Canfield model engine

[Tom Jamboretz]

      I'm making the Canfield model engine and have run in to another problem.

The fuel check valve does not hold the fuel back in the tank as it should. Has any one had this problem before and what is the solution. The check ball is 5/32" and does not weight enough.  Thanks far any help.

[cnr6400]

Is there room above the ball to add a small dead weight to help it seat? Sometimes the vibration on gas engines does not allow lightweight balls to stay on their seats.

When doing boiler check valves I usually place a brass bar on the (stainless steel) ball in the seat and give the brass bar a sharp whack with a light hammer. This seems to help the ball form a closer fitting seat.

Hope these ideas are of some use. 

[Dave Otto]

Hi Tom

Could you possibly add a light spring over the ball?

This is not much help for your problem, but not wanting to use the blocky chunk of brass mixer that Debolt designed. I chose to machine and install a Lunkenheimer mixer from Morrison & Marvin.
I think it looks much better and performs great, also I have never had any issues with it leaking fuel.

Dave

[Roger B]

There are quite a lot of important design details to get a reliable ball valve.

-The hole should be ~0.7 times the ball diameter.

-The ball lift should be constrained to ~0.2 times the ball diameter.

-For flat seat valves the bore of the hole should be smooth and round, at least reamed. The seat should be smooth, flat and perpendicular to the bore.

-For cone seatings the cone must be truely round and smooth. A drilled seating will not be reliable.

-The seating can be formed by putting a ball in place and giving it a tap via a suitable sized brass rod. Throw this ball away and use a new one for the valve.

[Admiral_dk]

Very nice to see some numbers Rodger    but one dimension is missing - the diameter of the bore the ball is moving inside ...?
I guess that there is some leeway here, bit too small and you have a flow problem - too big and the ball might not 'come down' well / fast enough .... 

[Roger B]

Thank you Per, I think the key here is the very restricted lift. If the ball rises 0.2d or less it is restrained in place by the orifice. I do have some documentation on this saved somewhere which I will look out (Tom, I hope we don't hijack your thread too much  )