Godless Commie

iTrader: (2)

1990 Miata, in horrible shape.
Rusted in places, no maintenance at all, severely underpowered engine...
Engine "rebuilt" once with wrong pistons, collapsed lifters, coolant passages completely blocked with rust and sediment.

We decided to save this car, because they are very rare over here in Turkey.
Car was stripped and all rust was repaired, new paint, etc.

I pulled the engine and tore it down, cleaned everything, used a drill to clear water passages, spent a ton of time getting everything right, tapped or chased every single thread, used new bolts or studs on everything.
Block and head were machined to specs, new (correct) pistons, bearings, rings were installed.

Knowing the car's history, I was extra careful when I assembled the engine. Did a proper reroute while I was at it, too.

Long story short, I installed everything back in the car, double checking every step.

Nearing the end of the project, I filled the engine with SAE 30 dino oil for break in, and filled the cooling system with about %40 antifreeze and %60 water.
I would fire up the engine a couple days later, after final checks.

Soon after I filled the coolant system, I noticed a pretty steady coolant leak.
The source of the leak was a complete surprise to me...
Head gasket was just dribbling away all along the intake side and the front of the engine. I tilted the nose of the car up, and air started purging out from the head gasket at the front.
I can see the front clearly, because I had deleted the front water outlet since I had reruoted the coolant flow.

I mean, you can just forget installing a head gasket and torque the head down on the block with nothing in between, and t should still be able to hold the coolant in the not yet running engine, right?

BTW, I torqued the head bolts to 58 lbs in 4 steps, and in the correct sequence. Have done it countless times before, never had an issue.

I immediately drained the coolant and left the car in the garage. I just need to be in a better mood to finish the job, so I took a break.

I started reading up on this problem once I got home. And, found some pretty interesting stuff.
Apparently, this is a more common problem than I would ever know.
Some suggest "running the engine up to temp" with no coolant, and then fill the system, once the engine heat cycles.
Some say start it with water only, heat cycle several times before adding antifreeze.
I am baffled.

I even found this:



By these people:



I will check and see if there is coolant in the cylinders, haven't done that yet.

Should I try the method explained above, or just yank the head, get a new gasket and try again?

technicalninja

I've been building stuff for 40+ years. Never had a head gasket leak like that.
"Machined" to me means both block and head have been machined. Block is decked from crank bore and head has been surfaced below 25 RMS for MLS.
I have put many back together without said machining and not had any leaks. I do check both surfaces with a straight edge and much over .003 warpage I have machined...
I've done "hail marys" up to .005 warpage with copper coat and had them work as well.
Once I built up a blown out section between two combustion chambers with JB weld and that puppy lasted 9 months. I guaranteed 6 minutes. It did not leak like yours.
The "run it till it seals" is new to me as well.

Something is WRONG in my "book".
I would take it back apart and find out what is going on.

The "cleaned out water passages with drill" worries me a bit.
A really bad corrosion problem can "pin hole" castings. I've seen a couple of trashed heads from this.

technicalninja

A piece of trash between the head gasket and block/head might cause this.
A "pinched" line up dowel might cause this.
Also verify you are not bottoming the head bolts out in the block (this is a stretch-pun intended).

Interested in what you find...

Godless Commie

iTrader: (2)

The head was pressure tested, and passed.

Godless Commie

iTrader: (2)

Now that I have had a bit of time to think about this issue, first thing I will do is remove the head bolts, add a couple washers and torque them back down.
If the head gasket holds coolant that way, I will shorten the head bolts when I install the new gasket.

For the record, the gasket is not MLS. It's a Victor-Reinz aftermarket gasket.

fooger03

iTrader: (2)

I had an issue with a used head that had overheated in its previous life. Leaked coolant into one of the cylinders like a sieve. In my case, it had warped and needed to be planed; I suspect your "machining" already covered this, but just wanted to throw it out there just in case it didn't.

Godless Commie

iTrader: (2)

Problem solved.

