TLDR;
========================
My 1.8 block is bored to 83.5mm (and has been for years). And for years, through a multitude of blow-ups and rebuilds, I have slapped on a 83mm bore head gasket and called it a day. But I've always had a "mysterious water leak" that I could never track down. Most times I blamed it on the garbage churbo's that I had been using because eventually I would be getting clouds of white smoke at idle.

Well, two years ago (I think?), after the last "blow-up" I decided to go for the gold and get a BW 6750 "real man's turbo" BUT i slapped on the same old 83mm Cometic MSL gasket and called it a day. Except that "mystery water leak" re-appeared. I tightened everything I could think of tightening and I thought I had tracked down the last of the leaks. But when the summer turned to winter and the temperatures dropped I started to notice those billowing clouds of white smoke again and I'm still losing coolant.

I can only think that I have a blown head gasket, and (I think) I confirmed that when I did a test for combustion gases in the cooling system and that tested positive.
====================
My question is simple; Should I be using a 84mm bore head gasket (since a 83.5mm bore gasket is as rare as unicorn tears)? I ran across this article which suggests that is what I should do to avoid the potential for creating hot spots in the combustion chamber at the expense of losing some compression ratio. Anyone care to weigh in?

 

afaik you should always use the same size as bore. Cometic has 83.5mm head gaskets for the BP.

 

I agree it’s best to match the gasket to the bore size.

However I do not think that is the cause of your coolant consumption issue, kinda sounds more like your head may be cracked.

 

The PN for the Cometic 83.5mm gasket is "special order" at Cometic and perpetually out of stock at vendors. So I'll be waiting until forever to get it.

La la la la...I don't hear this...la la la

 

Cracked head is a definite possibility, or even a cracked cylinder. May not be common, but if the usual remedies aren't working, maybe you are looking in the wrong places.

I would not be putting this back together until the cause has been identified with some degree of confidence.

Sorry

 

Okay, time to 'fess up.

This engine was witness to an "incident" two years ago at Road Atlanta where a valve burned through and blow torched a piston. That's when I ditched the churbo and went with the BW. At the time I didn't think about a cracked head, I just wanted to get a running car again - which I did...sorta. I should have been smarter and sent the whole thing out to be inspected ("shoulda, coulda, woulda"). Yeah, it's probably cracked somewhere.

I do have a spare head that was part of a full engine purchase I made many moons ago - it's been collecting dust in my garage for years. I probably should get that inspected (does anyone know of a shop in the greater ATL area that would do that?) before committing it to a rebuild.

As for the block being cracked, la la la...I don't hear that...la la la.

 

Hmmm, well a burnt valve and melted piston is more than likely a result of fuel or ignition timing problems, not really the turbo itself. But between the block and the head, in my experience it’s usually the head that cracks, not the block.

Are you able to pressure test the cooling system?

 

I wasn't saying that the valve/piston was caused by a turbo problem - only giving a more complete picture of what this setup has gone through. I'm praying that the block was unaffected by all of that drama (which means that I'm going to ignore common sense and precautionary measures and re-use the block).one day

EDIT - Adding this for the sake of completeness - and adding a relevant data point - the first time that I noticed the water leak was after driving home from work one day, I noticed that the overflow tank was bubbling furiously and that the CLT values were fluctuating wildly. After cool-down I checked and the radiator was almost bone dry. (It was at this point he knew he had fucked up) This is probably the event that either warped the head, exacerbated any damage from the melted valve, or both.
END EDIT

I did a pressure test on the cooling system when the coolant loss first raised it's ugly head about 6 months ago, and the system held pressure like a champ (which pissed me off at the time because I was looking for leaks and found none). But that is easy enough to re-do.

FWIW - I ordered a 84mm Cometic gasket and will slap that in with the dusty, old head and send it...

 

Ahh, I see. I doubt the block is cracked, between being iron and only a little overbored from stock, they're pretty stout. Plus, I think that if the coolant were getting into the combustion chamber directly, you'd have way more immediate problems with blowing hoses or pushing fluid out of the rad, since combustion pressure will massively outdo cooling system pressures. If I were to guess, I'd think it's probably getting into one of the ports, hard to say if intake or exhaust.

