Topics created by Anon1337

@fcwilt the motors run better at 60V when you push it

@T3P3Tony true, the board being 24V max is what keeps the size down (and the board is pretty big already)

one solution might be a double sided PCB and perhaps moving away from the "sticks to a stepper" design (which is weird anyway since most of the nice new extruders use pancake steppers anyways...)

@Anon1337 said in Magnetic Encoder on a non-ideal stepper motor:

@fcwilt https://www.duet3d.com/duet3magneticencoder

How neat. I didn't know they made those.

So either add a spacer or buy a longer magnet.

If you need to buy a longer magnet you should be able to find a site like this that has all sorts:

K&J Magnetics

Frederick

@dc42 So AC SSRs connected to the 3HC PWM outputs? How were the segments controlled? Was it automatic or do you have to manually choose what segments to heat?

@dc42 yes I too believe the segmentation isn't ideal but more of a workaround

I don't have experience with closeLoop steppers, but from what I gathered, it requires a powerful MCU to reach same speed levels as openLoop.
Which IMHO is funny, because the main reason why I'd use CL steppers is to increase speed.

I'd like to build a Delta based on my differential belt drive. It allows us to double up the effector speed, but without running the steppers at twice the revs.
At the same time, I can choose any reduction ratio when I need higher precision.
All that would run on a Duet2 with driver-expansion

@T3P3Tony No I'm not aware of it.
I just had a flashback, when RRF in the beginning tried to support laser diodes via M106/M107 commands, instead of the established codes.

@Anon1337 said in Chunkiest motor the 1HCL can drive:

Since the 1HCL operates the stepper as if it was a servo this means that as soon as I start applying torque on the shaft that goes against the stepper and the stepper is let's say half a step between the 1.8° it can apply the full 10Nm instead of the normal 7 Nm?
Is that correct?

No, the maximum torque will be 7Nm. 10Nm can only be achieved at full-step positions, and only if the driver is programmed to drive both phases at full current simultaneously.

In principle you could get up to 10Nm by increasing the peak coil current to sqrt(2) times the rated value, but that should only be done when the motor is moving at a reasonable speed. Perhaps we will implement that on the 1HCL in future, but for now we don't.

@Anon1337 the 1HCL will work with a wide range of 4 wire bipolar stepper motors with encoders (up to 6.3A RMS). We recommend using stepper motors with 1.8 degree step angle and an encoder with >1000PPR/4000CPR.

The 23CL builds the Duet 3 expansion board control technology + encoder into the motor casing itself( with an optional brake) so a 1HCL is not required - it can be connected directly to the CAN-FD bus.

For external controllers I recommend you use the 6XD to produce step/direction signals and select controllers that can accept these. The 1XD also allows connection to these sorts of controllers.

Whenever you have multiple motors moving the same gantry you need to take precautions to avoid the gantry getting out of tram. you can set the homing files up to move both motors at once when homing and ensure both sides of the axis touch off the homing switches before homing finishes. the 6XD and 1XD also have inputs for the "alarm"/"error" outputs that many closed loop external drivers have that should be connected and dealt with appropriately, for example, stop homing if the alarm on one or both motors are triggered.

@Anon1337 i'd call that a gross over simplification, but, "sure"

If at all possible, I would proto different board thicknesses.

You MIGHT also need to slightly change the code in attiny to account for scaling effects in the tri/spiral strain gauge.