I built a 100mm x 66mm type 4 engine. I had lapped in the cylinders to the heads and yet 3 out of the four leaked. I have pulled the engine apart again. Now, how do I get these heads to seal?

See my response on the Pelican board.

--DD

Hmmmmm........very possibly the problem is at the cylinder base, called the case register, or perhaps unequal thickness or burrs on the cylinder base shims. Check the pattern of the leakage to deduce if there's a slight difference in cylinder height causing the problem--could possibly correct it with additional very thin shims on the base. Be very wary of the shim packs -- check them carefully for stamping burrs--I had to smooth mine on a grinding wheel. Even so, one pack of four had three similar ones and one that was .004 thicker! Had to order another pack and then mic all of them to get a matching set of 4.

If I remember there is a problem with cast iron big bores not sealing because they distort.

This is a nagging question for me. I know the prevailing advice is to forgo head gaskets and do a direct cyl to head. My heads were done by a very reputible shop, and my cyls are also quality (96mm), but I'm not able to visualize how this will work and not leak without a gasket. I dont necessarily like the aluminum that comes with the Reinz kit, and am suprised that a thin copper gasket isnt used.

I'm not arguing for this, just trying to understand how this actually works. This is my first VW/Porsche aircooled engine, but I have rebuilt many a Yamaha aircooled engine, and typical water cooled Motors, Gm , Saab, BMW. Maybe I am just old school, but this freaks me out a bit, and dont want to have to redo....

My cylinders are birals, the heads have been resurfaced. I will check the deck height to see if we have a variation.

show us a photo of the heads. was the leak between the cylinders?

Birals? The AA Chinese ones?

If it won't seal up you have at least one of these variables, or maybe more than one:
-Case decks are not flat, 2.0 cases are notorious for this, as they have unsupported cases under the registers.

-Cylinder head decks are not flat.

-The cylinder sealing surfaces are not flat, or were machined irregularly when compared to the case seating surface.

-The top fin of the cylinders is contacting the cylinder head first, before the seating surfaces make positive contact.

-deck heights uneven, as previously stated. This could be from the cylinders having unequal seating lengths. All should measure 3.600" or 3.640", these two are the industry standards for case deck to cylinder head seating surface.

Chances are you have more of these going on than just one, the first thing I'd suspect are the cylinders.

The heads had been machined by Hoffman. I will check the cylinders and the registers. More to come.

Thanks

Boy, it would be great if you had access to a coordinate measuring machine to measure the head sealing surfaces to confirm that they are not only parallel but IN THE SAME PLANE WITHIN .001". With careful setup this can be done on a surface plate as well. Same thing with the case register surfaces. they have to be in the same plane. Cylinder heights for the pair on each side must be the same length, of course the other pair can be slightly different.
If your machine shop can't verify their work, go do it again.

Knowing what I know, I'd be concerned with the concentricity and plane of the cylinder deck surfaces more than anything.

I made this exact tool, but to check for cylinder/deck height differences and sold it for a bit. I sold maybe 40 of them before interest dried up.

-Aaron

I am going to vote the top fin is hitting the head, not enough room to seal properly.

That was on my list of potential issues. I agree that its a great probability.

I know....just reminded me of a time........

Just seconded your suggestion.

I do have a surface table and height gauge so I can check the spigots as a function of the case centerline. I hope to have it fully apart this weekend.

I checked the cylinder deck prior to removing the cylinders and they are all in the same plane. The block had been decked when the spigots were machined for the bigger jugs. I think my problem stems from replacing the head nuts with serrated nuts. I think the added drag on the nuts caused the tightening torque to be reached prior to proper clamping. I am going to go back to washers and nuts.

BTW I am going to a Web Cam 494. So I want to go with lighter valve springs. Can you use T1 springs in a T4. T4 springs just aren't that available.

If you go look at European Motorwerks he did an interesting modification using I think 356 lifters, much smaller, with machined fittings in the case.
You could send the case out for this modification to them, it removes a decent amount of weight.

Rich