active torque control?

So I'm working on a toshiba TB6560AHQ driver based electronics and I thought it might be useful to connect the torque control pins to the 644p, but I don't want to use 8 pins to do full control of all 4 drivers. With one line controlling both torque pins on the driver I can get 100% or 20% torque. With two lines I can get 100%, 75%, 50%, and 20%, with corresponding current draw. So here's the question for the firmware guys out there. Which axes might benefit from torque control during prints? (In between prints I assume you would use enable rather than torque control)

I'm currently thinking one line 100%/20% control for Z so you can lower the current between layers without losing steps. I'm not sure if any of the other axes would be able to use torque control during prints as they are almost always moving. So if you were going to add torque control to a firmware, which axes would you add it to and why?

Quote

Which axes might benefit from torque control during prints?

IMHO, none. Torque has to be sufficient to deal with friction and acceleration, and this is pretty much constant during a print. The only benefit I can see would be on the Z axis where going low torque instead of disabling them entirely might be better.


BTW., the Toshiba branch in Gen7's Github repo is still open ...

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Disclaimer: I'm not a firmware guy, but I think there could be some benefit to managing current from the controller.

I'm pretty sure my Z is disabled when it's not moving. Well, one of the firmwares did that, anyway. I'm about 7 hours from my printer right now, so I can't check. The Z axis screws won't back down. Well, they might, if you have a bunch of moves at just the right frequency. I don't think a RepRap can generate those frequencies or amplitudes of vibration though.

I could see bumping X and/or Y to a high current for acceleration, and drop to low current for holding a position or constant velocity. There is some electrical efficiency to be had but I don't know if there is enough processing time for the extra calculations.

As for Z, I thought the same thing at first, but now I'm not so sure. It looks to me like the drivers retain microstepping position while disabled, so that they come back to the last position upon enabling again as long as the step line hasn't been pulsed while disabled. So I'm thinking that disabling Z between layers is a lossless strategy too.

Re Gen7T, I am designing in eagle because that's all I know, and the pinout has had to change a bit to accommodate the new drivers and retain a single sided board, so I'm not sure if it's close enough to Gen7 anymore for you to want it in your repository.

The torqeue increase during acceleration is one of the things I thought might be useful. It seems like you could have high torque when you need it without overheating the motors by running them full on all the time. But I'm not a firmware guy either, so I don't know how easy/hard that would be.

Whenever the stepper motor goes from enable to disable, it snaps from whatever micro step position it is to a full step position.

It is bad practice to disable the Z steppers and has some potential for trouble with minimal gain.

---

www.Fablicator.com

Yes it does, but from reading the datasheet on the toshiba drivers it appears that the state is saved and even updates with the step input while disabled, so as long as you don't pulse the step line while disabled you should come back up upon enabling to the microstep position you were last at. I don't know what the pololu drivers do but I imagine similar functionality is fairly common across step/dir style drivers.

Edited 1 time(s). Last edit at 12/29/2011 05:47PM by bryanandaimee.

So here's my first attempt at a Gen7 Toshiba branch.

[forums.reprap.org]

Suggestions are welcome. I'll try to test it sometime soon. May be a couple weeks before I get it etched and built.

But a saved state just makes the problem worse, if I understand it correctly. if you are at 3/4 of a step when the motors are disabled, it will snap to the 4/4 state. then when you power up, the controller thinks it needs to be at 3/4 step, so it goes to the 7/4 position. so instead of being 1/4 away, your a full step away from where you left off.

and even if the controller could assume it would jump forward or backwards, like at 3/4 to 4/4 or 1/4 to 0/4, which way would it fall if it were at 1/2?

The only way it could work somewhat reliably is if the controller goes to a full step position before disabling, but then your z height will not be accurate and you might as well turn off microstepping.

Edited 2 time(s). Last edit at 01/05/2012 01:07PM by Buback.

Quote

then when you power up, the controller thinks it needs to be at 3/4 step, so it goes to the 7/4 position

To my understanding it'll go back to 3/4, because a full cycle of a stepper motor is two steps. If your assumption was true, there would be no way to change the direction of the motor's rotation.

Quote

you might as well turn off microstepping

Even if you'd always stop at full steps, microstepping has advantages. Like less noisy operation and less wear on the mechanics.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

So going from the nearest full step position (disabled) to the last microstep setting should always get you back to the pre-disable position.

I think so, but I don't disable my z-axis because the weight pressing on the screw plus the vibration from the X-axis motion might cause it to rotate with a stick / slip ratchet type action.

---

[www.hydraraptor.blogspot.com]

So maybe there would be a place for torque control for Z. Maybe a jumper or switch that would route the enable signal to the torque control pins instead, such that for Z the enable function in firmware would actually set 100% or 20% torque.