valve stems are on the same plane - both can wear/get pitted and the valve stems can get cupped which results in a condition where valve adjustment becomes difficult.

Minimally, The cam should spin freely without binding.

Most of my head gasket jobs were preventive maintenance, but I had one with excessive oil in the coolant yet no oil in the coolant (M20B25) - this engine has been in my DD for ten years and runs well. Another M20B25 was great except for fairly robust external oil leaking from the passenger side of the engine.

M20 heads are robust when lubed properly, but are prone to valve train wear including the guides over time especially when abused - much depends on oil additives (lack of) and how the engine was used - 200K is the mark where I would suspect maintenance should be carried out. The height of valve spring pressure and valve lift require some attention to oil choice - read read read about what oil to run and expect to have your head spinning with all of the varied opinions.

Pulling the cam is a slight rubiks cube event the first time, but possible on the bench without specialized tools. I would read up thoroughly before this procedure if you have not done it before. Save three eggs cartons (to organize stuff) for the job and you are ready to go.

[88Sedan]

Alright so here's the plan so far, in order:
1) degrease everything and take a series of short drives around my neighborhood after engine is hot. Each 3 minute drive I will use a light to see if I can positively locate leak points.
2) if HG appears problematic, do a dry and wet leak down test and compression check. If results are ok...
3) pull VC off and check for valvetrain wear, cam wear to determime if I need a new head/cam etc before next steps. At that point in time if wear isn't present, I would honestly skip the valve clearance check because the head is coming off so I would have to do it later anyway
4) head removal, disassembly, surfacing and rebuild head

If the last step gets taken, I am tempted to cough the $$ up for ARP head studs since these HG failures are not rare on m20's. I have even read of issues with new heads later having similar leaks only a few k miles later. Are the stock head bolts the weak point? The HG's? Or are alumimum i6 heads just more prone to this due to their length? What are peoples thoughts on the actual brands/types of HG's themselves? I know Subarus and BMWs are nothing alike, but their HG's were updated after earning them a pretty bad reputation for a few years.

[88Sedan]

Unfortunately I confirmed the HG is leaking. I will post compression test results and leak down results as time allows. Other things were leaking too, but thankfully the oil pan wasn't one of them.

I started looking at m20 head rebuild guides. Is cam wear common for these motors? My toyota cams look new at 160,000 miles so I am wondering what to expect with this motor. What about cracks im the heads? I will have to refamiliarize myself with valve spring height... I seem to remember you shim them to make up for differences in unloaded length? I can't remember if something gets checked in installed form.

What I am hoping for is good results with the tests above, then maybe I can get away with just HG and related seals (valve stem seals and cam seals too plus all the upper items). Of course valve guides will be checked too.

[digger]

Depends on history/mileage/usage of head.

It is very common for cams and rockers to wear, the head will crack over spilt milk so pressure test, skim surface and inspect and replace parts as required. it may need guides but probably not seats. If you need guides then a new VJ and resurface seat on valves is mandatory and recommended in any case (unless it was freshly rebuilt recently)

[88Sedan]

Sorry for my ignorance, crack over spilt milk? Is that a euphamism for yes they crack easily?

Trying to wrap my head around everything to make the best choice. My friend has an extra 885 head so technically I have the option to do a full head rebuild, just trying to save some bucks if nothing is sacraficed.

[packratbimmer]

Do the tests and add a cooling system pressure test. If you break the engine down without base data/information, you are blindly throwing darts. It's possible to throw a perfectly sealed rebuilt head onto a weak bottom end only to find your rings are weak. The good news is that our short blocks are good for 300K+ while the heads lag behind at 200K+ from my experience. You might find that getting the carbon removed from everything (including the valves and seats) will get you back to 90-99% with just a gasket kit and work.

[88Sedan]

So I got what appears to be very good results from compression checking and leak down testing. Here is what I did.

I let the engine idle at operating temperature for awhile, then pulled all ignition wires and spark plugs. I also zip tied the throttle body in open position. I also pulled the fuel pump relay, then one by one checked hot compression.

I checked the cylinders in front to back order 1-6. Cylinders 1-3 all read 180psi, and cylinders 4-6 all read 175psi. My pee break probably let the engine cool down a bit after 1-3, possibly explaining the slightly lower compression in the final 3 cylinders. So far so good.

I didn't run the engine any more in between compression checking and the leakdown testing.

For the leakdown tests, I went in the firing order of the engine to save time and turning the crank anymore than I need to. I made sure each piston was super close to TDC on compression stroke. I removed the air box to hear the intake side of things. I decided I wouldn't bother removing the downpipes unless I got bad results from testing. This turned out to be a good call.

Luckily I got super good results on this test too. All cylinders netted less than 5% leakdown, and all the audible hissing was out of the oil cap hole. I am ectstatic as this means I could get away without a full valve job.

My coolant looks untainted, and I have records showing it had full servicing during the last timing belt change. I don't have a way to pressure check it, but everything points to it being good.

I will continue to update as I finish things up. Now I have to place trust in a machine shop to do a good job surfacing my head. I struggle trusting shops for anything, and will eventually have to take a leap of faith that they won't unlevel the head during surfacing. I don't know where I'll be bringing it yet, but will include that info as things happen.

[packratbimmer]

Awesome testing - great news!!! Looks like you will come out of this unscathed. I feel your pain. It is super easy to EFF up a head while resurfacing it ( you don't want someone taking additional swipes to mop up their mistakes). That's why I try to find a mint head that does not need it. Here are some more points to consider:
1) Measure the head when it comes off and/or look at the divots - check for warping with a straight edge and make sure that the cam turns over freely without binding.
2) If you get it resurfaced, be aware that your valve timing could be off and will need to be kept in check by either using a thicker head gasket (OE BMW) or installing an adjustable sprocket.
3) Buy the best stretch bolts you can get - this is not a place to roll the dice. I think those dudes who advertise here have head bolts and Metric Mechanic might be a great place to check as well. Studs?

[jeffnhiscars]

With good compression and leak down tests im not clear on how you determined the HG was leaking. I'd try adding some oil dye and get a UV light. I thought I had a leaking hg and it turned out to be a cam seal

[88Sedan]

With good compression and leak down tests im not clear on how you determined the HG was leaking. I'd try adding some oil dye and get a UV light. I thought I had a leaking hg and it turned out to be a cam seal

That is a good point. I suppose theres no harm in doing that. I'm about to go on a trip, but will try and source some dye. Where do you find a UV light? Bezos?

[88Sedan]

I should also say, great picture. One more thing added: when PO had motor put in, every seal except HG was replaced. The package the kit came in was empty except HG, and every visible gasket looks good. But I still like your point and will investigate.

[jeffnhiscars]

The dye is from Tracerproducts.com & oreilly sells it for like $8/bottle. I did get my UV light from Jeff. You want one w some power. I wouldn't assume cause a gasket was installed that it can't leak. a friend of mine is a Porsche tech and has had to warranty swap 2 engines in the last month on 2022 921 turbo S car cause of a leaky rear main seal...that's an engine to fix a seal in a $200k car lol

My cam seal had less than 40k on it and for years I was convinced it was a head gasket. Last ditch effort to avoid that was the dye. While I was at it I did the water pump & obviously the timing belt plus...well...a pic is worth more