• Protip: Profile posts are public! Use Conversations to message other members privately. Everyone can see the content of a profile post.

Head Gasket Repair - Part 1

Joined
27 July 2007
Messages
7,454
Location
Denver, CO
As many of you know, my 1991 NSX suffered a head gasket failure at the 2012 NSXPO during the track event. You can read about my experience here. Well after a long discussion and much negotiating, I am very pleased to report Mile High Acura stepped up to the plate and agreed to repair the damage. They got rid of the guy who messed up my engine and hired a new tech, Michael M., who builds import race engines as a hobby and is a super competent tech. They even let me interview him before letting him touch my engine and he definitely knows his stuff. Since there are almost no threads like this on Prime, I thought I would chronicle the repair so other folks with questions might find a few answers here.


Day1: The engine is out! We made sure to use the 1997 service manual procedure, as the original 1991 manual calls for separating the ball joints, which is bad juju on the NSX. Also, my car has 1998 3.2 manifolds on it, so the 1997 procedure for removing the heads also applies.

IMG_2888.jpg


Day 2: Lots of parts are here from Honda. Replacement main hoses, oil cooler hose, plus a few extra bits I've been meaning to replace. New battery ground, new ignition switch to replace my spliced one (from aftermarket alarm). Also, new A/C idle pulley, since my 23-year old one is getting noisy. Also, new PCV hose for dressup appeal. Note, the head gasket kits which have part no. 06110-PR7-020 are actually packaged by Stone Gasket (Ishino), who is the OEM manufacturer for the NA1 gaskets. Interesting that the new intake manifold gaskets are white now instead of the original green. The kit has everything you need for one complete cylinder head, including new valve seals! So, you should buy two of these kits to do the engine.

DSCN2022.JPG


DSCN2031.JPG


DSCN2027.JPG



Also have new parts for the intake manifold overhaul, which I am doing myself to save cost. I will have a separate thread on that.

DSCN2036.JPG


Day 3: Top of the rear head ready for removal. This head was heavily varnished from old oil last year when I re-did the valve covers. Amazing what a year of daily driving with Mobil 1 can do to clean it- much better!

DSCN2037.JPG


Front head- still varnished, but vastly better than a year ago. Michael tells me the head with the PCV valve always is more dirty on all Hondas, not just the NSX. Makes sense. Also, you can see my new updated LMA's from Science of Speed, installed last December.

DSCN2038.JPG


First sign of trouble. I should have gotten a better angle, but you can see some corrosion at the seam on both heads. This is evidence of the head lifting and leaking coolant.

DSCN2039_nsxp.jpg


Rear head is off. Incredibly, no sign of any coolant leak anywhere on the deck. Strange. Pistons look good for 90,000 miles and we could still see the cross-hatching in all the cylinder walls. #1 and #3 do look like they were running a little rich at some point in their life.

DSCN2041.JPG


Rear head gasket. Again, no sign of blow through or damage (the tearing is from removal).

DSCN2042.JPG


Front head is off. Again, no signs of damage on the deck and the head gasket was intact.

DSCN2052.JPG


Rear head on the bench. Now, we start to get the picture. These pockets of corrosion on #1 are a clear sign of head lift. Coolant was seeping in and getting trapped here and corroding the aluminum. Front head had the same corrosion.

DSCN2044_nsxp.jpg


And finally, the smoking gun (no pun intended). All 16 of the head bolts had these rings of corrosion around them. This is clear evidence of the heads lifting completely off the block and letting coolant into the bolt sleeves, which rusted the bolts in this pattern. Think of the steam pressure in a boiler and immense force it can exert. Same thing here overcame the clamping load of 8 heavy bolts. Please make sure to check your coolant and make sure there is no air in your system or it can do this! When we compared these bolts to the new ones, several were at least 1 mm longer.

DSCN2045.JPG


Never thought I'd see my NSX like this. :( But, it is car porn for sure! :D

DSCN2049.JPG


More to follow later...
 
Also subscribed. I know you like to do things oem refreshed by why not upgrade the head gaskets since they are known to fail (especially in FI)?
 
car porn for sure bro. I love it keep it up she will be driving in no time.

Mobil1 is the best, also add a cup of sea foam will clean it up even more.

awesome.
 
Nice thread. Excuse my ignorance though, why would air in the coolant cause the heads to lift? Or for that matter cause your head bolts to stretch? I'm a little confused on that comment. Never the less, good luck! I hate seeing when my cars are all apart but soooo much better when it's all back together.
 
you are lucky to find a dealer that allows you that much access to the bays,and lets you do some of the work.As much as we often hate on dealers ,if you can forge a good relationship you have the comfort knowing they are supported by the manufacturer ,and have the resources to get the car to spec.
 
Also subscribed. I know you like to do things oem refreshed by why not upgrade the head gaskets since they are known to fail (especially in FI)?

Yes. I would 100% use cometic gaskets and not Oem.

