AMIS-30543 drivers

I'm really liking the look of the new AMIS-30543 drivers. With the SPI based control for micro stepping and current limiting we could really have some fun with these. Has anyone looked at dropping the current on a motor when it's not moving, or dropping the micro-stepping during high speed moves? Not sure if that second one is going to fly - changing the micro-stepping while the axis is in motion sounds like a recipe for missing steps, but limiting changes to when the motor is stationary sounds like it defeats the purpose.

The drivers are a bit pricey at the moment (~$20), but presumably the clones will soon arrive. I see there was already an announcement post [forums.reprap.org]

James.

Quote
JamesK
Has anyone looked at dropping the current on a motor when it's not moving, or dropping the micro-stepping during high speed moves?

My fork of RepRapFirmware (the firmware that runs on Duet electronics) already reduces the motor current to a user-configurable value when the motors have been idle for 30 seconds. The default is 30% of the normal configured current. The Duet doesn't use SPI-controlled drivers, it uses standard A4982 drivers with Vref fed from a DAC.

Variable microstepping would certainly be a possibility, but if the electronics has problems generating step pulses fast enough when doing high speed moves, it's probably easier to just send 2 or 4 step pulses at a time (which AFAIR Marlin already does). However, modern 32-bit electronics is quite capable of generating the step pulses at the required rate, up to at least 32x microstepping.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Sounds good. The duet seems like a nice controller, but I'm a little cautious about having the drivers soldered directly on the board. Perhaps with this new generation of spi configured drivers we can have the best of both worlds - software configurable current and modular/pluggable drivers. If the prices come down a bit I might look into using these with a cheap Due clone for my next build (that sounds laughable considering the slow progress on my current build, but I might as well bow to the inevitable addiction).

Quote
JamesK
Sounds good. The duet seems like a nice controller, but I'm a little cautious about having the drivers soldered directly on the board. Perhaps with this new generation of spi configured drivers we can have the best of both worlds - software configurable current and modular/pluggable drivers.

I am working on a low-cost 32-bit board design that has pluggable drivers with software-controlled current. The drivers are just like the standard stepstick-type drivers, but slightly longer to accommodate a 9th pin at one end of one side, which is the current control pin. Standard stepstick drivers can be modified to add the extra pin. It's easiest if you buy them with the header pins not soldered in, so that you can solder a 9-pin strip on that side. Then you remove the pot, and add a wire from where the pot wiper was to the extra pin. The board will also accept unmodified drivers too.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Brilliant idea! That will really let us extract the maximum value out of the existing drivers.

Why not take the idea of improving the stepstick design to the next level and implementing error sensing to the driver boards (the ics do already have it) imagine correcting lost steps...

Oh, soory, I just brealized that the AMIS-30543 pololu boards are exactly just that pololu2.0 boards! HOW FINE!

Now we just need a smoothie-compatible board where the new drivers would fit into.. something like an open source version of the azsmz with the new pololus

Hi!

Any news on this topic? I find it quite interesting what the drivers are capable of. I just ordered one to play around.
Also the AZSMZ board is OS now.

The AZSMZ is a very simple design / ripp-off of the smoothie-board...I think I would be able to design a modified version of the board with AMIS-30543, however I would not be able to write the code nor modify the smoothie ware.