paper: What Does “Stall” Mean, and Why You Should Use the 775pro

Thread created automatically to discuss a document in CD-Media.

What Does “Stall” Mean, and Why You Should Use the 775pro
by: asid61

A paper about how stalling a motor works and why you should use a 775pro over a BAG motor

Not the cleanest paper I’ve written, but I’m tired of sharing google drive links. This is an explanation of how stall works for DC motors, and how when people say “The BAG motors last longer stalled than 775pros do” they’re not looking at the full picture. I’ll make any additions, improvements, or clarifications as they come up.

What Does Stall Mean and Why You Should Use the 775pro.pdf (302 KB)

Interesting read, still going through. I noticed a lot of assumptions that I need to wrap my head around to see how well they apply to the situations I most commonly find our team’s design in.

One note too, there’s been some discussion recently on the exact nature of the current limiting behavior of the speed controllers in FRC. As far as I know now, the current limiting functionality is primarily targeted at battery-draw current limiting, not at motor power limiting. Buyer beware.

What kind of assumptions did you notice? I’d be glad to explain them.
We limited our elevator at 5A, 10A, 15A, and 20A at different points during the season and the Talon reported that those amounts of current were being drawn from the 12V supply. I’ve heard that this doesn’t translate into power dissipated by the motor, but for us it’s worked very well. In any case, a 775pro limited to 20A still has more output power than a BAG running at peak power.

The two that stood out to me at first:

0A drawn at runout - presuming frictionless motor & mechanism? I think this is fair enough since the majority of the conclusion is about behavior at stall.

The other one was saying Kt was the same between BAG and 775Pro. Does this hold throughout time, as the motors heat? The Vex charts show stall torque dropping off (sometimes by ~50%) as heat builds up.

I’m wondering if it’s possible to take the raw Vex data, and with some interpolation, draw a chart of current-draw over time for constant-torque curves. I feel like this would explain a lot… added to my rainy-day backlog :slight_smile:

Is anyone else getting a “ERR_RESPONSE_HEADERS_MULTIPLE_CONTENT_DISPOSITION” trying to download this file?

Yes. Seems broke somewhere

I’m also getting this error, can anyone post another link?

Yep. On a desktop browsing with Chrome 65.0.3325.181

Download worked with Firefox, was broken in Chrome.

Seems to be that Chrome cannot download files with commas in it like this one.

If you assume the adjusted kt are the same, then running a BAG motor and 20A and a 775pro at 20A will make the same amount of torque at the output. However, at 20A the 775 will be spinning a lot faster, approximately 1.5x as fast as the BAG, meaning it’s also putting out 1.5x more power.

Look at the motor curves for BAG and 775Pro and do this calculation:

BAG at 12V and 20A generates approx 147W mechanical power

775Pro at 12V and 20A generates approx 172W mechanical power

That’s a 17% difference, not 50%

You are correct, I must have zonked out on the math for that last bit. I’ve updated it to reflect that. I probably mixed up the RPM-adjusted kT and the actual kT.

I reuploaded the paper and it seems to download fine on Chrome.

The 775pro specs say that it’s perfect for applications where it won’t be stalled for long periods of time. What would be considered a “long period of time?”

Our team has two 775pros on their elevator, but they keep burning out. Or at least that seems to be the failure. The motors cease functioning and they are extremely hot and some of them have had pieces of carbon (brushes??) fall out of them.

Any thoughts or ideas?

VEX has done a fair amount of testing to tell you just how long “long” is. If you check the page for the 775proon motors.vex.com, the third chart shows you how long the motor will last at a given voltage. The point where the curves drop to around 0 is about where the motors will fail. For instance, 12v will last maybe 2 seconds, while 8v could get you 10 to 20 seconds.

Regarding burning out, my guess is that you’ve got too little reduction going on. This is a good point to look at graph #1 on the motor page. When operating, you want the motor to operate between its max power (most work is done by one motor) and its max efficiency points (the least electrical energy is turned into heat). Since the 775pro is fairly easy to smoke, you should go closer to max efficiency, so there’s less heat made to smoke your motor.

It is not necessary to stall a motor at all in order to burn it out. Check out the “peak power test” on the Vex 775pro page. You’ll see that if you run a 775 pro at the peak power speed for half a match, it’ll burn up. (This isn’t just a 775pro, but a fairly general behavior.) How much current does your motor draw?

Take a closer look at the Locked Rotor Stall Test graphs in the paper (provided by motors.vex.com). For each motor, depending on what voltage you use, you can see the torque provided by the motor per time. Where the motor torque drops off is where the motor “burns up” and is irreversibly damaged.

The whole point of this paper is that, if you design your mechanism properly to not need all 12V from the 775pro, you can use them in “long-term” stall conditions. For the 775pro, stall with anything over 6V (ie 66A) will burn the motor in a matter of seconds. At 6V, the motor lasts about 180 seconds (longer than the match). Make sure your mechanism is geared properly so that it doesn’t need more than 6V at stall to maintain the desired force, and you should be fine.

edit: double sniped :yikes:

The Power Distribution Panel will also let you track your current draw to see if you are going over the safe limits, but that can be hard to program for many teams.
The logs can be hard to find for many teams, but there is literally no programming required to check current draws. It’s available in the default log files.

(How do I know? SVR’s wonderful CSA taught me how to read the logs three months ago. And I’m barely even a controls novice - so if I can do it, anyone can!)

This link only discusses CPU/connectivity issues, but current draws are available through the same interface:
https://wpilib.screenstepslive.com/s/currentCS/m/troubleshooting/l/599678-driver-station-log-file-viewer

Not sure, but I will find out.

I’m picking up a lot of good information in this thread, I really appreciate it!

I’m talking about using the current draw you get over CAN to do a closed-loop that limits the amount of current you draw from the PDP for each motor controller. That’s a fair bit trickier than just checking the logs, unfortunately. Without that kind of control you can check to see if you’re exceeding the current limits, but you can’t stop it from spiking without sacrificing some efficiency.

I first fully got my head around all this data after watching this video by 1678 Citrus Circuits.