2 Motors on One Axle?

I know this topic has been brought up before, but I specifically want to know if there is a major benefit to using two motors on one axle as opposed to putting an idler and another motor driven axle underneath. I know there is a risk of “candycaning” the axle if the motors run in opposite directions, so what makes it worth it? Finally, do you just estimate the axle length that will fit into both motors and cut it?

It’s better because it reduces weight and friction. Fewer parts and less moving things. It’s pretty difficult to ‘candycane’ a shaft like this because the motors don’t have enough torque to really get a good twist going. To cut it you just estimate the length.

The 4 we have done, when we had to cut it we test fit it and trimmed one end of the axle with a Dremel until it fit perfect. But I think we managed to get a 2" axle to fit between 2 motors with shaft couplers, on support columns spaced 1.5" apart without cutting the 2" axle. (If I’m remembering right.)

I actually disagree: if the motors are running at full power in opposite directions, the axle doesn’t stand a chance from my experience.

To cut it, I would recommend putting the axle on with 1 motor as normal first, then sliding the other motor onto the other end of the axle and measuring the distance between where the motor screw threads are supposed to be and where they are while hanging from the axle. That’s how much you would need to cut off.

Huh. I’ve never actually done this so I wouldn’t really know. Remember that if you put 2 motors on one shaft the motors need to be reversed.

I usually don’t mess with the axle length, I’ll just space one motor back a little bit if the axle is too long. :stuck_out_tongue:

I would say that’s the best approach to putting 2 motors on 1 axle.

I like to avoid cutting anything whenever reasonably possible to preserve whole pieces for future years. The fact that our sister team cuts things way too often contributes to this mentality.

Yep, that’s why we haven’t had to order metal all year. We have had to order more axles, though, given that we have 3 teams again this season.

Cutting shafts for this length is sustainable though since you can probably bet that you need shafts of this length to put motors on both ends in the future.

Yes and no…
Generally, with a well-built robot, you wouldn’t necessarily need to put 2 motors on an axle.
Though, if 2 motors on a single axle is a common practice in your organization/team, I can see how it would be relatively sustainable (assuming people are willing to look for the right length before cutting another axle).

I have never seen 2 393 motors on an axle running against each other twist the axle. I have seen it cause a burn out and trip the PTC. So long as you program it correctly, I see no disadvantage of two motors on one axle vs two axles with each having its own motor. Every time you add an axle, you are also adding friction. Every time you add an axle, you are adding something else that can fail (axles bend under stress).

Whenever you want two motors turning something it is better to have them both on the same axle if that can be accomplished.

Calculating the length is not too difficult. #Brian gave a real easy way to do it. There are other ways to calculate it as well, but Brian’s method is probably as easy and accurate as any.

The axle penetrates the motor housing approximately to the seam between the sides of the motor and the screw plate. You can mock up your assembly, and measure between those two seams.

We are coming up with a list of off-season experiments. One of the experiments we are going to do is find out the optimal way to combine motors. Do you get more torque by putting them on one axle or using a idler gear? Do you get more torque if you mechanically couple the motors or purposefully not mechanically couple the motors? What we are trying to get at is: what is the impact of slightly varying motor speeds when mechanically coupling them? Does one take the brunt of the work while the other doesn’t add anything? We seem to have observed that this could be the case - when we add a second motor, we don’t “apparently” get twice the torque when mechanically coupling them.

But, like I said, this is an off-season experiment that I want to do very methodically and not depend on just powers of observation.

In the spirit of learning, I would encourage other teams to set up similar simple experiments to test hypotheses. Internet strangers are one thing - and this forum probably has slightly more robotically educated internet strangers - but how do you validate an internet strager’s response unless you try it out for yourself?!

Looking at this from a different perspective I think that teams should also evaluate how easy it is to test and repair/replace motors on a single axle.

  • Is it easy to test each motor?
  • Is it easy to replace a motor between matches?
  • Should you have spare shafts of that length if one is damaged?

We had teams in NBN struggle with motors paired on shafts with flywheels in between. It can be hard to get to the screws and really difficult if one is stripped.

Seems like I end up doing this a lot and found that the axle basically inserts into the motor to the seam. If you put you axle across both motors and draw a line just inside the seams it works pretty well once you round off the ends of the cut side.

Replace the screws that hold the motor together with a zip tie. The motor housing itself has nice flanges that keep everything aligned anyways. With this you can very easily change out gears and shafts without pulling off a motor entirely.

We’ve seen best results with two motors on one axle; the fit of the idler gear can cause significant friction loss. (In particular, our experience is that it is a bad idea to gang motors with metal pinions running against identical metal idlers. Way too much friction. Odd, because properly-shaped involute gearing should roll one against the other and have very low friction. In VEX gearing, there is less friction if one of the gears is plastic. But I digress.) One tricky problem with single-axle mounting is getting the correct axle length, however. If it is ever so slightly too long, the axle presses into the motor’s PTO (Power Take Off) coupler too far. This also can cause significant friction losses, and accounts for why some people don’t see gains they expect when coupled onto a single shaft. Also note that if the axle is slightly too short, it won’t press far enough into the PTO coupler. Then, the axle can eventually become rounded off on the end. It will appear to work until it’s under load. Then, the motor coupler will spin around the axle. After a while, the motor on the “spun” end of the axle won’t be contributing anything.

Theory says it won’t work that way, and that the work will be done relatively equally once friction and electrical losses from each of the individual motors is accounted for. I have to say, what we’ve seen in practice matches the theory. That is, both seem to contribute nearly equally. Think of it this way: For simplicity, assume both motors are being driven equally. That is, you’ve set them to the same “power” level. The more efficient motor (the one that “wants” to turn faster) will take up some load. And therefore slow down, since it’s not an infinitely powerful device. When it slows down, the speed of the system will drop down to a point where the less efficient motor begins to contribute. This allows the entire system to speed up. An equilibrium will be reached where each motor is contributing, though the more efficient motor will be contributing slightly more to the rotational work than the less efficient motor. The additional loss of the less efficient motor will equal the additional work done by the more efficient motor. But this should be a small number.

Sounds like fun. And a worthy goal.

After spending a few minutes figuring out who you are and what your background is, I’d believe you.

But that’s just me. And, overall, point taken.

[edited to clarify]