Heavily researched and considered this. Based on discussions with NSX experts here and in Japan, we determined that MLS gaskets were not needed for my plans with the car. Honda switched to the MLS for the NA2 because of concerns over the reduced thickness of the wall between cylinders. There is nothing inherently wrong with the NA1 composite gasket on a 3.0. There are plenty of NSX's out there running hard on original factory gaskets, including supercharged and turbo cars. Indeed, our own LarryB has no concerns with using the composite and has done so on several cars without issue.

Also, if you look above, the gasket actually did not fail in this case. Instead, the head bolts could not hold under the extreme steam pressure from the air in the system. Even under these conditions during several trips to the track, the gasket held together so well that the car was completely driveable if you kept the revs under 5,000. Honda reliability. :D

In addition to LarryB, Kaz recommended MLS only for a high-compression race engine or FI. Since I have no plans for either, composite was fine for me. If Larry and Kaz give you the same answer, you can pretty much take it as gospel for the NSX. Also, MLS gaskets require additional prep, as the surfaces need to be polished/machined to almost mirror-like level to promote a good seal. Also, you need more clamping force, so ARP studs are recommended. That's almost $1,000 more cost for capability I don't need.

- - - Updated - - -

Nice thread. Excuse my ignorance though, why would air in the coolant cause the heads to lift? Or for that matter cause your head bolts to stretch? I'm a little confused on that comment. Never the less, good luck! I hate seeing when my cars are all apart but soooo much better when it's all back together.

The air finds it way to the open deck around the cylinders and just floats there. Under high heat/load, those air pockets boil the coolant and cause a steam flash. If it is a single air bubble, it often will perforate the head gasket and warp the head in that one spot. In my case, the car sucked in so much air through the loose bleeders that the bubbles were spread all over the deck, so they lifted the heads uniformly. Actually this was fortunate, since it is the isolated bubbles that can bust a gasket and/or twist a head.
 
you are lucky to find a dealer that allows you that much access to the bays,and lets you do some of the work.As much as we often hate on dealers ,if you can forge a good relationship you have the comfort knowing they are supported by the manufacturer ,and have the resources to get the car to spec.
Even more lucky that the dealer agreed on fixing this. I would not find any dealer in Europe who would do this. If something goes wrong -> they have the dumb customer who pays what they call him for.
 
Head Gasket Repair - Part 2

IMAG0347.jpg


After much cleaning and scrubbing, Michael got the heads all pretty. We measured with the straight edge per the service manual and THE HEADS WERE FLAT! :D Only a couple spots had barely a 0.001" gap, which is within Honda's tolerance spec (up to 0.002" is ok with the composite gaskets). So, good news that I don't have to get the heads skimmed.

IMAG0341.jpg


Rear head fitted and cleaned to this level. Nice. Michael also installed all new valve seals, which turned out to be a good thing as there was oil on top of some of the valves. Something for Primers to think about, as my car has 90k miles. It was not smoking that I could see, but based on this info, I was not far away from that. Finally, he did a nice job cleaning up the ports and the combustion chamber. Not technically part of the repair, but he is meticulous and cares about this engine. Great tech.

IMAG0342.jpg

IMAG0343.jpg

IMAG0344.jpg


Torquing the rear head down. Michael was absolutely obsessed with getting a perfect torque on each bolt in the proper sequence. We used liberal amounts of fresh oil (you can see the cup) and assembly lube, but a few of the bolts were still chattery. Michael did not stop until he got each one just right. Also, I was drooling over the SnapOn digital torque wrench. As this car is tracked and Kaz has indicated the factory bolts have loosened as much as 5 degrees during the NSX engine development, we decided to torque to 60 lb/ft instead of the specified 56 lb/ft to add a safety margin. We torqued in three steps: 20 -40-60 following the factory pattern. Also, we used high-tack spray on the gasket, as it will flow under clamping load to fill in the surface imperfections in the metal and improve the seal- just more safety margin. A trick Michael uses on his turbo race engines.

IMAG0353.jpg


Front head cleaned and mounted. Still some varnish, but Michael says to keep driving it with Mobil 1 and it will clean off. Also, we found that my PCV valve was bad and replaced with new, so that was contributing to the problem too. Cam journals were mirror-like with no visible wear. Yay Mobil 1!

IMAG03551.jpg


Front spool valve mounted with new gasket and bolts. Note new rocker shaft plugs and washers- I replaced all 8 as some were leaking.

IMAG0349.jpg


Same deal on the rear head. Nice and shiny.
 
I'm glad an acura dealership actually stepped and owned up to their mistake and had a competent tech work on your car.
 
Last edited:
Looks like you might have snapped one of the heat shield bolts on the rear header. Here's an idea you might want to try. Take a punch, or reasonable facsimile, and whack what's left of the bolt remaining in the nut, which will break the nut loose from the bracket. Then install an M8 speed nut from your local auto parts store. http://autoplicity.com/masterimages/full/GAPI3/50/509417.gif.ashx?maxwidth=328&maxheight=228

Funny you mentioned this. We fired it up and it threw a rear O2 code- there is a crack in the rear header right where that broken bolt is. :( Checking with Mark @ Pride to see if he can scrape up a set for me. :D Anyone have the Pride headers? Feedback?
 
