Stirling engines - Ball bearings

From what I've read, shiny wire does a better job of being a bearing on a tin can Stirling engine, but I want to see just how much extra friction a proper ball bearing race or two would add to the mix.

The friction levels are very low on these little engines.

They have to be or they just don't work.

But I want to try to make a more robust version, and perhaps even make one that is capable of doing some work.

Perhaps.

But probably not.

Any attempt to make a bigger tin can Stirling engine would also involve a lot more weight. And more weight means more wear and friction. Ball bearings will be the solution, as long as there is enough power left over to overcome the extra friction that bearings have.

That sounds a little contradictory, but bearings are really good at dealing with extra load on the rotating surfaces, but they involve a little extra friction than say... hovering in space.

So, with this in mind, I looked into small bearings.

They cost a lot.

A 12mm (internal) ball bearing race is a very standard thing that industry makes. They cost around $2 each retail, and are a weighty, very strong thing you might find in a motor bike, or in the centre of a bicycle wheel. They carry a lot of weight, and last for ever. But they have way too much friction.

The little bearings I want that are only around 1.5mm in diameter (internal) all seem to cost around $20 each. They are nothing like the robust 12mm ball bearing races I looked at.

Tiny.

Fragile. (actually not really, but compared to the 12mm version...)

Fragile.

And expensive.

But my fishing real has a dozen of the things. They cant really cost that much. It must just be that there is no real retail trade in them. I need three or four, so I thought of buying a $20 K-mart fishing reel, and pulling it to bits, and that's probably what I'll end up doing. But in the mean time, I really want to know if a bearing will be too costly in terms of friction.

I found a little computer cooling fan in my electronics junk drawer. I figured that should have at least one bearing.

These little fans don't put a lot of stress on their little bearings, but they last for ever and spin really fast, with little friction.

Last for ever, fast.

Perfect.

I started by removing all the bits that didn't look like a bearing.

How hard could it be.

Very, it turns out.

That's the little bearing inside the small brass tube.




I spent a lot of time and energy trying to get it out.

I started by trying to knock out the pin by gently placing a centre punch (made of thick wire) on the centre axle, and smacking it with a hammer as hard as I could.

That didn't work so well.

Actually that didn't work at all.

I finally got it to give some ground by taking a hacksaw to it, and cutting through all the hard plastic surrounds that held the coils to the little motor.

This left me with a much more manageable bit of kit, that even looked like it might finally surrender it's bearings.

In fact, this would be perfect if I could just get the pin out of the centre, so I could put the Stirling engine's wire crank through the centre.

Centre.

Difficult

I put one end in over the opening of a little socket so the punch could get through and hit it hard.

Nothing happened, so I turned it over and hit it again.

That's my trusty hammer at the top of the frame.

Camera shy hammer.

This time it worked.

There's a little flange on one end of the pin that was making it impossible to tap out.

An amazingly strong little flange.

I hit it hard.

A lot.



Now that the shaft was out, I needed to knock out the bearings. I tried inverting the socket so it's outside fit inside the brass sleeve, and rested against the bearing. I hit it again.

And again.

And some more.

In the end I just kept cutting.

The hacksaw was the only thing making any progress.

Who would think there would be a time where a hammer failed me.

There's no real danger of damaging the little bearings here, as the brass sleeve is soft and bearings are made of insanely tough... stuff.

The bearings came out with ease, once the brass was cut through.

It turns out that the brass sleeve is really a brass sleeve with a divider in the middle.

No amount of hammering was ever going to get the bearings out.





Oddly, after all that hammering, the bearings still roll reasonably.

Reasonably.

The reason the little fan was part of my electronics junk drawer in the first place, was because it no longer turned. I think it was the cooling fan that I replaced on my rectifier, so it was never going to work perfectly.

The bearings spun freely enough after rotating them for a while with my drill. but there was a little bump in each revolution.

This pic is of the new ball bearing raced tin can Stirling engine running on my stove-top at around 200rpm.






200rpm is around the same speed on the same heat setting as it ran before, but it no longer runs from the heat of a single candle at only 36rpm, so the bearings have increased friction a little.

The brass sleeve was replaced with a cable tie for each bearing, and it turns out that cable ties fit nicely into my adjustable screw in bearing mounts.

Having the adjustable wire crankshaft I made turned out to be worthwhile, as I would never have been able to make just this one change. The smallest change in the crank shaft, in either the displacer crank, or the power piston crank makes a lot of difference as to how well my little Stirling engine runs, so it would be very hard to tell if it didn't work because of the bearings, or because of the different crank I would have been forced to make.

So.

A success as far as this little engine up goes. The bearings are small, but should be more reliable than the a plain wire on wire bearing. And it seems the friction loss is only around a quarter of a candle.

An interesting addition to this learning thing.





120 Things in 20 years measures the amount of friction in a tin can Stirling engine in "candles". I think I just invented a new metric.