Got better results with Mega 2560?

Hi there!

So I'm fiddling around with this quite a while now. Before I bought the Duet 0.6 I was using the clearly underpowered Mega 2560 / Ramps combination for my delta printer.
But somehow my printer was doing better before the "upgrade". A very noticable difference is the finish of the object, as you can see here:

[imgur.com]

The carriage was printed with the RAMPS combo, it is more glossy and layers are way better aligned on top of each other. I printed the carriage with repetier and almost no calibration. I got it running and just wanted to print the final parts, so I printed the carriages, idlers, platform and hotend mount. Everything printed well.

Now I spent a long time trying to calibrate my duet with dc42's fork of reprap firmware. And I still love it, but I just can't get good results. Every print is coming out matte and with other flaws. Tall objects are not square at all and more.

Using ABS, I tried following temperatures: 210, 220, 230, 240, 250 and even 260°C. Found that 230 - 240 works best. Tried speeds from 30 to 100 mm/s, extrusion multiplier from 0,80 to 1,20 (0,80 works best).
0,40 nozzle, 0,48 extrusion width. Tried layer heights 0,1 / 0,2 / 0,3.

My config.g file:

[pastebin.com]

Thank you

I suspect the tower corrections in your M665 command, "X1 Y-1", are the culprit, if your delta is an equilateral triangle. Very odd to have a negative correction value in there, and may explain why tall prints are not square.

I agree, those exact +1 and -1 degree tower corrections look very odd.

Are you using the same motor currents, speeds and accelerations in config.g that you have configured on your RAMPS boards and in Repetier?

---

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 remember printing out the advanced delta calibration object and my X-Z angle was 121° and Y-Z 119, so I thought it was right to put +1 -1 there. Before that it was set to -2 -2 by the auto delta calibration if I recall correctly.

For the currents, with the RAMPS I used around 0,50 to 0,55 VREF, which should result in a current of around 1 amp, which I am using now aswell. Acceleration was set to 2000 before, now it is 3000, with speed you mean maximum feedrate / speeds I suppose, if so that was set to 500 mm/s before now it's 300 mm/s ( 18000 mm/min )

The XZ angle in your print does not relate directly to the tower angle corrections you specify to the firmware. Also, if auto calibration was suggesting corrections of 2 degrees, that is higher than usual and I suspect a geometric error in your build - although it's not impossible that your towers really are that far out, if your horizontal extrusions are not all exactly the same length.

Please explain what you are trying to compare in the two photos. The only meaningful comparison is two prints using different firmware of the same object using the same slicer, layer height, speeds and filament - and I can't see that in your photos.

---

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].

This is for every object I printed before, and every object I printed after the Duet upgrade.
Now, layers seem smaller. I notice thicker ripples on the outside of prints done with RAMPS, even tho I am using the same layer height now. Now, every print is matte, no matter what or how I print. With RAMPS every print was glossy. It#s bothering me because I think my prints look odd now, with RAMPS they lookey like every other print I saw elsewhere.

Also, my prints were stronger with RAMPS. I think it's really weird but if I print something now and try to bend it, it snaps right off, without bending that much. If I bend something I printed with RAMPS, it will bend like crazy before snapping off.

Nothing changed, just the motherboard and firmware. It took like one to a few days to change the board / firmware. I thought the filament was wet so I put it in the oven for 3 hours @ 80°C, nothing changed. I just can't wrap my head around it.

It sounds to me that you are extruding at a different (probably lower) temperature now than before. RRF and Marlin use different thermistor models, so they may read different temperatures for the same thermistor resistance. The thermistor series resistors and ADC offsets also may be slightly different between your RAMPS board and the Duet. Try increasing extrusion temperature by 5C or 10C. Also, tell us what hot end thermistor you are using, and what parameters you have in the M305 P1 command in config.g.

---

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 tried 220 to 250°C almost completely in steps of 5°C. It is too much at about 250°C but there is not really much of a difference inbetween.
The thermistor was preinstalled, all I know is "100 Karat NTC B 3950". As it came without a fix in its socket, I surrounded it with thermal compound.

M305 P1 T100000 B4388 R4700 H0 L0

As everyone might have seen already, the problem is solved - obviously. As my previous post states, I was using 4388 as B value for the thermistor, while it should be 3950.
Can't believe I haven't noticed this before, sorry for bothering with a seemingly obvious problem. But thank you for pointing me in the right direction.