Those of you running Duet electronics may be interested in my new release of RepRapFirmware. For users of delta printers, the main improvement is that you can now include the X and Y tower position corrections in the M665 command in config.g. More details at [forums.reprap.org].

I have also set up a wiki page on configuring and calibrating a delta running RepRapFirmware at [reprap.org].

Edited 1 time(s). Last edit at 07/31/2015 08:01AM 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].

Thank yor for additional M665 params. It is now possible to save calibration results and print without bed leveling.

Also, results of calibration is slightly differs, is it ok?

I hardcoded "M665 R105.08 L214.350 B90 H236.31 X-0.20 Y-0.98" and "M666 X-1.26 Y0.56 Z0.70" in condig file and then start G32 tests:

First start after M665 and M666 filled with all params:

16:25:20 M666 Endstop adjustments X-1.30 Y0.58 Z0.72
16:25:19 M665 Diagonal 214.35, delta radius 105.08, homed height 236.29, bed radius 90.0, X -0.16°, Y -0.62°
16:25:13 G32 Calibrated 6 factors using 16 points, deviation before 0.059 after 0.047

16:26:25 M666 Endstop adjustments X-1.28 Y0.60 Z0.68
16:26:23 M665 Diagonal 214.35, delta radius 105.08, homed height 236.30, bed radius 90.0, X -0.07°, Y -0.88°
16:26:17 G32 Calibrated 6 factors using 16 points, deviation before 0.047 after 0.034

16:27:25 M666 Endstop adjustments X-1.27 Y0.53 Z0.74
16:27:24 M665 Diagonal 214.35, delta radius 105.09, homed height 236.28, bed radius 90.0, X -0.16°, Y -0.85°
16:27:19 G32 Calibrated 6 factors using 16 points, deviation before 0.048 after 0.043

16:28:30 M666 Endstop adjustments X-1.27 Y0.54 Z0.73
16:28:28 M665 Diagonal 214.35, delta radius 105.18, homed height 236.31, bed radius 90.0, X -0.12°, Y -0.95°
16:28:21 G32 Calibrated 6 factors using 16 points, deviation before 0.048 after 0.043

16:29:29 M666 Endstop adjustments X-1.25 Y0.52 Z0.72
16:29:27 M665 Diagonal 214.35, delta radius 105.19, homed height 236.33, bed radius 90.0, X -0.20°, Y -0.89°
16:29:22 G32 Calibrated 6 factors using 16 points, deviation before 0.056 after 0.052

16:30:29 M666 Endstop adjustments X-1.25 Y0.57 Z0.68
16:30:28 M665 Diagonal 214.35, delta radius 105.04, homed height 236.31, bed radius 90.0, X -0.17°, Y -0.88°
16:30:22 G32 Calibrated 6 factors using 16 points, deviation before 0.054 after 0.050

Reboot and start again:

16:31:59 M666 Endstop adjustments X-1.28 Y0.60 Z0.68
16:31:57 M665 Diagonal 214.35, delta radius 104.99, homed height 236.29, bed radius 90.0, X -0.12°, Y -0.82°
16:31:53 G32 Calibrated 6 factors using 16 points, deviation before 0.039 after 0.033

16:33:01 M666 Endstop adjustments X-1.28 Y0.55 Z0.73
16:32:59 M665 Diagonal 214.35, delta radius 105.09, homed height 236.29, bed radius 90.0, X -0.15°, Y -0.69°
16:32:54 G32 Calibrated 6 factors using 16 points, deviation before 0.051 after 0.044

Also, web-interface affects print on Duet+RepRapFirmware like graphic LCD affects RAMPS1.4+atmega2560. There is audible clicks every 250 ms (as web ui refreshes) then printer moving hotend from X0 Y0 Z230 to X0 Y0 Z1 with speed F15000. (my setting is 100 steps/mm). This clicks become hard to hear at F9000-F12000 speeds. Also, if something is printing and I go to settings page of web ui then printer pauses print for about 300-600 ms causing melted plastic blobs.

Quote
roboprint
Thank yor for additional M665 params. It is now possible to save calibration results and print without bed leveling.

Also, results of calibration is slightly differs, is it ok?

If just the endstop corrections jump around, it suggests that the homing switches are giving inconsistent results. Reducing the fine homing speed in homedelta.g may help. If the delta radius and tower corrections jump around as well, it suggests either a mechanical issue (e.g. loose joints), or that the Z probe is giving inconsistent results. Reducing the Z probing speed (Z parameter in M210 command) may help if it is the Z probe.

Another way of checking the consistency is to make a copy of bed.g with the S6 parameter on the final G30 command changed to S-1. This will report the height errors at the probe points without changing anything. Run this file (e.g. print it just like a normal gcode file) multiple times and see how consistent the results are.

