Problem in design of a piston and cylinder for elevated temperature

[vahidp60]

Hi friends,
I have a cylinder with inner dia 40mm, and outer 100mm. A piston has to slide in the cylinder to inject a molten metal in 600 centigrade degree. Both of the parts are made from H13 tool steel. I was considering about using H7/g6 tolerance between two parts. But the problem is that when these parts bring into those high temperatures they stick each other, preventing the piston from slinding.
I analyzed the two parts using ABAQUS (FEA software). Since the material used for the parts are the same (both H13 tool steel), the analysis shows no or a little change in tolerances between two parts in elevated temperatures. But in practice the problem remains, and the parts stick to each other. I was wondering if the material looses its hardness in those temperature that causes a severe wear in the first stages of sliding, and then the parts stick.
I am really confused.
More discussion about this problem can be found here: thread1103-250208

 

[btrueblood]

If both cylinder and piston are at the same temperatures, then they should expand the same amount, and a slip fit at room temp. will stil be a slip fit at high temperature.

The problem is, in practice, it's unlikely you have the same temperature across the piston face and throughout the cylinder. I.e., the i.d. of the cylinder may be at 600 C, but the o.d. of the cylinder may be quite a bit less (due to heat transfer to its surroundings), and the o.d. is restraining the cylinder i.d. from fully expanding. Also, the piston face likely has less of a heat transfer path, thus gets hotter than adjacent cylinder wall.

In any case, having knowledge of the actual cylinder wall temperatures, and if possible the piston face temp's., will allow you to calculate the actual expansion of each component in service.

Lacking that knowledge, you will have to just cut and try. I.e., machine/grind a bit off the piston o.d. and see if it works.

A last wild idea - ever heard of piston rings?

 

[israelkk]

Please do not double post.

 

[vahidp60]

Dear Israelkk, in the previous post Mr. KENAT suggested me posting my question here in order to find larger audience of mech engineers. That is why I reasked my question here!

Dear btrueblood,
The condition that you described is the case where the heat flows from the inner side to the outer face. But in our situation the heat is generated using band heaters that surround the outer dia of the cylinder, i.e. heat flow is from o.d to i.d.
In this case the cylinder expands more while the piston du to its lower temperature expands less. And this should encrease the tolerance. But in practice the problem remains.

 

[btrueblood]

Ok, so you have the reverse. The point I wanted to make is: what temperature is the piston at, and what temp. is the cylinder at, when the piston is stroked? If you don't know, then you can't calculate what the tolerances should be to keep a sliding fit in operation.

How long is the piston, and how well is it guided relative to the bore? Could it be cocking and binding?

What alloy are you melting (sounds like it might be Zamak? Aluminum?) and are you getting any oxide buildup that could be causing trouble?

 

[KENAT]

israelkk, I did indeed suggest vahidp60 post here as I thought that some members that don't frequent GD&T much might have useful input. While I realize double posting is generally discouraged I thought this an exception and vahidp60 did put a link to the other post.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

http://eng-tips.com/market.cfm?

What is Engineering anyway: faq1088-1484

 

[gwolf2]

>Lacking that knowledge, you will have to just cut and try.

I've done a lot of high temp conductivity, expansion, and fit stuff. This one smells a bit tricky because you can't get a good thermal measurement or calculation on the hot side because its molten metal.

It's probably a cut and try job as btrueblood says.

 

[DesignerGuy16]

Have you thought about adding heaters to the cylinder to bring it to operating temperature more evenly? A couple rod heats or a coil heater would do the job to even out the heat your seeing, and maybe help with your binding problem.

I used to design aluminum extrusion dies, and in that industry they would heat the dies and backup pieces to temperature before every running aluminum through them. Just a thought.

James Spisich
Design Engineer, CSWP