The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support ExtremePSI
Please Support STM Tuned

2 Cometic Head Gaskets Fail the Same Way - Ideas?

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

VETDRMS

15+ Year Contributor
340
21
Jun 1, 2006
Vancouver, Washington
As the title implies I just finished removing my 2nd Cometic gasket after it started to pressurize the coolant system.

The first failure happened at relatively low boost (24-25psi) and conservative timing for E100 (14* total, 7-8* @ 5000).

Both the head/deck were prepped for MLS to a mirror finish. The block deck was checked for flatness. I am using A1 H11 studs torqued to 95 ft/lbs (30-60-95).

The first gasket was a Cometic 87mm (.066" thick), the second gasket was 87.5mm (also .066" thick). Copper spray was used on both.

The first time it failed it was mostly at the back of cylinder #2, right in the middle. After cleaning the deck surface it also leaked in the front of #2 and the front/back of #3. Each point of failure was directly in the middle and directly in line across the cylinder (front to back).

Upon the first disassembly there was some overhang of the gasket into the chamber due to too much material being taken off when deshrouding the valves. The head was welded up, seats replaced, and re-machined.

It failed at the same boost/timing on the 2nd gasket. It held fine at 22psi. It failed in the same spots as well.

I've attached a few photos of the first failure, but the second one is about the same. Any ideas of what could be causing this?

Also, what other gaskets suggestions are there for the 2.4?

I was considering using an OEM Galant composite and having the head machined for SS o-rings. Can this be done on the 2.4? I have heard that it is difficult and not recommended to machine the 2.4 block for o-rings due to the larger bore.

Thanks, hopefully the third time will be more charming. OMG
 

Attachments

  • 2-Front.jpg
    2-Front.jpg
    40.7 KB · Views: 504
  • 2-Back.jpg
    2-Back.jpg
    45.6 KB · Views: 556
  • 3-Front.jpg
    3-Front.jpg
    38.7 KB · Views: 474
  • 3-Back.jpg
    3-Back.jpg
    43.9 KB · Views: 462
please post a couple pics of the head deck surface and the block deck surface.
 
The only pic I have of the head readily available was before it was lapped.

did you take the valve cover off and recheck the torque on the studs after going through a few warmup and cooldown cycles before boosting?

Yes, the studs were re-torqued after the first full heat cycle. I also re-torqued them after the first sign of pushing coolant.
 

Attachments

  • LappedDeck.jpg
    LappedDeck.jpg
    59.9 KB · Views: 381
  • 1111101136.jpg
    1111101136.jpg
    34.6 KB · Views: 402
I would try not using copper spray on the MLS's. Ive had better experiences without it. Composite gaskets I use it on, but not MLS.
 
I would try not using copper spray on the MLS's. Ive had better experiences without it. Composite gaskets I use it on, but not MLS.

i was under the impression with a cometic MLS you actually aren't suppose to use copper spray? am i wrong in this research?
 
Vetdrms... thanks, but both are hard to read the surface, the head looks like it was cut with a segment stone grinder or a carbide rotory broach. it is hard to tell from the angle.

the surface still may be an issue.

if you look thu this thread, you will see the surface finish I perfer for a MLS
http://www.dsmtuners.com/forums/new...ce-4g63t-cylinder-head-mls-head-gasket-2.html

The head was cut with, I believe, a 14 tooth Storm Vulcan decking machine. My friend runs the shop and installed all new cutters before doing the head at the slowest setting. The head was then lapped with 1200 grit using a large (maybe 10x10x3 machined steel block). The block deck was prepped the same way.

I might have some better pictures of it at home.

i was under the impression with a cometic MLS you actually aren't suppose to use copper spray? am i wrong in this research?

I used a very light coat, I have seen both recommendations for its use on MLS and against. I also talked with a tech at Cometic, and, while they said they don't specifically recommend it, it should not be detrimental.

perhaps undetected detonation using e100?

Pretty unlikely given the boost/airflow and timing. The E100 should be very detonation resistant. I am using non-projected tip BR7ES NGK plugs. I pick up some noise on the knock sensor that results in .4-1* of knock retard here and there, but no huge spikes ever. The most knock retard I ever saw was 1.8*.
 
Last edited:
have the head resurfaced... thats a rotory broach.. I will bet it is still to rough can you drag a finger nail across it and feel the ridges in the head.
I would also say that is why you have the swirll pattern on your gasket

with having to set 14 cutting teeth is time consuming, and can take several hours to index them even close to equal and still be off by .0005

no disrespect to your friend or a fellow machinist, but try having the head surfaced on a Comec or DCM that takes a single insert

also you may stop at a office supply and get a sheet or 2 of NCR paper ( I think Victor Gaskets has some head sized) and with out the head gasket torque the head to the block and read the pressure readings
 
Thanks. Does the point of failure (side of gasket) have any relation to the surface responsible for the failure? There are no signs that any of the leaking was on the head side of the gasket, only the block side.