Quote
roboprint
Also, web-interface affects print on Duet+RepRapFirmware like graphic LCD affects RAMPS1.4+atmega2560. There is audible clicks every 250 ms (as web ui refreshes) then printer moving hotend from X0 Y0 Z230 to X0 Y0 Z1 with speed F15000. (my setting is 100 steps/mm). This clicks become hard to hear at F9000-F12000 speeds. Also, if something is printing and I go to settings page of web ui then printer pauses print for about 300-600 ms causing melted plastic blobs.

I'll pass that report on to Christian Hammacher, who does most of the work on the web interface code these days. My guess is that going to the settings page is causing the config.g file to be read from the SD card and sent to the web interface, and that is introducing too much latency. I don't think it should be hard to fix.

---

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].

David, after moving to Duet electronics I'm experiencing some type of issues that I can not explain to myself. I'm using Kossel Mini with your fork RepRapFirmware 1.09d-dc42.
There is part of my config:

M665 R105.08 L214.350 B90 H236.31 X-0.20 Y-0.98
M666 X-1.26 Y0.56 Z0.70				; put your endstop adjustments here, or let auto calibration find them
M92 X100 Y100 Z100					; Set axis steps/mm
M906 X1000 Y1000 Z1000 E700			; Set motor currents (mA)
M201 X5000 Y5000 Z5000 E5000		; Accelerations (mm/s^2)
M203 X20000 Y20000 Z20000 E3600		; Maximum speeds (mm/min)
M566 X1200 Y1200 Z1200 E1200		; Maximum instant speed changes mm/minute

Then I am printing OpenSCAD's cube ( cube([50, 50, 50]); ) using Slic3r's vase mode (If you remember, we used it in other topic) - everything is OK. Walls surface is very smooth and clean.

But then I print cylinder ( cylinder(r = 25, h = 30, $fn = 1024); or cylinder(r = 25, h = 30, $fn = 128); no difference ) using Slic3r vase mode, then walls are not clean, with ridges and valleys, also this bad pattern changes in different height. Same effect then I using Cura vase mode.



Also I have tried to print vase (the same g-code as I used before with Repetier firmware). Now my vase printed using Duet contains 6 symmetric vertical areas with blobs. No such 6 areas of blobls exist on older vases printed using Repetier firmware (and ramps/arduino electronics):


Mechanically printer is the same. I'm only moved from Repetier + Adruino 2560 + RAMPS 1.4 + A4988 drivers to Duet 0.6 board with RepRapFirmware 1.09d-dc42.

Also I tried different power supply (12V 30A LED power supply) - no changes.

What could it be? What should I check?

1. Can you confirm that you are printing from SD card?

2. Are the accelerations and instant speed changes in your config.g the same ones you were using with Marlin?

My delta is currently in pieces, but should be working again tomorrow, and I'll try printing the spiral vase cylinder then.

---

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].

1) Yes, I'm printing from SD card (I upload files manually via FTP and then select needed file via WebUI and print). Upload speed for card is => 300 KB/second, download speed is much faster (don't remember, about 900 KB/sec)

2) No, I have changed this values after moving to Duet board:

Original parameters from Repetier firmware EEPROM:

2500.000 = Print acceleration
2500.000 = Travel acceleration
30.000 = Max. jerk

Thank you for reminding me about accelerations, I'll check the original values from Repetier using Duet.

update: also I tried to remove X and Y parameters (X-0.20 Y-0.98) from M665 command (I thought this can affect quality), but without success.

Edited 1 time(s). Last edit at 08/15/2015 04:35PM by roboprint.

Is the 30.0 max jerk figure that you used for Repetier in mm/sec or mm/min, or something else? I'm guessing it's mm/sec but would like to be sure.

I just tried the spiral vase print from the model "cylinder(r = 25, h = 30, $fn = 1024);". Unfortunately, slic3r 1.2.9 makes a pig's ear of slicing it. On each turn of the spiral it raises the head by about 5mm, then drops it by almost the same amount at the end of the turn. And repeats that for every layer. Same if I set $fn=128.

If the bad pattern changes with height, it looks like a slicing issue to me.

Can you post the STL model for the vase?

---

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'm sorry, I forgot to specify mm/sec.

I have attached zip-file with openscad file, stl file and sliced g-code (I'm using 0.3 mm height and 150% infill for first layer), all settings are available at the end of g-code file.

Also I can attach Cura slicing profile for this model.

For vase I'm using this model: [www.thingiverse.com]
I do not remember what uniform scale I used for model, but real printed height of model is about 38.5 mm.

Edited 1 time(s). Last edit at 08/15/2015 06:29PM by roboprint.

Also I have tried to view (in 3D tab) slic3r's g-code using [gcode.ws] site and I can not see anything that would make me worry.

Edited 1 time(s). Last edit at 08/15/2015 06:24PM by roboprint.