But if you've got the head off it might not be the hardest thing to spot. Was the head resurfaced after that overheating incident? Could also just be warped a little from that event. How did the old head gasket look? Any obviously worn/burnt segments between the passages?

 

Check head hardness. Overheating can soften the head, and it then can't hold a HG.

 

The old HG looked a little suspect between the edge of the water jacket and #1 cylinder, but me not being a "HG whisperer" (and in too big a rush to get back on the road), I couldn't say for sure.
Not having a Brinell Hardness Tester in my garage, how would I go about doing this?

FWIW: Another data point; In the rebuild I installed EGT thermocouples into the exhaust manifold. I kinda ignored them until about 2 months ago when I started looking at them and tweaking the Fuel Trims to try to get the EGTs within shooting distance of one another. I was mostly successful - EXCEPT for #1. No matter how much I tweaked the #1 trim, it was always higher than the others across most of the measured RPM range and wouldn't respond to "reasonable" trims (I felt that any fuel trim number greater than +/- 10% is "unreasonable").

My question here is; Would water in the cylinder cause a higher EGT? My intuition tells me that it would have an opposite effect, i.e. cooling the combustion. Just asking. All of that is kinda "magic" to me.

 

I don't know for sure but I agree with your assumption that water would lower the EGTs. Have you swapped that injector to another cylinder to see if the odd trim stays with the cylinder or the injector? Do you still have the OEM fuel pressure damper on the rail?

Hard to say if they're all related issues but they definitely could be.

 

I did A/B the injectors (and the coils, just for ***** and giggles), and the "high" stayed with that cylinder. I'm only speculating here, but right now, I'm willing to lay out all of my "dirty laundry".

 

Hmmm, interesting. Have you a/b tested the thermocouple? I assume they are all installed the same distance from the head?

 

All are at the same distance from the head (approx 1 inch from the flange of a Kraken hi-mount EM) and protruding the same into the port (I had to adjust that, but they are all the same). I did not A/B the thermocouples. As far as "testing" them...no I didn't do that. They all came from the same source and were "set and forget".

 

My block is bored to 83.5 and I have been running a real Mazda gasket for at least a couple years on this motor now. 0 issues with coolant or oil consumption. Pretty sure it was always standard practice to use a stock gasket for only a 0.5mm overbore.
As for the high EGT with a coolant leak. Who knows what is actually going on in there when you have a tiny breach like that? You have coolant coming in during vacuum, but charge, combustion gases and exhaust going the other way during boost. Must be pretty complex in there, all happening thousands of times per minute. Sounds like you're temp probe is telling you that there is a problem to me.
But WTF do I know?

 

Exactly, WTF do I know as well.

My current theory (until I lift the head and inspect) is that I warped (and/or cracked) the head at the "Road Atlanta Incident" and finally the HG gave up the ghost. So, I'll check it (and my spare head) for straightness, choose the "best one" (or send them both out for re-surfacing), put on the 84mm gasket (and cause a new set of problems) and send it.

The head comes off this weekend and the gasket should be here on Monday. "News at 11".

 

Ok, I just measured a stock Mazda gasket I have in my garage. It is close to 84mm, bigger than 83.5mm, just used a caliper. I suggest asking local shops who they recommend for cylinder head service, when you find a couple recommending the same shop, that's the one to use. Ask them to check hardness and pressure, shouldn't cost too much.

 

Well, the head is off and it looks like there was a "party" happening between 2 and 3;



It seem like #3 bore the brunt of the coolant inflow (the exhaust valves are more wet than any other cylinder). I'm going to measure the flatness of the head tomorrow and see what's up/

Opinions are welcome

 

Oh yeah, 2 and 3 definitely getting into mischief. Interesting that the #4 exhaust valves look so markedly different than the other 3, despite your report that cyl 1 had the erratic fuel trims/EGTS. Curious to hear if/how much the head is warped.