Which Gearbox Should I Use?

[Monochron]

Originally Posted by SeanFitz

So if it's about temperature, would the various components near the breaker on the bot heating up affect weather or not it trips? Also, once the breaker does trip, is it true that it becomes more likely to trip again at a lower temp?

Very little heat is going to be transferred through air to the breaker. So no, as long as your breaker is thermally isolated you won't see any difference.

To your question about the 40 seconds being cumulative, that number is an estimate based on continuous stall conditions. If you stall for 20 seconds, then wait for a minute and stall for another 20 seconds, it likely won't trip but it is absolutely possible. Giving the breaker time to cool down will help, but each breaker is going to be manufactured with slight differences. Doing your best to avoid stall conditions (ie. ensure that you wheels will slip on the carpet instead of freezing) is the best thing you can do to avoid tripping.

As for your second question, I don't believe there have been experiments to track the propensity for earlier tripping after a breaker has tripped once. I believe, though, that anecdotal evidence suggests that they DO trip quicker after tripping once.

[SeanFitz]

Originally Posted by Monochron

Very little heat is going to be transferred through air to the breaker. So no, as long as your breaker is thermally isolated you won't see any difference.

To your question about the 40 seconds being cumulative, that number is an estimate based on continuous stall conditions. If you stall for 20 seconds, then wait for a minute and stall for another 20 seconds, it likely won't trip but it is absolutely possible. Giving the breaker time to cool down will help, but each breaker is going to be manufactured with slight differences. Doing your best to avoid stall conditions (ie. ensure that you wheels will slip on the carpet instead of freezing) is the best thing you can do to avoid tripping.

As for your second question, I don't believe there have been experiments to track the propensity for earlier tripping after a breaker has tripped once. I believe, though, that anecdotal evidence suggests that they DO trip quicker after tripping once.

So if the breaker trips, would I be better off letting it cool and then continuing on or should I try and replace it? The answer to this would factor into the placement of the electronics (I.E. how hard it is to get to each part)

[Oblarg]

Re: main breaker trips, our team had a lot of experience with this over the past season with an aggressively-geared 6-CIM SS drive. Way more than we would have liked. Here's a rough outline of what we discovered:

Six stalled CIMs will pull around 40 amps/CIM from a fully-charged FRC battery. Note that this is significantly lower than the rated stall current of a CIM - this is because as you draw more current, the effective voltage of your battery drops. The standard FRC main breaker, from our testing, could sustain this for anywhere from 10-30 seconds. The effects are "quasi-cumulative," in that since the breaker is triggered by heat buildup, and heat dissipates over time, you regain some of that time when you remove the current draw, but in our experience, not very much over the time scales involved in an FRC match.

Six slipping CIMs on our robot (120lbs w/ bumpers, drive geared 6.11:1) pulled around 30 amps/CIM and could be sustained almost indefinitely without breaker trip over the course of an FRC match. Our robot in this configuration was just on the very edge of being able to slip the wheels, so this was essentially the "worst-case" scenario for wheel-slipping, and other configurations would likely draw less.

What can be discerned from this is that as long as you can be absolutely sure that you will not stall your drive motors while pushing, you should not have breaker problems with a 6CIM drive. Unfortunately, figuring out precisely what gearing gives you that certainty is nontrivial, as it depends critically on several subtle factors (how much current your robot is drawing apart from the drive, how long the wires leading to your drive motors are, how fresh you expect your battery to be in-match, among others). You could try to calculate this beforehand (the math is not hard), but I'd recommend simply erring on the side of caution and picking a relatively slow low-gear (no faster than 7FPS*). Do that, and you should have no worries about breaker trouble.

Your fast gear can be however fast your drivers feel comfortable with. Just make sure you do not get into a pushing match while in your fast gear, or you can (and will) trip the breaker.

* How I arrived at that number: Our robot was just able to slip the wheels at 14fps gearing while weighing 120lbs. Our wheel COF was 1.1. For a "worst-case" scenario, assume a 150lb robot with a 1.5 wheel COF. Both factors scale linearly with required torque, so our gearing would have to be 14FPS * (120/150) * (1.1/1.5) = ~8 FPS. Reducing this to 7 FPS gives you a bit of overhead for not-quite-charged batteries/other stuff pulling current/whathaveyou. You could probably claw back a bit by having better (read: shorter and/or fatter) wiring than what we had, or by having exceptional drive efficiency (though, subjectively, I thought ours was pretty good on the robot in question).