I removed the head and inspected the gasket. The metal rings around the cylinder bores were too proud.
By that, I mean they were at least 0.5 mm taller than the gasket surface, which meant the head was not making contact with the gasket around the edges where the water passages are.



That explained the coolant leak, but the not the cause.

I had torqued the head bolts in the correct sequence to 58 lbs in 4 steps, using my trusty torque wrench.
My. Trusty. Torque wrench...

It was an "a-ha" moment...
I immediately secured the 1/2 square end of the torque wrench in the bench vise, measured exactly 1 foot on the handle from the center of the 1/2 square end, hooked up my digital hook scale to that mark, set the wrench to 50 lbs and started pulling on the scale.
It clicked on 27 lbs!

Damn thing was way off, by %46.
I had torqued the head bolts only to 31 lbs. Apparently not enough to crush the metal rings. Pretty much the equivalent of hitting the cylinder head with a purse.

I calibrated the wrench using the digital hook scale, took less than 10 minutes.

Reinstalled everything, torqued head bolts to 62 lbs to be on the safe side, started and heat cycled the engine, no leaks.

tl;dr make damn sure your torque wrench is properly calibrated.

(Engine runs great, btw)

turbofan

iTrader: (1)

What a story, haha.

DNMakinson

iTrader: (3)

Nothing like going back to the basics. Thanks for the report.

technicalninja

Glad it was simple.
One of my rules is "always test your test equipment"...
I have 5 torque wrenches acquired during 40 years of wrenching.
I've had my expensive Snap-on 1/2" 30-200 recalibrated 3 times professionally.
I've spent as much as the torque wrench originally cost on recalibrations.

What I've done is use it to check/test the other ones.
They are never far off but I run them back down to first torque (30 lbs on the snappy) immediately after using it and NEVER leave them "spun" up.

The one that appears to be the most accurate over time is an old (maybe 50 years)"triple beam" that has no moving parts!
Don't throw triple beams away...

What's up with the 45 degree cuts on the upper surface of the intake valves?

Godless Commie

iTrader: (2)

technicalninja

Yep, I know about that. Best is a tiny little radius.
The "margins" in the picture you posted are too small on the intake valves.
The resulting edge does not help flow (especially on the intake) and will reduce the life expectancy of the valve.
Your exhaust valves have a tiny little bit of this radius going on and are "perfect" IMO.
In a stock NA application the intakes will work fine, they just will not last 100K miles.

The most common mistake on head work is "going too far" and someone has done this with the top cuts on the intake valves.
The top cut helps flow by the exhaust valve more than an intake; think about direction of flow...
The exhaust valves look excellent in this regard.
But the margin is critical for valve longevity. I throw valves away when they have no margin. The margin in the illustration is fine.
The hotter the chamber temps are the more critical this is. FI increases chamber temps significantly and your application is NA...
What you have pictured will work fine on a NA stock rebuild but if you do a serious head you will want a different approach.

On the valves itself:
Back cuts (the 30 degree cut above the 45) help intakes more than exhaust. When it is really done right it will be two angles instead of one. That illustration needs a tiny 20 degree cut above the 30 (you can see the "hump" in the illustration). Any transition above 15 degrees "shears" flow and creates turbulence.
Top cuts help exhaust more than intakes and when it is really done right it's a radius (the illustration mentions this).

On the seats:
Intake seats should be multi angle (more than three) but throat diameter is the most critical aspect. Most American engines have throats that are too small and most imports have throats that are too big. Gen 3 and up GM motors are the exception to this rule, They have big enough throats to begin with.
The fix on American stuff is to open the throats by porting and blending and on the imports the fix is to install bigger valves WITHOUT hogging the ports out further. Intake throat diameter should be approximately 87% of valve diameter.

Throat diameter is less critical on exhaust (87% rule works fine here) but the seat should be a smooth radius with the only straight part the actual seat for the valve. In an endurance or street motor the 'flat" (seating surface) should be normal width to allow proper valve to seat heat transfer. In an "all out" maximum effort set up the seat flat is so small to almost appear non-existent. This doesn't live long...

