48 Pan Check Valve Seating Problem

[harleygreg]

Hi Guys

My ’48 Pan has been idle for some months and had some pints of oil in the crankcase, which I drained.

I have the proper spring installed in the oil pump check valve (1 x 8/32” long). It seems the ball bearing is not seating, and oil appears to drip from the crankcase drain hole every 5 seconds.

Is there a way to seat the ball, without taking the oil pump off the motor? I’ve tried tapping it with a bolt and mallet but with no discernible result.

Thanks

Greg

[sswaney]

I did a seach of the Forum using "check ball". There's a few threads available. Then one under Panheads seem to be the most informative.

[T. Cotten]

I did a seach of the Forum using "check ball". There's a few threads available. Then one under Panheads seem to be the most informative.

A link would be nice, SSwaney!

Here's one: https://forum.antiquemotorcycle.org/...ting-ball-seat

....Cotten

[harleygreg]

MMMMMnnn! - interesting, I'll have to think about this one then?

[T. Cotten]

MMMMMnnn! - interesting, I'll have to think about this one then?

Do you have a lathe and a diamond dresser, Greg?

....Cotten

[sswaney]

https://forum.antiquemotorcycle.org/...mp-check-valve

[harleygreg]

Thanks Cotten and sswaney, I need to check the flow rate as it seems to have reduced significantly, and check if there is any flow through the pump shaft by sealing the ball bearing hole. I've got a couple of ideas and will come back in a week or so, thanks

Greg

[droptopford]

Let it puke. Add oil as needed.

[T. Cotten]

Thanks Cotten and sswaney, I need to check the flow rate as it seems to have reduced significantly, and check if there is any flow through the pump shaft by sealing the ball bearing hole. I've got a couple of ideas and will come back in a week or so, thanks

Greg

Any luck, Greg?

Sidevalves are notorious for leaking through the adjacent cam cover bushing, so I'm curious what you set up to study an OHV.

In an earlier life before digital cameras,,,

Pumpseal.jpg

...I installed a late-model seal in a damaged cast iron body.

....Cotten.

[harleygreg]

Hi Cotten

I'm losing just over 1/4 of a pint per week, with no use, to the crankcase. This was measured after warming the bike up for fifteen minutes, and then leaving it to stand for week in cold weather, and then draining the crankcase.

Having measured the oil, I intended to remove the ball, and seal the hole, to see whether the leakage was through the ball seal or maybe through the oil pump. The seal didn't work. Given that all the forum threads seem to refer to the ball seal and not the pump, I'm assuming that that is where the problem is?

Also 1/4 pint per week would be the maximum, as if the bike were used every day, the leakage would be less, and probably negligible - do you think the 1/4 pint is about right, or would you consider that excessive?

The solution could be an inline valve if the bike is not used that much. I could only find the following, as the one from 'feked' is too industrial, and the fittings are for flexible hose, though I'd need a chromed one.
https://www.vintagecarparts.co.uk/pr...ng-in-line-tap
That could be fitted on the pipe from the tank to the pump.

The only other solution I thought of is pictured. I spoke to the guys at Henkel adhesives, and they reckon Loctite 648 would do the trick, if the washer was brass or 9488 if fibre, both would be high temp adhesive/sealant, impervious to oil etc. Maybe a bench test with your setup would prove it? I've used 648 to help seal the intake nipples with good results.

What do you think?oil loss 1.jpg

Greg

[T. Cotten]

You are being a glutton for punishment, Greg!

If you are going to attempt to dress a new seat insert, why not just dress the original seat as I have described?

And eliminate all the variables of a self-destructing washer, adhesive goober, spring tension, etc.

....Cotten
PS: The original seat never had a wide angle on it like your drawing until it was badly worn... Think how a fiber washer will wear compared to cast iron or steel.

[harleygreg]

OK, OK, I'll have to take it apart!

Thanks

Greg

[T. Cotten]

Might as well get your tooling together first, Greg!

Can you diamond-dress a common die-grinder stone on your lathe? (That's often forbidden on a pristine piece of equipment...)

It must not only have a pilot, and the end dressed 'flat', but the OD must fit into the "well"

CHEKGRND.jpg
As the drawing shows, the worn seat must be removed, and the pilot has only a slim chance of concentricity, as the Factory threads are often not.
Thus the end of the stone must be flat, not conical or otherwise. Please note that if the ball rests upon the mill-profile of a stone-ground surface, then there will always be gaps larger than an oil molecule, as big as they are. So the new seat should start out as narrow as possible, like motor valves that 'beat themselves in' to burnish the final finish. In this case, there is no need for any width at all to dissipate heat.

As I posted previously, a freshly-ground sharp edge produced the best results with my bubble apparatus, at less than 4psi. Any further attempts at anything only made it worse. I can only presume that the Factory produced a thin edge of contact to begin with: a simple hole, and the ball itself did the rest.

I invite any and everybody to come up with a better approach. (Here are a few of my failures, with the exception of the flat stone on the right),..

CHEKfails.jpg

..But please bench-test it.

....Cotten

[BigLakeBob]

In my opinion it does not matter if the angle is an inch long, a round ball will only hit one tangent point on that straight plane (meaning the angle line whether it's 1/2 degree or 89 1/2 degrees or any where in between for example). The largest point on a diameter of any given roll or ball can only touch one point when contacting a flat plane. Period, simple geometry. An angled cutting then lapping tool should work.

I originally did the Easyrider's tech tip with welding a ball to a welding rod, hand ground several recesses for the lapping compound to go in. Used it many times until I lapped a groove in it. I then made a one piece tool and ground a 3/16 R on the end and installed grooves again. My bikes don't leak, at least nothing noticeable, this method has always worked for me. I do it on the bike, you don't need much compound, then just clean with a q-tip, no big deal. I plan to make some tools out of some 3/8 dia brass balls but haven't got to it yet. Bob

[BigLakeBob]

I was just thinking, just because there is a new angle put on, maybe the angle is not round, this would make the ball not seal. Hard to do by hand. Could get 'wollard'. Again, I've had real good luck lapping by hand with a fine compound, 400 to 600. My 2 cents, good luck.