Cylinder head flatness - is milling really necessary

[Zed Head]

I decided to mess with my old engine parts, and put them back together for easier storage.  It's an N42/N42 combination, that ran well, but had leaky valve seals and a coolant leak from the back of the head gasket.  After I took the head off I used a long straight level and measured .007" of upward bow in the center of the head.  By the FSM, that's out of spec. (.004" maximum) and the bottom should be reground to under .002" warpage.  

 

I set the head on the block and rechecked flatness, measuring the gap in the middle.  I got .006" this time, which makes sense since the contact points are closer together because my front cover is lower than the deck.  The head was only touching on the ends of the block.

 

So then I inserted a single head bolt in the center and with a few easy twists of a wrench, not even using the end of the wrench handle for leverage, was able to reduce the clearance to zero.  The head was easily flexed down to good contact, end to end.

 

So, it seems to me, that if there are no surface defects and the "warpage" is gradual, that a person could probably just bolt up an out-of-spec. head and run it.  The only concern might be stretching the head gasket as the ends of the head push out when the center comes down.

 

Just wondering.  More food for thought.  Has anyone out there just said "screw it" and run a warped head, and what were the results?

[Chickenman]

Interesting question. I " think " you could get away with it. While .007" is beyond factory specs, it is a long cylinder head and the head is made out of aluminum.

 

The head gasket is the problem. A flat steel plate doesn't compress, but a Multi Layer head gasket does.

 

There is a way to " fudge " things though. Use variable head bolt torquing. Get some brand new " Turbo " head bolts and torque in normal spiral sequence. But on the final tightening, torque the head bolts near the center 5 ft lbs more than the outer head bolts. Use the Turbo head bolts as they will take the extra torque.

 

That should give you an even compression of the head gasket, or " more even " than if all of the head bolts were torqued to the same amount.

[Chickenman]

If you're putting the head and block into storage, here's an interesting experiment. Bolt the head to a block with no head gasket. Bolt down the center part of the head only. Try 40 ft/lbs torque.

 

Come back in 6 months or so and check the head and see if it's straightened out. I'm betting it will.

[Mark Maras]

I like Chickenman's idea. I think it would work fine. I would check the torque a couple of times after it has a few miles on it. What head gasket were you thinking of using? I think I'd go with the compressible one.

[Zed Head]

 

There is a way to " fudge " things though. Use variable head bolt torquing. Get some brand new " Turbo " head bolts and torque in normal spiral sequence. But on the final tightening, torque the head bolts near the center 5 ft lbs more than the outer head bolts. Use the Turbo head bolts as they will take the extra torque.

 

That should give you an even compression of the head gasket, or " more even " than if all of the head bolts were torqued to the same amount.

Good point on balancing torques, for even pressure, down the length of the head.  Measuring the amount of torque required, with a torque wrench and working down to it, on the center bolts to get to "flat" could give a number to be added to the final torques.  And turbo head bolts to allow the increase without overstretching a bolt.

 

Some day I'll probably try it.  The main reason I put much effort in to looking at it was because the engine ran fine, and the head gasket showed no signs of leakage, at the cylinders, or at the various fluid passages.  Logic says it worked before, why mess with it.  I can see planing a blown head gasket head, to be sure you have good sealing surfaces, but this one would seem a waste.  

[rcb280z]

And you have the time to do the above test.

 

One question though, hopefully its not a dumb one, when the head is warped that much doesn't it affect the cam? Wont the cam also be warped?

It seems logical to me that the cam would be or is it that the cam towers are affected?

Edited October 18, 2015 by rcb280z

[Zed Head]

I like Chickenman's idea. I think it would work fine. I would check the torque a couple of times after it has a few miles on it. What head gasket were you thinking of using? I think I'd go with the compressible one.

I'd go with an OEM style with compressible steel fire rings.  Nothing fancy. You can spend some time looking at heads and head gaskets and how they work.  They're really three gaskets combined in one assembly.

[Zed Head]

And you have the time to do the above test.

 

When the head is warped that much doesn't it affect the cam? Wont the cam also be warped?

That's the interesting part of the whole question.  When the head was bolted on at the factory everything was flat and in-spec. and turned freely, unbent.  But when the head is unbolted from the engine, that's when it will warp, due to internal stresses.  It certainly wasn't warped sitting on top of the engine, unless the deck was warped also.

 

Then it's planed/milled to flatness, but the cam tower surfaces are still warped.  That's the factory way.  The second way is to mill top and bottom, and remount the cam and towers.  But then your valve guides and valves are slightly akimbo.  Seems like to do it "right" you have to remachine every surface, including valve guides and seats.  And it f you're not going to do it right, you might as well just bend it back to where it was when it left the factory, using calculated torque values.

 

The whole milling/planing/flattening head thing has always bothered me.  You always end up with something that's not quite right.

[Zed Head]

I may try the test.  I ran in to one problem with a head bolt though, it wouldn't thread correctly (the one opposite it was fine).  I had already reinstalled the sprocket and chain so didn't want to take it off again.  But winter is coming and I will get bored so it will probably happen.  I have the starting numbers so why not.

 

I'm not sure that aluminum sitting in its elastic region on the stress-strain curve will relax to flat with no heat-cycling though.  It would be odd if it did.

[rcb280z]

That's the interesting part of the whole question.  When the head was bolted on at the factory everything was flat and in-spec. and turned freely, unbent.  But when the head is unbolted from the engine, that's when it will warp, due to internal stresses.  It certainly wasn't warped sitting on top of the engine, unless the deck was warped also.

 

Then it's planed/milled to flatness, but the cam tower surfaces are still warped.  That's the factory way.  The second way is to mill top and bottom, and remount the cam and towers.  But then your valve guides and valves are slightly akimbo.  Seems like to do it "right" you have to remachine every surface, including valve guides and seats.  And it f you're not going to do it right, you might as well just bend it back to where it was when it left the factory, using calculated torque values.

 

The whole milling/planing/flattening head thing has always bothered me.  You always end up with something that's not quite righ

Ok, that makes sense. All along I thought all you had to do was mil the head, HG surface, and if the cam was warped have it straightend. I mean within reason of course.

Edited October 18, 2015 by rcb280z

[Patcon]

My question for Zed Head is this, with the warped head laying on its side will the cam still turn with out binding. I suspect it will. I would think the combined bearing tolerances from the middle tower and one end or the other easily exceeds the .007 warped. Plus when its bolted down most of that warpage goes away. So even if you just shave the bottom of the head to dress it flat the cam would still operate properly once it was bolted down

[Zed Head]

That makes sense.  In sum, the essence of my question is "what's the difference between a head that's bent down to flatness using head bolts and a head that's milled to a flat surface".   The answer I get is "not much".  Especially when you look at how each cylinder has its own fire ring.

 

Full disclosure - I work in R&D, where questioning the status quo is how you justify your job.  Of course, better and cheaper is the end goal.  Part of cheaper is no wasted effort though.

 

I could be missing something.  Open to criticism/discussion.

 

 

 

Edited for too many "thoughs".

Edited October 18, 2015 by Zed Head