Talon braking after power if off

With Jaguars, we could set a jumper to use braking mode, which would remain effective even when power is removed (e.g. at end of match). Talons can be set to brake mode, either by CAN or using the B/C CAL button.

I’m wondering if the Talon braking mode is effective when power is removed. “Saved” could mean that it is effective, or just remembered for the next time it is powered up.

Braking mode could be useful for letting your robot come gently down to the floor after climbing (depending on other factors such as gear reduction), but coasting might allow the robot to fall too fast.

I’m not sure if what you’re referring to here is when the robot is entirely powered off, or after the match is completed, but with the robot powered on and disabled.
Brake mode will work just as well in disabled mode as teleop, but will do absolutely nothing with the robot powered off.

Yes, I meant disabled, not the whole robot powered off. Did you find that information documented somewhere, and/or have you actually tested this behavior?

You can change the neutral mode, and therefore enable/disable the brake mode during the disable loop. Robot does not have to be enabled for you to modify it (mentioned in section 16.2 in talon SRX software reference manual).

Also check out our example below…

Be sure to update Talon SRX firmware to latest via our SDK installer.

Checkout section 4.3 in the Talon SRX software reference manual for the brake API.

Based only on experience , the answer is “sort of”.

When the power is removed the motor controller is off, period. However, when you spin the motor it’s connected to, the motor becomes a generator. Spin it fast enough and the motor controller will power up and apply braking (assuming braking was already enabled). We see this all the time when pushing unpowered robots across the floor. We can see the motor controllers for the drive motors coming online. Push too fast and the robot slams on the brakes.

While we haven’t seen any damaged motor controllers from doing this, the current path has to be through the inherent D-S body diode within the output MOSFETs and probably isn’t the best idea.

So you’re saying one could program the robot to disable brake and thus fall down exactly 1 second after match ends?

This is something I certainly want to know. I don’t think our lift is going to have a good clutch mechanism so, I want to have the lift brake turn off 5 seconds (just in case) after the match ends. That way, we’re safely on the ground to be picked up and carried off the field.

Of course, I want to test this before competition so that the robot doesn’t come crashing to the ground.

You can change the neutral brake/coast setting while the robot is disabled.

Keep in mind the “brake” mode of the Talon isn’t a physical brake. It merely shorts the motor leads together, which causes the motor to resist changes in the rotor position when neutral (orange LEDs).

It’s not a physical brake that ensures no motion whatsoever, and should not be relied upon for such a purpose.

With that said, there are definite advantages/conveniences to changing a motor’s resistance to back-driving during disable.

Make sense?

1 Like

This is great. I expect we will set the talon to brake mode, finish the climb, and wait 5-10 seconds after end of match (to be determined by testing), then possibly change to coast mode (also to be determined by testing). That way, the robot comes down slowly to at least touch the floor. There may be enough friction in the system (we are using a high gear reduction on the mini-CIM) that once the robot is touching the floor, there isn’t enough weight on the rope to continue back-driving. If that’s the case, maybe going to coast mode will allow it to continue coming down and rest level on the floor. This would make removal safer and faster.

This is a great question, but why don’t you just remove the rope from the airship while it’s still attached to the robot? We’re just going to turn the motor to unwind the rope after we get back to the pit.

How would you remove the knot from the DAVIT if the robot is pressing on the TOUCHPAD?

Our rope stretches, so it will be under significant tension when the climb is complete. If we have any mechanism that holds the robot up while disabled (ratchet, worm drive, etc.) then we expect the robot to be difficult to remove. A gentle-release mechanism solves this problem nicely.


The tension in the rope is the same whether it is stretchy or not.

I guess I was thinking of what people with hooks did last year - they’d lift the bot up and the hook would detach off of the bar. Same idea this year - two people lift the bot up while it’s still hanging from the rope (releasing the stress of the rope) and you’d be able to detach it after that.
I see the point of the gentle-release mechanism, though.

If the rope is stretchy, it will still pushing hard against the davit, so I think it will be difficult to lift the robot. Removing the pin at the attachment point to disengage the rope will work, but I still feel better about having the robot already down at that point.

How do I go about setting up this in Java on WPILib? I’ve checked but, I don’t see any method to change the brake settings in the TalonSRX class there.

EDIT: Nevermind, I just found it in the CANTalon Class.

Section 4 in Talon SRX Software Reference manual for anyone still looking.