Problems with design, or "KitBot's math seems wrong"

[bardd]

I've been trying to design my own drivetrain, and based my math on this tutorial. I designed for 30 amps, and the result was pathetic: the robot could only go at around 5ft/s, and could barely push a baby if it tried real hard. I tried messing with the current, bringing it as close to 40amps as I dared, but the results weren't any better.
I was sure I did something wrong, so I did the same math on 2013's KitBot. The result showed each motor will draw 141 amps. Being sure I was wrong again, I used JVN's calculator, which said exactly the same. Even with two motors per gearbox, twice what was supplied in the KOP, each motor would draw 72 or so amps.

My question is, how is this possible? CIM motors are always connected to a 40amp fuse, so how doesn't it break when the robot tries to move? How does any KOP bot ever move without breaking anyway? There has to be something I'm missing here, since I saw hundreds of fully-functioning and certainly-moving KitBots, including our own. Also, how is it possible the standard, unaltered KitBot has every motor draw 141amps when a CIM's stall current is 133amps?
I narrowed it down to some jaguar/victor magic, since every other aspect of the system, as far as I understand it, can't cause this, while I don't really know how a motor controller works.

More than wanting to satisfy my curiosity, I would also like hearing how I should factor whatever it is into a design (of anything, not necessarily my drive).

Thanks in advance.

(tl;dr KitBot has a weird current quirk where the motors draw so much current that it should make the fuses on the PDB break/jump, but they never do. Why?)

[AllenGregoryIV]

Someone is going to come along and give a much better explanation than I am but I'll start you off.

A couple things to consider
1. Circuit breakers aren't rated for when they trip, instead they are rated for how much current they will allow indefinitely. That means that you can draw 40A from a 40A breaker for forever. Draw more current than the rating and eventually it will trip but it's a matter of time and how much current you are drawing. A short circuit will have them trip immediately. 45A will take longer. You can look up the data sheet for each breaker and find a graph of when they trip.

2. Most drivetrains aren't ever in a stall condition (pulling full current for very long). The kitbot breakers do trigger often if you put a two CIM kitbot into any sort of pushing match where the motors attempt to get near stall current.

This year we ran a 6 CIM single speed drivetrain that was geared for around 14-15fps (in our last three events). At stall each CIM was drawing somewhere around 55A+. Our driver was trained to not stall the robot for too long because it wasn't traction limited and we would blow the main breaker. We tripped the main breaker once at Razorback and browned out the radio once at IRI but other than that we never had issues.

[Gregor]

Originally Posted by AllenGregoryIV

Someone is going to come along and give a much better explanation than I am but I'll start you off.

A couple things to consider
1. Circuit breakers aren't rated for when they trip, instead they are rated for how much current they will allow indefinitely. That means that you can draw 40A from a 40A breaker for forever. Draw more current than the rating and eventually it will trip but it's a matter of time and how much current you are drawing. A short circuit will have them trip immediately. 45A will take longer. You can look up the data sheet for each breaker and find a graph of when they trip.

Here's the data sheet for our 40A breakers.

http://www.snapaction.net/pdf/MX5%20Spec%20Sheet.pdf

[bardd]

After reading your comments and having a short conversation with Gregor, everything's clear to me now. I'll sum up what I just learned so everyone can learn from my mistakes and correct me if I misunderstood:

1. While driving at a constant speed, the motors only draw the current needed for them to spin at the angular velocity they already have (e.g. if the wheels spin at 400 RPM and the gearing is 10:1, the motor will only draw the current needed to spin the motor at 4000RPM).
2. The time where you need to worry about breakers breaking is during acceleration. You need to design your drive so that when accelerating the motors don't draw too much current for too long.
3. Breakers can withstand more current than the current specified on them, but only for a certain amount of time, that can be found on the breakers specksheet.
4. Breakers=/=fuses. When a fuse breaks it's permanent and the fuse needs to be replaced, while a breaker will re-set some time after breaking (actually knew that before, but I use the word "fuse" for both out of habit so I'm taking this chance to clarify).
5. When designing a drive, don't use a tutorial that's basically for non-drive mechanisms.

If I got any of this wrong, or there's an important point that you think should be mentioned, feel free.

Thanks, Allen and Gregor!

[Mr V]

Originally Posted by bardd

[*]Breakers=/=fuses. When a fuse breaks it's permanent and the fuse needs to be replaced, while a breaker will re-set some time after breaking (actually knew that before, but I use the word "fuse" for both out of habit so I'm taking this chance to clarify).

A little clarification on Circuit breakers. There are different types available. The ones that we use in the power distribution board on FIRST robots are of the self resetting type. After a rest period they will as the name implies reset themselves. The main breaker we use is a manual reset style. So once it trips you must reset it by moving the lever.

The majority of circuit breakers and all of the FIRST legal are a type of time delay style that will allow massive over current situations for a short time. There are also fast acting circuit breakers that will trip quicker or with less of an over current situation.

Fuses on the other hand are most commonly fast acting but you can get time delay fuses in some styles too.

Determining the amount of power that a motor will draw is a little more complicated than looking at the max current at a particular RPM. That is valid when it is supplied a constant power supply at the rated voltage. However with a motor controller that is providing pulsed power the average current through the motor can be lower than the max. Yes it might be pulling 80 amps at a particular rpm when power is applied but if the motor controller is commanded to be operating at 50% to achieve that rpm then the average current would only be 40 amps. The circuit breaker is then getting a "rest" period where it can cool between the times it is supplying that 80a so it will more or less act as if it was only carrying 40 amps and should be able to do so more or less indefinitely.

[Ether]

Originally Posted by Mr V

Yes it might be pulling 80 amps at a particular rpm when power is applied but if the motor controller is commanded to be operating at 50% to achieve that rpm then the average current would only be 40 amps. The circuit breaker is then getting a "rest" period where it can cool between the times it is supplying that 80a so it will more or less act as if it was only carrying 40 amps and should be able to do so more or less indefinitely.

Put a small resistance in series with the breaker¹ and display the voltage drop across it (which will be proportional to the current) with an oscilloscope. You might be surprised at what you see.

¹or just measure the voltage drop across 1 foot of wire coming from the breaker