Why don't ball bearings conduct electricity?

[mmagliaro]

Jim,
Your test description is outstanding!
I was already figuring on doing something just like that (try them out in a mocked-up chassis and roll it on a slope).

I was just going to try rolling it down a sloped piece of track right onto the layout somewhere, and measuring
how far it rolls with sleeve vs ball bearings.  Your idea of jacking up the ramp and measuring how much slope each
bearing version needs to start rolling is also a good test.   The two methods don't exactly measure the same thing.
With your idea, the chassis does have to overcome whatever standing friction the two methods would have.
I am betting that neither is enough to care about when a motor pushes on that shaft through a gear.  Once the engine
overcomes its "sticky" and starts moving, we have the constant rolling friction... and that's another matter.   If that
goes down a lot with ball bearings, I could see how that would make the engine run more
uniformly at super low speed.   

I think I want to try the slope and see how far it rolls.

And besides, this will let me make a "scrap" chassis, so I'll be better at it before I make my real one.

////////////////////
As an aside... I completed driver #5 last night.   This is quite the brutal task, but I have only one more to go.

Speaking of building things twice:

I had #5 almost completely done except for a few spokes.  Then I made the bone-headed move
of trying to put a spoke in while the wheel was just taped down to the bench, but not on my brass alignment jig.
I had so many spokes in already, I wasn't worried about concentricity. 

What I didn't think about is that without being on the brass jig, the wheel is sitting in open air, and heat from the soldering
iron happily travels right through that wheel with great speed.  So great, in fact, that a second or two after I touched the hot
iron to it to solder in the new spoke, about 6 other joints immediately softened and the whole wheel moved, bent, and
pretty much fell apart.   

So... out came all the spokes and hub, back on the jig, built the whole thing over again.

Ironically, when I built it the first time, the run-out tested at .003".  That's pretty good, but it was the "worst"
one so far (the others are all about .002").  When I built it the second time, it came in at the amazing
figure of only  UNDER .001" runout.   In fact, the little Starrett gauge can measure under .001, and I'd say it's about
.0005" !!!  So it went from "worst" to "best".

Sometimes a disaster turns into a blessing.

[Lemosteam]

I see a ball bearing chassis with wheel wipers in your future...

Sorry @mmagliaro I could not resist (ance) .  See what I did there??

[nscaleSPF2]

Max,

Ok, I understand your plan to evaluate the friction of the mule chassis, but as a dyed-in-the-wool techno geek, I owe it to you to explain the limitations of your approach.

First, let's understand the definitions of break away friction and rolling friction.  Suppose that you and your friends are trying to push your son's hot rod (which sits on a level driveway) into the garage.  When you first start to push it, there is a large amount of resistance before it just starts to move.  This is called break away friction.  As you and your friends continue to push, the resistance goes down.  This is the rolling resistance of the car.

Our case, with the mule, is exactly the same.  There is going to be some initial resistance, which will go down when it starts to move down the track.  Maybe there will be a big difference between the break away and the rolling friction, maybe not, but the point is that they could be different.  You won't know until you measure them.  I can imagine a type of wheel wiper, for example, that requires significantly more effort to start moving, but less to keep rolling.

One of the issues with your proposed test method is that the results will include the effects of both break away and rolling friction, without a way to tell how much of each is present.  I know, and agree, that the break away friction will not be an issue, given the friction of the motor.  The point is that (I think) you are after an accurate measure of the rolling friction.  I don't believe that your method is going to tell you that.

I applaud your efforts, Max.  I just want you to obtain the best results you can, given the effort that it is going to take to make the mule chassis.

There.  I've said my piece.  I'm going to shut up now.