[SeanFitz]

Originally Posted by Oblarg

I'd recommend simply erring on the side of caution and picking a relatively slow low-gear (no faster than 7FPS*). Do that, and you should have no worries about breaker trouble.

Using the calculator from earlier in the thread, the gearbox I'm currently thinking of buying is going to have a top speed of 4.8FPS, so I guess that works out fine for me.

The only thing that worries me is that the calculator says the maximum current draw at maximum is 155 amps. Inj your opinion, is this number realistic or is it underestimated?

Also, your FIRST (hah, pun) hand experience with a 6-CIM chassis is very helpful.

[Oblarg]

Originally Posted by SeanFitz

Using the calculator from earlier in the thread, the gearbox I'm currently thinking of buying is going to have a top speed of 4.8FPS, so I guess that works out fine for me.

The only thing that worries me is that the calculator says the maximum current draw at maximum is 155 amps. Inj your opinion, is this number realistic or is it underestimated?

Also, your FIRST (hah, pun) hand experience with a 6-CIM chassis is very helpful.

I don't know how detailed the calculator you're using is, but that sounds realistic. That's not at all a worrisome number, either - the FRC breaker can take 155 amps for essentially a whole match. During the times last year when our robot was not running into problems, we were pulling ~180 amps in pushing matches, and that was about the worst-case possible while still slipping the wheels. It was when we got near the 240-amp mark that things went south, and I only ever saw that type of current draw when the motors were stalled. You will not stall your drive motors when geared for 5FPS, barring some sort of cataclysmic failure.

Make sure that if you're going for a 4.8FPS gearbox that it's a two-speed. That's far too slow to function well as a single-speed drive.

[SeanFitz]

Originally Posted by Oblarg

I don't know how detailed the calculator you're using is, but that sounds realistic. That's not at all a worrisome number, either - the FRC breaker can take 155 amps for essentially a whole match. During the times last year when our robot was not running into problems, we were pulling ~180 amps in pushing matches, and that was about the worst-case possible while still slipping the wheels. It was when we got near the 240-amp mark that things went south, and I only ever saw that type of current draw when the motors were stalled. You will not stall your drive motors when geared for 5FPS, barring some sort of cataclysmic failure.

Make sure that if you're going for a 4.8FPS gearbox that it's a two-speed. That's far too slow to function well as a single-speed drive.

The link for the calculator is here. If I'm reading it right, with a VEX 3-CIM ball shifter (yeah, it will be two-speed) with a 3.68 spread and a 24-60 third stage, the total amps at max are 155, but that's with a 1.4 wheel cof. Take a look and see if the calculator looks right.

On another note, what did you guys use for wheels last year? Is there anything you would change?

[DampRobot]

Originally Posted by SeanFitz

So if the breaker trips, would I be better off letting it cool and then continuing on or should I try and replace it? The answer to this would factor into the placement of the electronics (I.E. how hard it is to get to each part)

Yes, once you trip it once, it's easier to trip the second time. I've got this from personal experience, and from talking to top teams that do deal with breaker trips a lot.

Whether or not you should replace it after every trip is a difficult question to answer. I know some teams that bring a lot of breakers to competition and know they're good, so they swap it out any time they trip their breaker in a match. On the other hand, when we had to switch out a breaker, we didn't have any extras ourselves, so we had to borrow one. Bad decision. For whatever reason, the breaker we got tripped a lot more easily than the one we had replaced, even though (according to the team we borrowed it from) it was new out of the box.

At this point, I'd guess that the breakers have some Q/C issues, and top teams just test their spare breakers to make sure they're good to go before swapping them out. So, yes, already tripped breakers are easier to trip, but if you can't be sure of the quality of the breaker you're switching them with, I'd be cautious about making the swap.

