Migrating from Smoothie to Duet

I've ordered a Duet0.85 from TP3D and waiting for it to ship, just trying to do my homework on what I need to do to get it set up. I have a working smoothie config, how much can I simply carry over? Max Z, delta radius, arm length I assume. Do trims work the same way? Anything else I should watch out for?
I have a somewhat unique bed setup, I have M42 and M43 set up in smoothieware to turn one of the FET's on or off. This energizes a relay which controls power into a standalone temp controller with thermocouple that takes care of bed temp. I only turn the bed on and off via program. I know that RepRap firmware has a different code for M43, is it a matter of M42 S1/M42 S0 to turn on and off respectively? Conversely can I get a thermocouple input into the Duet? Smoothie can use an amplifier chip over SPI for this purpose, don't know if the Duet can do anything similar.
One of (several) reasons to try the Duet is the pressure advance function. I've read a few posts in this forum about it, but still not 100% on it's operation. From what I read it begins a retract before a line ending. Am I correct in understanding the amount is dependant on pressure built in the hotend? When does this material get "un-retracted"? What interaction if any does this have with my slicer generated retracts?

To answer your questions:

1. Yes you can carry your delta parameters over. However, I believe the M666 endstop corrections work in the opposite sense.

2. Yes M42 is used to turn output pins both on and off.

3. Dan Newman's fork of RepRapFirmware already supports up to four MAX31855 thermocouple interface boards connected to the SPI pins of the expansion connector. His code is in the 1.09p-alpha version of my fork but is untested and probably doesn't work yet. Thermocouple support should be fully working in the full 1.09p release when that is complete..

4. Yes the amount of retraction (and whether retraction happens at all) at the end of an extruding move depends on the amount of pressure reduction required. You may need to reduce the amount of retraction configured in your slicer, to avoid over-retraction. The "un-retraction" happens during the start of the next extruding move. In fact at the nozzle, unless you have the pressure compensation constant set too high, there shouldn't be any retraction due to pressure compensation, because the retraction done at the extruder will be taken up by the Bowden tube.

HTH David

---

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].

Thanks for the response, I'm looking forward to sometime moving the bed control to the board and the SPI chip was what I was thinking of using with the smoothie. Good to know it's being worked on.
Do you mean invert the signs on the trim values? I currently use negative trim on the smoothie, which I think moves the carriages down towards the bed. Reprap Firmware does the opposite?
As far as retraction goes, I have been playing a lot with the coast values in S3D and found that as I use those my retract value drops as well. I started with a retract of 4mm and basically exchanged 1.5mm into coast, so coast was 1.5mm and retract was 2.5mm. Pressure advance works in a similar way, but not in the same units then?

In RepRapFirmware, positive M666 trim is the same as increasing the homed height, which moves the nozzle closer to the bed. So invert the signs. I may change this sometime to be the same as Marlin and Smoothie, but not until I do a release with a new version number a opposed to just a different suffix letter.

If you have a good Z probe then you can use auto calibration. Autocal normalises the endstop corrections to average zero.

Coast to end does some of what pressure advance does, in that it relieves some of the pressure in the Bowden tube by the end of the extruding move. But it doesn't increase the pressure at the start of the next move to compensate, neither does it actually reverse the direction of extruder movement before the end of the move where necessary. Whether such reversal is necessary depends on the deceleration at the end of the move and the pressure advance factor.

---

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].

Got the board yesterday and got most things wired tonight. I notice that there is no M301 by default in config.g Is that something I need to add? I have values I used for the smoothie, I assume that they are no practical use here?

Most people find the default PID parameters in RepRapFirmware are a good starting point for commonly used hot ends. You can add an M301 command later if you find significant temperature overshoot or undershoot when heating your hot end. The bed defaults to bang-bang but can be switched to PID if you need to.

---

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].

Thank you for all of your help, and your hard work on the firmware. Still running 1.09m at the moment, I'll grab p as soon as it gets out of alpha. Working rather well so far, and I haven't even hooked up the ethernet yet. I also need to make a z probe, although my printer is very square and the bed is very flat and level.

If you are looking for a Z probe, have you considered my mini differential IR sensor? See [escher3d.com], and the fitting instructions for more details including compatible bed surfaces.

---

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].

I have, but it doesn't meet my needs thank you. I am planning on making a probe which works like that Renishaw probes used on mills and lathes only smaller and not with the IR transmitter. At the moment I am using a PEI sheet over aluminum plate and I think I might have issue with the adhesive film underneath causing odd readings, I also don't want to be restricted in the future as to what I can and can't use.

Quote
Koko76
I have, but it doesn't meet my needs thank you. I am planning on making a probe which works like that Renishaw probes used on mills and lathes only smaller and not with the IR transmitter. At the moment I am using a PEI sheet over aluminum plate and I think I might have issue with the adhesive film underneath causing odd readings, I also don't want to be restricted in the future as to what I can and can't use.

Thin PEI works very well with my IR sensor if you spray the bottom of the sheet black with heat resistant paint. I use thin PEI sheet on glass. I blogged about various surfaces I tried at [miscsolutions.wordpress.com].

---

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].

I read your blog, and again sorry not interested. I have no desire to pull up the existing PEI to spray paint it and I don't personally feel that IR is the right choice for this application. I have tools and knowledge suitable for building a probe that will repeat within .003mm or so, I do this kind of work regularly.

Koko76 - Care to share how you will build the probe? I'm interested in using my delta as a 3d CMM of sorts.

Mike

The probe is relatively low on my to do list. My machine is very square, and the towers located very well. It performs quite nicely with manual calibration. The principle will be the same as the Renishaw units, six balls arranged in a circle and a "spider" with 3 rods sticking out. Assuming that the same rod goes back to the same two balls (the entire spider doesn't twist) there is only one orientation the spider can take to the six balls. By looking at continuity between the rods and balls you can see if the spider has been disturbed from rest.

Ok, I'm on the latest and greatest firmware 1.09r-dc42 and I now have the ability to talk to the board over ethernet. My web interface says it's 1.06, which I think is a few steps back and most/many of the fields in the web interface are unpopulated with any sort of data. I was able to upload and start a print, but not many fields are filled in. What do I need to do in order to enable these? Do I need to update or wipe out the www directory on the SD card? Apologies if this is documented somewhere and I missed it, a link would be greatly appreciated in that case.

Update your www file I had the same problem sounds like it at least



[github.com]

Thank you, that has the web interface working ok!
Dc42, wondering if the following is possible, I'd like to simply turn the bed output either on or off, with no temp regulation. Is there a way to accomplish this in G code? I would like to use the heater output to turn a relay on or off.

There is no gcode to do that in my fork of RRF, but I think chrishamm's fork has a gcode that lets you set the state of any output pin.

Don't forget the flyback diode in parallel with the relay.

Edited 1 time(s). Last edit at 01/30/2016 02:46AM by dc42.

---

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].