The motor on my tower has been working fine before with no problems but after I used it for a bit it has started stalling and won’t turn and I have no idea why. I have another motor for the indexer that still works perfectly fine. My best guess is it has something to do with all of the friction on that axel because I try to turn it by hand but it doesn’t turn easily while the other motor does. Could it also have something to do with the fact that the non working motor doesn’t have a metal insert while the working one does?
Heres a video of my problem. I tried to get all angles so you could see whats going on. The axel is connected to a big sprocket which is connected to a smaller one that turns the flywheel, but if I disconnect the chain the axel still does not spin well which makes me believe the connection to the flywheel is not the problem
This looks like a clear friction issue to me. Either the bearings are not screwed in properly, the axle is bent, or (it could just be parallax error) but the axle doesn’t appear to be in the right holes across from each other.
This could all be mitigated with more bracing and proper alignment. I suggest you disengage the axle from the motor and make sure the motor works properly without a load. Disconnect the chain from the sprocket as well. If the roller can’t freespin for several seconds, it is definitely a friction issue. But correlation does not necessarily equal causation, so the root of the problem could potentioally cause other issues despite you solving a symptom of the problem.
For this reason, you should use the 5-why method
This problem-solving technique was recommended to me by a retired engineer. It’s gone a long way at troubleshooting pretty much anything (even outside of robotics).
Thanks, if I remember correctly this rod was a little bent before I used it so that could definitely be the problem. Also what does free spin mean? Because all the bearings I have on the axels really prevent it from moving much. Here is a video of my flywheel spinning by hand, its not spinning long at all and I’d like to know if that is also a friction issue. Also thanks for the tip about the 5 whys, I’ll definitely be incorporating that into my engineering process!
Alright, so this is looking more and more like a friction issue. Typically you only want 2 contact points on anything that rotates. The four bearings you have are just redundancy points. You should only use two of them at most. I’m gonna do a bit of explaining for this, but the solution will be at the bottom of the post.
Explanation
I saw in your reveal that you bent the c-channel on the chassis to fit bearings and collars. There are a few workarounds for that. I typically would avoid ANY form of bending because it morphs different parts of the metal and changes the material chemistry. Aluminum and steel fatigue over time with bending.
So, I cut out notches in the c-channel, but you could also sand-down the bearings and they fit perfectly. Vex produces bearings so that they can only fit in the middle holes of the c-channel or the exterior of the c-channel. Bearings physically do not fit properly where the c-channel “lips” are (aka flanges). The holes right next to these flanges are not meant for *unmodified bearings It can be misleading because it looks like it fits, but trust me it does not. If you pay close attention to the bearing flats, there are small divots that protrude out the back end. These are supposed to snap into the c-channel holes to ensure proper alignment.
Here’s how I make sure a bearing is properly inline. I screw it in as best as possible with the divots. Then I put in an axle and spin the axle in the open hole with my bare hand. If there is any resistance, I loosen the screws and readjust. Do this for every bearing. This is simply a manufacturing flaw that you have to account for when building. There are new bearings that self-align better than the current ones, but it seems you don’t have access to those.
As for your particular robot, get rid of the “interior” bearings. Those are the bearings on the towers that are tucked next to a c-channel flange. Leave the ones that are on the outside part of the c-channel and make sure to align them properly with the process I explained above or something similar to it.
Freespin (in the context of vex): when you manually rotate some object on your robot and allow it to spin without assistance. Anything that rotates on the robot should be able to do so with minimal resistance and minimal force. The force from your hand should be enough to allow it to spin freely with its own momentum. I don’t know how else to explain it without video.
The bearing seems to fit fine on the side of the c channel when I am testing it. Is there a small difference that is unnoticeable until more bearings are used?
Oh yeah normal bearings don’t fit on the insides of c channels. You can shave the side down a bit which is what we do or try and put the bearing on the other side of the c channel. The hole doesn’t exactly line up with the bearing when they are attached on the inside of c channels like that.
so the reason it can free spin with one bearing like that is because the shaft is allowed a lot of slop and wiggle room. if you have two bearings like that, the shaft isn’t able to wiggle and it will make it not be able to free spin.
also check the straightness of your axle, though looking at your issue it’s almost entirely caused by bearing problems.
if you want to have the bearing on those holes, put it on the outside of the c channel, where it does fit, not on the inside.
Is that steel? The gap between steel flanges is slightly larger than aluminum for whatever reason. But yes, the example you showed, there is no real difference. I was under the impression that you were using aluminum. Either way, you should only have two contact points.
Also, just something to note, you shouldn’t test the axle freespin with a gear or anything with a large radius for that matter. The force you apply might be minimal, but it’s multiplied by a factor of the magnitude of the radius. Basically, you can’t get a good measure of the friction with that set up. So just use the bare axle.
Edit: the freespin example you showed is a good example of a freespin. However, you must do it with two bearings across the width of the towers. The goal should be to get those rollers on the tower to spin as freely as the example in the video.
and like @mvas said, your best fix will be to remove all but the two outer bearings. you should only have 2 bearings on each axle, especially on a light load, low friction axle like on a roller. any more bearings just cause extra friction.
another thing you can do to decrease the friction on bearings slightly is to drill out the bearing hole to just a tiny bit larger. it will increase your slop as well as decrease your friction, but slop is better than friction on rollers like that imo. with perfect build quality, you shouldn’t need to do this, but it can help.
I wanted to make a few clarifications to a couple of statements I made (so not to confuse anyone). You should test freespin with a wheel or a gear because that will allow you to see the overall resistance in the system in comparison to momentum.
But to get an accurate gauge of the actual friction acting on the axle, use only the axle to get a feel for the amount of resistance. It should be negligible in both cases.
Unless I’m mistaken(I’m on mb), you seem to be transferring power through the 12" shaft. It would be better to avoid this as it could twist the shaft and warp the frame(which may be the cause of your friction)
While this is true, there is hardly a load on this system. Additionally, there is no other way to transfer power with Vex parts for rubber band rollers. You could do something bulky or impractical with screw joints, but I wouldn’t recommend that for this particular case.
So yes, this statement is true:
But only because the shaft is experiencing a high load due to friction. Once that issue is resolved, it shouldn’t matter.
I’d like to understand this problem. Are you saying that I shouldn’t connect 12’’ shafts to the motor because those are likely to twist? If so, how could I avoid this problem? I know that I probably won’t have to worry about this because I don’t have much of a load, but I’d still like to know what I don’t have to worry about if that makes sense lol
I was considering this, but it’s not ideal for intaking. The main benefit of using rubber bands is their compliancy which could minimze the amount of space the subsystems take up (since the ball could “dig” into the rubber bands). This is impossible with standoffs. There is also contact loss from the irregular shape of the standoffs as opposed to circular rubber bands.
TL;DR space efficiency and energy transfer decrease with standoffs. The reason it was done in TP was that the front intake had to span the width of the robot and the low strength axle needed to be reinforced. Again, this could be completely eliminated with a high strength axle.