Head Gasket Repair - Part 3

IMG_34462.jpg


Time to set the timing belt. We used bright yellow paint so that there was no error or doubt. Michael also removed the #1 spark plug so we could confirm #1 TDC for sure. Screwdriver in this pic shows #1 TDC on the crank marker.

IMAG03711.jpg


Rear head marks lined up on the cam wheels.

IMG_92481.jpg


Rear head marks double-confirmed on the head side.

IMG_43731.jpg


Front head marks lined up and double-confirmed.

Tensioned the belt and with the valves adjusted all the way out, we turned the crank 1 revolution to make sure the marks lined up properly when back at #1 TDC. They did. :D

Unfortunately, once the engine was back in the car, we fired it up for the test and were greeted with a ear-splitting roar from the rear manifold and a rear O2 sensor code. Turns out I had a crack in the manifold near the broken heat shield bolt and all the twisting we tried on that bolt pulled the crack open. So, we are down until my new Pride headers arrive- ETA Thursday hopefully.
 
Hey Honcho, great thread. I've learned a few things from it. How has everything held up since your did this back in 2013?

I just removed my heads today. I had a coolant hose burst, and overheated the motor. The #3 cylinder had 26% leak, so here I am.

I was wondering how you removed all the gasket gunk? I was thinking a thin paint scraper, but don't want to mar the aluminum head surface. Any advice?

I hope your motor is running well and you're enjoying your NSX!

Mike B

 
Mike B - Honcho sold his car awhile ago and bought a GTR. AFAIK, the rebuilt engine is still going great for the new owner.

Sorry to hear about your engine.

What HG are you putting in your rebuilt engine? Do you have a C30 with the OEM paper HGs, or do you have the C32 with the OEM MLS HGs? If you have a C30, are you keeping the paper HG like Honcho did, or are you planning on running the Honda MLS or an aftermarket MLS? If you go to MLS, you will need a smoother HG mating surface on the block and heads according to the HG manufacturer specs. Therefore, you can gently scrape with a plastic scraper and then use a mild scotchbrite pad (I forget if red or green is the least mild) to clean it up.

Also, you may not be aware, but there have been quite a few instances of the head stud threads pulling out of the block. It depends how much you were planning on doing to your engine, but it seems to be more common these days to go ahead and Timesert the threads in the block.

See this thread:
http://www.nsxprime.com/forum/showthread.php/161326-Help-leaking-coolant?

I have a Timesert kit if it is something you are interested in doing. If so, please PM me.

Dave
 
Interesting old read as I refresh my '91 3.0 too. I have followed the SM and removed the axles...no bad "juju" found. Replaced the boots while they were out. Plan on using the MLS head gaskets from SOS, since the graphite composite gaskets are old technology and did leak into the head bolt tunnels (not due to head gasket failure, per se, but due to its age and technology limitations). I haven't measured my old head bolt lengths, but would hope they are stretched (and narrowed) since that's what they are designed to do during the torque sequence. Anyway, michaelbrat, I used several new and sharp razor blades to gently remove the graphite gasket remnants. Hard to remove graphite was finished off with a brass brush attachment on a Dremel tool at relatively low RPM.
 
Interesting old read as I refresh my '91 3.0 too. I have followed the SM and removed the axles...no bad "juju" found. Replaced the boots while they were out. Plan on using the MLS head gaskets from SOS, since the graphite composite gaskets are old technology and did leak into the head bolt tunnels (not due to head gasket failure, per se, but due to its age and technology limitations). I haven't measured my old head bolt lengths, but would hope they are stretched (and narrowed) since that's what they are designed to do during the torque sequence. Anyway, michaelbrat, I used several new and sharp razor blades to gently remove the graphite gasket remnants. Hard to remove graphite was finished off with a brass brush attachment on a Dremel tool at relatively low RPM.

The NSX head bolts and even rod bolts are not torque-to-yield (TTY) and therefore are in the elastic deformation zone under tension and return to their nominal dimensions once unloaded. These may therefore be reused if you'd like. I'm not sure what you'd compare your measured bolt lengths to anyways, since we do not know what length they started out at. Some engine manufacturers do use to TTY bolts as they are more consistent. However, the drawback is that they can only be torqued once and then thrown away (which usually isn't an issue).

On my rebuild, I bought the ARP head studs and ARP rod bolts, but you can also reuse your OEM ones.

Dave

Dave
 
I purchased new head bolts since all down-side bolts showed surface pitting from coolant exposure. Mac Attack, your comment makes sense, since there is no step in the NSX SM to measure the diameter of the used head bolts like there is in my AP1 S2000 SM. Apparently, the S2000 head bolts are TTY.

I took a few measurements of new versus used head bolts. Both have similar O.D.s, but the used bolts are up to 1mm longer than the new. They are stretched to some extent during the torque sequence.
 
Last edited:
I was wondering how you removed all the gasket gunk?

I have found it very easy to use a wire wheel on a drill to clean off the block. If you overheated due to a coolant hose, the heads need to go to the machine shop for flattening anyway, they will take care of the heads for you;)..

HTH,
LarryB
 
Back
Top