In terms of placement, keep in mind that you are going to have to access the switch to turn the robot off, so you don't want it in some crazy hard to get to place anyway. On the other hand, you do want really short wire runs between the battery, the breaker, and the PDB to minimize resistance, which usually means putting the breaker in the middle of the robot (which can be harder to get to). So, like most things in engineering, it's a tradeoff.

[Oblarg]

Originally Posted by SeanFitz

The link for the calculator is here. If I'm reading it right, with a VEX 3-CIM ball shifter (yeah, it will be two-speed) with a 3.68 spread and a 24-60 third stage, the total amps at max are 155, but that's with a 1.4 wheel cof. Take a look and see if the calculator looks right.

On another note, what did you guys use for wheels last year? Is there anything you would change?

Playing around with the drive calculator, the numbers look pretty reasonable. It overestimates the max current on our configuration last year by a fair margin, for what it's worth.

We used the KOP wheels - our drive was a modified AM14U. If I were to re-do that drive, I'd have splurged for two-speed transmissions and used wheels with plaction tread. Our experience was not bad, but it was a solid indicator that 6-CIM single speed is not as great an improvement from 4-CIM single speed as one would think, since if you're geared to move reasonably fast you're still limited in pushing ability by the main breaker (in a sense, you're just trading problems with the drive breakers for problems with the main breaker, and the only real gain is acceleration). To really benefit from the improved acceleration of 6CIMs and still be able to push, you really do need a two-speed transmission.

Edit: Upon further playing with the drive calculator, it appears that even for a traction-limited drive it is using the static wheel COF for the "max" numbers which is...well, I hesitate to say that it's wrong because that might be what they intend but if that's what they intend then it's rather misleading because that's not what you're going to see in a pushing match. The numbers appear dead-on if you set the static COF to a realistic dynamic COF.

[thatprogrammer]

You may need to use a software limit on how much deceleration can happen at once with your robot. Going from high speed to reverse can blow the main breaker if their isn't a SLIGHT delay *like 1/5th a second*. You may want to look into cheesy drive to improve control at high speeds.

[Monochron]

Originally Posted by SeanFitz

So if the breaker trips, would I be better off letting it cool and then continuing on or should I try and replace it? The answer to this would factor into the placement of the electronics (I.E. how hard it is to get to each part)

Like I said, the consensus is that breakers do trip faster after being tripped once. If you are asking if it is worth the work/cost/etc. to switch after each break then that is a question only you and your team can answer. If your robot never really reaches serious heating conditions in gameplay, but you had some strange electrical problem or otherwise that caused the trip, it may not be worth it.

If your robot may trip the breaker in normal play though, the risks of continuing to run on the same tripped breaker likely outweigh any work/cost benefits.

[SeanFitz]

Originally Posted by Oblarg

Edit: Upon further playing with the drive calculator, it appears that even for a traction-limited drive it is using the static wheel COF for the "max" numbers which is...well, I hesitate to say that it's wrong because that might be what they intend but if that's what they intend then it's rather misleading because that's not what you're going to see in a pushing match. The numbers appear dead-on if you set the static COF to a realistic dynamic COF.

From your experience, what would be a realistic dynamic COF for some high friction wheels in a pushing match on carpet?

[Oblarg]

Originally Posted by SeanFitz

From your experience, what would be a realistic dynamic COF for some high friction wheels in a pushing match on carpet?

For the HiGrip wheels, I believe it's somewhere around .7-.8. I do not know what it is for roughtop or wedgetop tread.

[SeanFitz]

Originally Posted by Oblarg

For the HiGrip wheels, I believe it's somewhere around .7-.8. I do not know what it is for roughtop or wedgetop tread.

Probably only a little more, right. Like .9-1.1?

[Oblarg]

Originally Posted by SeanFitz

Probably only a little more, right. Like .9-1.1?

Scaling it proportionally to the static COF seems like a reasonable thing to do without additional information, yeah.

[SeanFitz]

Originally Posted by Oblarg

Scaling it proportionally to the static COF seems like a reasonable thing to do without additional information, yeah.

Just for the sake of filling out the calculator correctly, what do you think the kinetic COF would be when the static is around 1? 0.7-0.8?