The angles of the seats (everyone thinks 45 degree is the only way to go) can also have a significant effect. Running the intake seats at 30 degrees instead of 45 helps initiate flow sooner which means better cylinder filling. They are less reliable (IMO) and will need touching up sooner but just changing this angle 15 degrees is worth 3-5% more power. This trick works on NA applications only, force feeding the engine negates this. David Vizard is my source for this info; this can be verified with a bit of internet searching.
The exhaust seats flow best at 45 degrees on a smooth radius for both NA and FI.

The finish inside the ports is important. Intakes should be rough, the finish left by a carbide cutter is best, an 80 grit porting roll is almost too fine. A rough finish helps laminar flow. The exhaust needs to be as smooth as you can get it to discourage carbon fouling. The finish achieved with ceramic coating is best.

Another minor issue in your picture is the amount the spark plug extends into the chamber. I would shim them a bit to reduce this (no threads showing at all) and rotate the plug 180 degrees to have the ground strap positioned between the intake valves. This will reduce the thermal load on the ground strap. This is called "indexing" the plug. It's not nearly as important on a 4 valve crossflow head as it is in a 2 valve Siamese head. The ground strap should always face AWAY from the exhaust valve.
This doesn't make a rats *** of difference in your application but IS important in a 400HP BP build. Variable thickness copper washers to change the "clocking" of the plugs are available.

I would also deburr the edges of the chamber that the head milling created. This is most notable on the intake side. I wouldn't cut an actual radius just make the transition between the milled surface and the combustion chamber "smooth". It almost looks like this has been done on the exhaust side of the chamber.

I received a neg cat because someone thought I didn't know about top cuts.
I do.
I'm best at flow and temperature control. I've been actively porting heads for 30+ years. I'm "happiest" when I'm covered in aluminum dust and the air-powered die grinder is frozen to my glove...
I've sneezed aluminum for days after, it's probably bad for my health.

Anyone who has ported an aluminum head will understand...

DNMakinson

iTrader: (3)

Google did not help me find triple beam? I understand old beam style torque wrenches, but what is this triple beam of which you speak?

sonofthehill

iTrader: (1)

I think triple beam is a joke on the beam torque wrench. Like the old scales vs digital.

sonofthehill

iTrader: (1)

I personally always use a beam wrench on anything critical.

technicalninja

Triple beams is what I call them because 30+ years ago it was more common to see them with 2 fixed together main beams and one indicating beam.
I'm talking about normal beam torque wrenches, I am speaking in "outdated" language which I am prone to do...
I was not joking.

Why did I receive negative cats on the previous post?
When I'm wrong I'll post my retraction and learn from the mistake...
I will NEVER edit the incorrect post out. DNMakinson got me on one bigtime. It's still out there. He was right!
The above info is correct from my point of view.
I have decades of experience modifying heads.
If someone disagrees with the information please post why.
I will respectfully research your viewpoint and possibly learn something new.
To me learning something new is better than getting a new tool.
If I agree with you I will do so publicly on this forum and thank you for it...

sonofthehill

iTrader: (1)

I thought it was funny Triple Beam, reminds me of some old rap song, "Weighed it on the triple beam."
Don't mind the neg cats once in awhile, "Anyone" can give them, even me. I sometimes get them because people disagree with my opinion or I am unaware of a certain product.

sonofthehill

iTrader: (1)

All the cats I ever scored from had a triple beam which they had invariably stolen from their local high school.

Godless Commie

iTrader: (2)

I was a bit thrown off with the whole triple beam thing but I just thought it was a language limitation moment for me.
Negcats are a complete mystery, that I just don't get.

technicalninja

You have it far better than I.
"Triple beams" means an Ohaus 700 series scale. I OWN one. It's old too. I use it to weigh internal engine parts not for clandestine business dealings...
I'm using that term WRONG! I will not do that again, Thanks!

It was a "language limitation" for me, NOT you!
Sonofthehill set me right..