Aircraft cylinder construction

[rmeyers]

Here's a question that I thought that I knew the answer to, but didn't

How is the cylinder pressure held in the cylinder? In other words, what seals the cylinder head to the cylinder?

I know of course that the head is screwed to the cylinder barrel, but what forms the seal?

1. Is the top of the barrel, what would be the head gasket surface on an automotive engine, screwed hard against a matching 'deck surface' machined in the head?
No, that's not it. That is what I always kind of assumed was happening but after examining several cylinders, and a few Google Images, you can see that the barrel definitely does not seal that way.

2. Do the threads themselves, possibly with some kind of sealer, provide the seal?

3. Does the outer part of the head screw down and form a seal against the top barrel fin?

4. On the outside of the barrel, above the threads, the cylinder is smooth for maybe a half of an inch from the top of the cylinder to the top of the threads. Does this area perhaps form an interference fit in the AL head and so provide the needed seal?

5. Something I've missed completely?

Number 4 is getting my vote as of now, but I'm ready to wrong again!

Randy

[Walt]

http://eci.aero/pdf/12-2.pdf

[DanH]

Excellent document Walt...thanks.

[Brantel]

The charts on page 3 are interesting!

[TX7A]

...also found on the ECi website.

http://www.eci.aero/pdf/SeparationAn...essnaPilot.pdf

[BillL]

I have several Conti heads cut in half being used for bookends. The entire outer barrel/head interface is coated with cooked fuel. I don't think they seal perfectly, all the time. The threads all appear to have yielded in the assembly process.

I watched the assembly a few times. When they are assembled the head comes from an oven at 400F and the barrels are fresh machined and not coated. A heavy pair of handles are attached to the base of the barrel and it is used to suspend the threaded end in liquid nitrogen. After a short while and the boiling is low, the guy picks up the barrel starts the thread, and gives it a good spin and in a few revolutions it thunks home. No RPM, no torque, just the honed skill of the assembler that made this work. I always wondered how that was going to workout in the long run.

The barrel base flange was drilled and finished after attachment to the head.

Another separation failure mode is present too. If the barrels are too thin and the piston has too much ovality, the barrel below the head will deform due to side loads and then as the piston passes the ridge in the barrel reinforced by the head, it bends the barrel steel it separates. This occurs more on highly boosted engines.

Sad state of affairs for the poor owners of these recalled Conti engines.