Not Again! H/G Failure on Forged 2.5 Build
 

Annoyingly after 14 months and 15k miles the headgaskets have gone on my forged 2.5 rebuilt Hawk Sti engine :mad:. The gaskets fitted were RCM 0.78mm with the 11mm RCM studs, and ACL race bearings + Malhe forged pistons.

I can't afford to send it back to the shop so I've started the repair myself. As a starter for 10 I have a few questions:

- any other things to look for that might have caused the gaskets to go again?
- with the acl bearing set, is it likely the bottom end is ok and does not need any work?
- what determines required head gasket thickness, is thicker better/worse, or is to do with head skimming?
- any other useful tips for stripping it down!?

Cheers!

Has the car seen any track work or overheated at any point?

no track but lots and lots of tuning 'pulls' and I mean lots, in the last few months leading up to Christmas. Overheating actually only started last week (Thursday), but contamination of the coolant has been happening for a while, though I was hoping it was something else (naively), as I might have added a small amount of the wrong type of coolant at some point.

Often caused by either over-boosting or airlocked coolant.
What's your max boost set at, and have you/has anyone done any work on the cooling setup?

Wow that must be doing your head in? :(

I don't think I've spotted a H/G failure thread after a forged rebuild.

Where the heads skimmed and did you have any block work done, e.g pinned or cdb?

but overheating as the cause not the symptom, as air lock or ran low on coolant?

I'm pretty sure them rcm 11mm studs are not upto much apart from stock power as I had them and did hg too, with them 1635 studs if same
What power you running?

Aye through my tuning exploits I've hit mid 20s psi and been runnin at about 22psi with an mbc for the last 2 months.

As for power not sure exaclty as it a my own tune on the road, but i reckon about 350 ish. Im using the std vf43 turbo which from what lots of people have said shouldn't lift the heads.

The engine was built with rcm high tensile head studs and was claimed to be good to 420hp (rods as the weak link) however if you read their website the high tensile ones are only good to 350hp, so if these were the ones fitted that might be it.

Interesting, I was getting a little paranoid for a minute :) my forged build should be ready in a couple of weeks. My ultimate goal would be 400 maybe 420 with an SC42 or similar.

My spec is CDB and 14/12 rcm ultimate studs, please tell me I've gone in the right direction ;)

I would have thought if you have been doing "lots" of power runs at 1.7-1.8+bar then this would be enough to lift the heads?

Mine are 11mm ones. The 14/12 i think those are getting towards the real deal according to their site.

It's quite possible you were overdoing it a bit on the tuning runs and unfortunately paid the price

Exactly, 350 rated but I've been informed by a reputable specialist they they have ran 500 without issue, I'm not convinced.
Imo the stock headbolts are stronger, the extreme can cope with more or arp2000 and cosworth etc but them 1635 in not fully convinced tbh.
With the extra tourque your 2.5 does and mi e with 400+ at times with 2 bar.
There's no other weak link in mine.
So going 14mm next and overspeccing

I think the 14-12 are ok, just saves you drilling heads. But proper 14mm will obviously cope without bounds

I was thinking this to, especially if the head studs are lower than expected spec.

Given that I'm not looking for more than 400hp and want to keep the 11mm stud size, what's the best option, rcm ultimate, cosworth or other? From a quick look the rcm ultimate ones appear to have the highest strength, maybe?

:D just minor head work for the 14mm if needed, but I doubt I will go that far. Anyway I set myself a budget on the build of 3.5k and it actually came in at 3.6k :norty: should be well safe upto 450bhp I would think?

If you are DIY mapping what are you using for knock detection?
You need to be using detcans as det can destroy headgaskets no matter what head studs you are using.

I have a knock sensor permanently fitted to the block and use det cans and logger when doing tuning work and a logger for most of the normal driving. I've spent a lot of time logging my daily commute and other drives to smooth out my timing map. Also been using a mbc to adjust to various boost levels to hit different rpms and load cases to test the robustness of the map.
Got a wideband to and spent ages sorting the maf scaling.

The car was running bob on aside from it contaminating the coolant and now overheating!

Does anyone have advice on gasket thickness?

Curret was 0.78mm, should I stick with that or does it depend on if the heads are warped and need skimming?

Its no good guessing until its stripped out, it may well not be the head gaskets, a split liner could be an issue here also.


Were the heads & block faces skimmed on the rebuild ? what sort of finish of the skimming achieved ? were the head studs, if new, were run in on initial torqueing up ?
You also mentioned adding the wrong type of coolant, the Cosworth / RCM gaskets are very funny on the coolant you can run with them, it causes the outer finish of the gasket to pickle up, & detach from the steel layer.


As for gasket thickness, it will require the engine to cc'd to establish the compression ratio, as a rule, with std compression Mahle pistons fitted, block & heads skimmed etc, a 1.1mm gasket is the usual fitment to maintain std OE comp ratio.

Thanks for the advice. As for exactly what was done at the rebuild I will need to ask the guy who built it for the specifics.

For the coolant I'm not really sure because the bottle isn't very forethcoming to whats in it, it was about 200 ml I added about 6 months ago. But that's an interesting point that I guess you can see the evidence of?

Naive question possibly, but what is "cc'd" compression checked?

If it were me for a peace of mind, I'd drill the heads considering it's small work and either try swap or buy proper 14s just for peace of mind.
11 rcm extremes, arp2000 or cosworth high tensile.

And as above. On a 2.5 the liners can crack as their pretty thin and block surface needs skimming on the 2.5s from what I've read.

Personally I think I chose the wrong studs and paid the price for it but the seller was t going to admit that as I was questioning it with them but the usual they've ran loads with them, but if rcm only rate them to 350 :wonder:
Anyway, learn and move on I suppose.

It's measuring your piston tdc volume, head combustion chamber volume, head gasket volume, with a burette to establish its static compression ratio.

... The "cc" referring to the volume in cubic centimetres (cm^3)

To continue this thread... the engine is out and I'm considering the scope of work required. It's common practice to replace the main bearings when the H/G go first time as we know, but what is actual mechanism of failure that means the bearings need to be changed, is it because of oil contamination?

Generally would I be looking to replace the ACL uprated ones as a result of this second gasket failure?

Obviously that massively changes the scope (and cost) of the work required!

Cheers

You really need to strip out to the bare block & get the faces skimmed, heads also, you need to start on a complete fresh level surface, otherwise they will fail again, quite quickly.
You can also asses the bearings, as a rule the Subaru 2.5 gasket issue rarely see's coolant in the oil, the biggest issue with bearing failure after a head gasket issue is usually down to the mileage when the failure occurs, the b/end bearings by this time are worn, stripping & refitting the heads will lead to contaminate the oil system with debris, from various sources, from cleaning up the fire faces, cam cover silicon residue etc, its just enough to tip the bearings over the edge, usually 4 to 6 weeks down the line.

Absolutely nothing wrong with using any of the 11mm studs mentioned in this thread. I'd be interested to know what they were torqued too

Mine we're torqued upto rcm spec sheet that comes with them.

Can't comment on the torque condition of mine as the work was done about 15 months ago by a shop.

As for RCM bolts, from my understanding they used to sell one spec of high tensile head stud, but now they sell 2 specs, 'high tensile' and 'extreme', rated 350hp and 450hp respectively. That makes me wonder if they had a problem with old single spec range (i.e. it was sold as rated as higher than it could cope with), so down graded it and brought in the extreme ones also....... :wonder:

From my personal experience, research and discussions with experts I'd suggest always use 90ftlbs on 11mm.