Duet 3 NEMA 23 Closed loop with Z axis also using Elec Brake

[deckingman]

@bearer So just to be clear to the OP, what you are saying is that he'd be better off buying the Duet 2 rather than the Duet 3 that it looks like he's already ordered (unless he breaks out the soldering iron etc). Just trying to clarify things for the OP.

[A Former User]

@deckingman said in Duet 3 NEMA 23 Closed loop with Z axis also using Elec Brake:

@bearer So just to be clear to the OP, what you are saying is that he'd be better off buying the Duet 2 rather than the Duet 3 that it looks like he's already ordered (unless he breaks out the soldering iron etc). Just trying to clarify things for the OP.

That is my understanding, yes. Can double check when I get home if no one beats me to it.

[gallaghersart]

Thanks for your replies,

I figured some coding would need to be done, I have coded many different types before. I have not investigated how to make a new firmware for the duet 3 yet. Modifying / altering duet 3 board to get me to a protype is not a concern at this time. Soldering or anything else was expected.

I need the extra amps for the motors and that is why I choose the duet 3. I also figured I would have extra pins that could be used for other uses.

I also want to have a spring plate / mini piston in the hot end to release pressure in mixing / melting chamber. No Retraction of filament by filament motor. So yes, this would need to also be coded.

Along with a few other things not yet used or programmed into Duet 3.

I am figuring that down the road production of this system would / may require Duet making a custom board for my plans.

The duet 3 is the closest option at this time for a solution for my future plans as far as I know. Just a starting point that is something I can work with.

Thanks!
`mike

[A Former User]

@gallaghersart said in Duet 3 NEMA 23 Closed loop with Z axis also using Elec Brake:

I need the extra amps for the motors and that is why I choose the duet 3.

But if you're using the smart steppers you power them directly and control them with logic level signals; no load on the Duet at all. (or did I get that bit wrong?)

I also figured I would have extra pins that could be used for other uses.

This is the big difference between the two; besides the IO ports with headers there aren't adittional logic level pins like the Duet 2 has. The 26pin header is only a SPI bus towards the Raspberry Pi.

My opinion is still unless you need the functions provided by the Raspberry Pi or the CAN bus, you'd be better of with a Duet 2. (Edit: which will be getting the same Raspberry Pi interface at some point in the undetermined future)

Schematics are online to review
https://github.com/Duet3D/Duet3-Mainboard-6HC
https://github.com/T3P3/Duet/tree/master/Duet2/Duet2v1.04

[gallaghersart]

@bearer
You are correct on external power, originally i was just going to use a NEMA 23 without any brake or closed loop.

But for CNC versions i want those options, then FDM system may just be basic nema23.

`mike

[A Former User]

Fair enough

the Duet 3 is more expandable, but mostly in terms of the available expansion boards for the CAN bus, while the Duet 2 has the 50 pin expansion port with IO pins.

Now its getting interesting!

[dc42]

For the future:

• We have prototype CAN-connected external stepper driver boards ordered, they should arrive soon
• We plan to make a "CNC version" of the Duet 3 main board supporting external stepper drivers

[gallaghersart]

@dc42
Music to my ears, thanks for the info, I will defiantly be looking forward to these and any other future capabilities.
`mike

[Andme]

@dc42 When is it anticipated that the CNC version of the board will be available?

[dc42]

@Andme said in Duet 3 NEMA 23 Closed loop with Z axis also using Elec Brake:

@dc42 When is it anticipated that the CNC version of the board will be available?

Probably not until Q4 this year.

[hippyengineer]

@dc42 I know Covid has affected your operation but are you still on track for a CNC board soon?

[dc42]

Unfortunately that project has been delayed, but it is next on our list. We have done some preparatory work for it and we expect it to be straightforward.

Meanwhile, I took a few hours out yesterday to try our prototype closed loop expansion board with a larger motor.

[gallaghersart]

Thanks for the update!

Looking forward to making this my next project.

In the image is the lower board the “prototype closed loop expansion board” if so, what board is above it?

Is it me or is that image of the drive so mammoth compared to the electronics? Reminds of the little clown cars that hold so many.

Thanks for update!
`mike

[dc42]

Yes that board is the prototype closed loop board. The board above it is a Duet 3 Mini prototype. And yes, the stepper motor really is that big, it's this one https://www.omc-stepperonline.com/p-series-nema-34-closed-loop-stepper-motor-8-5nm-1203-94oz-in-with-electromagnetic-brake.html.

[djstree]

@dc42 look forwarward to seeing them. Im eager to use ClearPath servos on my project with a Duet3.

By closed loop do you have input for 1vpp from linear encoders? Or SSI/BissC/1vpp for absolute encoders?

[dc42]

@djstree said in Duet 3 NEMA 23 Closed loop with Z axis also using Elec Brake:

By closed loop do you have input for 1vpp from linear encoders? Or SSI/BissC/1vpp for absolute encoders?

It has a quadrature input that can be used either for a digital linear encoder or for a rotary encoder built into a motor, as well as a separate SPI connector intended for a magnetic encoder that we will produce for attaching to the back of a Nema 17 or Nema 23 motor.

[djstree]

@dc42 BiSS C-mode would be good to have on the support list too. Open source protocol.
https://www.renishaw.com/media/pdf/en/de4d6a54313b48c3a4dadef7207ac4c1.pdf

"BiSS encoder output with Sin-Cos 1Vpp: BiSS encoders can also provide a 1V p-to-p sin/cos output for real-time control, since the on-demand absolute encoder data can come in too slowly for many control loops."

[dc42]

@djstree said in Duet 3 NEMA 23 Closed loop with Z axis also using Elec Brake:

@dc42 BiSS C-mode would be good to have on the support list too. Open source protocol.
https://www.renishaw.com/media/pdf/en/de4d6a54313b48c3a4dadef7207ac4c1.pdf

"BiSS encoder output with Sin-Cos 1Vpp: BiSS encoders can also provide a 1V p-to-p sin/cos output for real-time control, since the on-demand absolute encoder data can come in too slowly for many control loops."

Stepper motors intended for closed loop control invariably have quadrature outputs. Analog encoder output sounds very retro to me and right now I do not intend to support it. I designed the quadrature decoder on the board to support high data rates, including the data rates provided by Renishaw digital encoders with a resolution of 100 counts/mm at speeds in excess of 1m/sec. This is way faster than the quadrature decoders used in other open source closed loop stepper motor controllers.

I guess analog encoding might be useful for BLDC motor applications where speed is more important than precision.

[djstree]

@dc42 Thanks for the update! I've just been looking at linear motors / linears scales for a new project requiring 3.2 meter Y travel (moving gantry) on a 1.3m wide granite bed. Tough to decide what to go for IP55 wise.

[Mirage]

I am definitely keeping a close eye on this new expansion board. My new XY gantry for NEMA23 is ready to take the beating! And clearpaths are in stock