Hello, this is my first year and we have a competition in a few weeks. I have been having trouble with my flywheel design. We are using the 5" wheels and 2 393 motors I put the turbo gears into. I don’t know what gear ratio we currently have but I can post a picture. If you can, help me find out if the ratio works or doesn’t and maybe post some pictures of your ratio or how to build one that would work
This is also my first year as a mentor. We did not adjust the internal gear ratio on the 393 motors. We used the green plastic gears and went 60 to 12 (2x) with a compound gear to achieve a 25:1 ratio. I was told by an experienced student that the motor is 100 rpm by default. This seems to shoot far enough, but we are not accurate (yet).
Do you know how far it shoots?
It is not consistent, it can shoot well beyond the net at high speed. The feed it critical and ours is not yet consistent.
There are lots of posts on this forum about various gear ratios and motors. Once you get that sorted out, also be sure to consider the consistency of the feed angle of your balls into the flywheels. See for example this discussion:
https://vexforum.com/t/vex-flywheel-launcher-accuracy/30150/1
If you have the 393 motors with turbo gears and experience problems with them burning out, you might try (among other things), trying out speed or torque gears and making sure your gear ratio isn’t too high. For reference, for our flywheels, we’ve been using 2 torque motors and a 1:17.5 gear ratio, and our flywheels can run longer than 2 minutes at shooting speed (which is only about 80 power, but that also depends on your angle). We also have a variation of the flywheel that works using speed motors.
Another thing to consider is that once you get the flywheel assembled, you might have trouble with friction between your axles and bearings. To help with this, we made the holes where the axles go through the bearings slightly bigger using a drill; otherwise, the holes are too tight.
Lastly, another important part of being able to shoot well is on the programming side of things. If you find that you have to wait a long time between shots, consider investigating flywheel velocity control - jpearman and some others have some very good posts about that which have helped us.
Can you post a picture of your flywheels?
For the turbo gears, the jump of gear ratio internally may be too much for the torque requirements you have on the entire compound gear train. You generally have to step up the gear ratio gradually over the gear train.
Something about the ratio per step should be within reason based upon the overall number of steps in the compound gear train. Going too large of a jump in a step is too much stress.
http://www.daerospace.com/MechanicalSystems/GearsGen.php
I am not sure the gear ratio of the turbo gears themselves and how big of a step that is compared to the previous gear set inside the motor.
Yes, next time I’m in the lab which will be in a few hours.
Thank you I will be in our workshop in a few hours too and I can post a picture of our current setup.
I will leave more comments after seeing a picture. Considering the high gear ratios for flywheels this year, turbo internals seem quite excessive for a flywheel. I would recommend using high speed internals, which come with motors when you buy them.
This changes your internal ratio from a 2.4:1 (turbo) to a 1.6:1 (high speed).
Turbo motors can be used for a flywheels but they have an output of 240 rpm. You just need to just adjust the gear ratio accordingly to get the desired speed. For example, an external gearing of 10:1 with turbo motors would work quite well. This year, double flywheels usually have around a theoretical rpm of 2000-2500 and single flywheels around 2800-3500. You also posted that you are only using two motors for your shooter. I would recommend using four motors as it greatly reduces the risk of motors burning out and will also give you a better recovery time.
I am quite aware of this. My thought behind this switch is that it is much easier to switch the internal gears as opposed to your entire external gear setup. Sorry for any confusion.
@Felis_Orom Here’s a picture of our gear box. The two motors are 393s geared for torque (the default configuration). The last axle (farthest left) has two 5-inch wheels on it. Left to right, that’s an 84-tooth hear (with ratchet) to a 12-tooth gear, and then a 15-tooth sprocket (high-strength chain) to a 6-tooth sprocket. We considered removing the chain, but we saw a post from jpearman (I don’t have a link easily accessible) essentially saying that the chain could help reduce vibration; it does seem to have helped.
Our flywheels run at around 4000 RPM to shoot across the field.
A 1:17.5 ratio running at 4000 RPM sounds wrong. It should be in the range of around 1700 with no stress and 1400 in actual play. Mine on 1:35 does not even reach 3100 because of the motor stress and battery voltage due to an applied torque.
A double flywheel mostly runs at around 1300- 2000 max in a match. My single flywheel is shooting above the net with enough force to break it at 2100 RPM.
In actual building, a 1:7 external with Turbo gearing and 5 in wheels with break nets in play on a double flywheel. This is because a double flywheel only needs to spin at almost half the rate as a single flywheel.I would consider 1:25 too high because you run the risk of stalling easily unless it is a single flywheel. I’d stick between 1:16.8 (what I described) and 1:21 to keep motor stress down. Also, remember that wheel size matters, a larger wheel with shoot farther and put more stress on motors than a smaller one.
That number was based on the RPM readout from our PID program, but it’s entirely possible that there’s (significant) error in there because
- IEMs on torque motors sometimes read more like 620ish ticks/rev
- Short sampling period (20ms)
- Our other programmer wrote the encoder ticks to RPM calculation, and I trusted that it’s right; there’s a slight chance there’s an error in there, especially since consistent error won’t matter all that much (once it’s up to speed).
Thanks for pointing that out; I’ll check that calculation next time I’m working with our PID code. Right now I’m focusing on TBH because it’s easier to get tuned, and we don’t have a lot of time before our next competition. This issue won’t apply to the TBH code because that uses the RPM of our drive gear (which is also where the encoders are; I guess scaling up the calculation would increase the sensitivity, but I don’t think that’s an issue for us).
IMEs are supposed to read 627.2 ticks a rev anyway. The proper conversion should be :
IMErpm = (deltaEncoder) / deltaT(in miliseconds) *1000 * 60 /627.2 * externalRatio
Also, definitely try TBH. It is amazing once you get a deadband and proper gain on. I have it working, and I am trying to get it to recover faster (this being solved with just a gain adjustment) and more consistently.