I will inquire about the type of milling machine used at a shop recommended to me that is local.
 
yes it does since it is the block side that would be the side I would look at close .

found the NCR paper I was telling you about

NAPA gaskets by victor reinz
catalong number nmg-300-04
march 2004

2004 master gasket catalog

pg 1243

Carbonless impression paper
part number JV311 2 sheets 10" x 30"

there is an easy and accurate method to check the flatness of mating sirfaces sealed by gaskets-carbonless impression paper.
This method is particulary useful for checking the flatness of the cylinder head and block where sealing is critical.
 
Thanks, I will pick some of that up and report back.
 
That head doesn't look like it was surfaced on a broach, the OP described what sounds like an old reliable carbide head 85b. Also, there is nothing wrong with a broach or a wet stone surfacer if they are dressed properly, and you can get sub-50ra finish even on aluminum.

Where did you buy the gaskets from? I ask, because they are more than likely cometic copies. I ran into this exact same problem on a 2.3 I built for a friend last year. Both surfaces were fresh with new ARP's and it kept on blowing exactly where yours was along with in between the cylinders. I had to eat these HG jobs myself, and after the second one blew after 100 miles of highway driving, I put a Mitsu MLS on it for him and there have been zero problems since. I got the parts house I bought the gaskets from to exchange one of the old blown ones for a new one, and the only text on the replacement is "High Performance 4g63" in white paint. Definitely not a Cometic, but it was sure sold as one!
 

Attachments

  • 2011-08-09_17-46-44_259.jpg
    2011-08-09_17-46-44_259.jpg
    38.5 KB · Views: 306
The attached picture is what the cutter that was used looks like. It has carbide cutters.

The first gasket was from ExtremePSI and it came in real-looking Cometic packaging, the second gasket was a custom order from Cometic. (Cometic charged $48 for 3 day shipping! LOL)
 

Attachments

  • mcgraw-85-1.gif
    mcgraw-85-1.gif
    61.6 KB · Views: 2,024
Yep, that is the 85b blockmaster. That machine set the par for surfacing for many years until the new cbn machines took over. You can buy a cbn conversion head for that machine, so it has the potential to be as good as anything on the market today!

I might have missed it, but did you mention how many miles were on each gasket before it failed?
 
The first gasket had probably 1800 miles on it. The 2nd one had 600. The first 1000 miles weren't at higher boost.
 
Well, although the surfaces were prepped, Bogus's suggestion of impression paper is a great idea to see exactly how your head is clamping down onto the block.
 
They both let go at 25psi, but held fine at 22-23psi. This is low 50lb/min airflow range. It was supposed to go on the dyno, scales, then track, but this happened.

At 25psi I was logging 0-90mph in 6.3s (speedo gear matches tire size). It is a full weight auto 2g with power everything plus me, i'm guessing 3600lbs.

Does the head need to be torqued down to spec (95 ft/lbs), or can i just tighten it up on the paper to say 20 ft/lbs?
 
torque it like you would be installing the head, so 90ft/lbs

wow a old block master, I havent used one in 15 years!
there is so many variables in that machine, when the inserts are set up from how the blade is sharpened, the angles, how many times each blade has been sharpened. the wear in the cutter head bearings, any warp in the tooth fixture.

but in all fairness, it boils down to the care and mantinace of the equipment, and the machinist that uses it.

the block master can be and still is a good machine to use. just like the old Van Norman 570 it still cuts great for cast iron heads. But my feelings on alum heads, and MLS gasket sealing... use a more modern machine.
 
I took some pictures of the head, gasket, and block last night.

Head, #2, #3:

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.



Gasket, #2 back, #2 front, #3 back, #3 front:

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.



Block:

You must be logged in to view this image or video.


Also, here is the head before it was installed after being decked and lapped:

You must be logged in to view this image or video.
 
Hard to call from the pics looks like you are not getting the proper clamp force around that area. and like you, I ask why? from the pics it looks good and for my own info, would you measuer the depth of the oil trough from the cotner bolt hole?
http://www.dsmtuners.com/forums/newbie-forum/341028-4g63t-head-oil-port-mod.html

You can see where the fire-ring imprint is "light" in some places, which, to me suggests less pressure in these areas. It is pretty obvious at two of the fail points (front of #2 and #3). I hope the straight edge will be available this afternoon so I can check the head/block tonight.

Should this be checked from end-to-end or front-to-back?

I will measure the depth of the oil trough. With the amount of material taken off the head I'm sure it is not very deep. I didn't worry about it as I am using a .066" thick gasket.
 
I could see it was light.

have you picked up the impression paper yet?

I would check it both ways, paying carefull attention to the "ball cap" shaped water ports, from what I see this is where the water was comming from.

Whqat I have learned is that trough should be appox .120 deep including the HG thickness
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top