Filament feeder slipping

Quote
rayhicks
it looks like version 57t has a boundary issue maybe - causing the low voltage on short to appear as a large negative value (not that this is going to matter in normal use, but I thought I'd bring it to your attention dc42)

The spreadsheet gives exactly the same values. The problem is that the thermistor theoretically has a finite resistance even at infinite temperature, so a zero resistance doesn't make sense. But for tidiness, I'll change the next version so that a resistance below the theoretical minimum reports a temperature of 2000C.

---

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].

Ah yes - I wrote that before I'd downloaded the xls

Cheers

Ray

Groneorm and KevM - it may be that your problem is related another one which appeared today, if so there's a potential fix... and it's fairly straighforward to find out (using one resistor and a multimeter to measure its resistance and the voltage across the thermistor connector when open circuit and when the resistor is connected to it) - please take a look at this thread: [forums.reprap.org]

Ray

I tried bumping up the motor current to 1200 as suggested here earlier, it still struggled and it sounded awful and the bit it printed out was pretty messy.

I'm clearly not as electrically minded as the rest of you guys, all this talk of resistors thermistors and ohms just baffles me,

is there not a quick fix for the novice who just wants to print stuff ? I've not had a successful print in about 3 weeks now, I just want to curl up in a ball and cry.

KevM - if it IS the same problem (and chances are that it IS if you have tried all else), you should be able to increase your extruder temperature by 10-20 degrees and get on and print - just ignore the fact that you're using say 205, when people say it should be 185 (the problem is that your hot end THINKS it's at 205 when it's really at 185) etc. I've printed PLA at 240 by accident - it doesn't print very well, but it does no harm


Cheers

Ray

Edited 1 time(s). Last edit at 03/11/2014 06:56PM by rayhicks.

Quote
dc42

Quote
dc42
If you are using my latest firmware 057t - dc42 then you can reprogram the thermistor resistance at 25 C and thermistor beta. See the checkin comment for details.

You can find 057t-dc42 and other revisions at [github.com]. From there, follow the link to the file you want, then click Raw to download it.

Quote
rayhicks
@groneorm, the release notes that describe setting the R₂₅ and beta values are here: [github.com] (scroll down the readme until near the end)
Ray

Well the .bin files has never been a problem finding. Have upgraded to 057t-dc42 now..
It's how to use the thermistor correction values I'm very curious about..

The only information I can find is this:

- Implemented M301 (hot end PID parameters) and M304 (bed PID parameters) commands. Extended these commands to allow setting of thermistor
  resistance at 25C (R parameter) and thermistor beta (B parameter). A negative P parameter means don't use PID, use bang-bang control.

What's the syntax?? Say for example that I calculate my corrected beta value to be: 4100

M301 ????


I have checked with a couple of resistors in place of thermistor and Duet reports:

R	C
100K	22.7C
39K	45.9
33K	50

Did not find any multimeter atm so I couldn't confirm the exact R values. Resistor connected to HE wireloom.

Seems to be off about 10% at 100K then?

The other way around.. Room temperature at 25C, Duet reports 22,7 to the web i/f.

So I guess that thermistor is about right but Duet is somehow getting it wrong?

So little time, so much to learn..

Edited 2 time(s). Last edit at 03/11/2014 10:15PM by groneorm.

@groneorm, you can find the syntax for gcodes at [reprap.org]. I have added R and B parameters to the M301 command as I said in the commit comment. So if you just wanted to change the beta to 4100 you would use M301 B4100.

However, it now appears that the hot end temperature reporting error error at low temperatures is probably caused mostly by offset and gain errors in the ADC. See this thread [forums.reprap.org]. So I intend to add more parameters to the M301 command to allow these to be corrected.

---

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].

@ groneorm, I think that knowing the exact resistor value would be useful, it would also be be necessary to know what the open-circuit voltage of the thermistor connector is, and what that voltage becomes when the known resistors are added. It looks like your major problem is one of gain, and a three-point measurement would be useful (using resistors of 100k, 10k, and 1k or lower (the thermistor is around 750-250 ohms at PLA to ABS temperatures) - so if you can get use of a multimeter please do. The information would be useful feedback for any calibration approach - remember that for now you can just set your hotend to a higher than expected value to print.

@dc42 - are there any spare ADC pins brought out on the expansion header? it should be possible to design a small plugin board with a 0.1% resistive divider ladder on it with a couple of taps as in this app note [www.atmel.com] then poll this on startup to get the data for a calibration routine, rather than having people guess or have to experiment to find what corrections their board needs. Alternatively, people could hook up their own close tolerance resistors to one of the existing inputs , monitor the temperature readout and use a spread-sheet to determine the parameters I guess. This second approach presumes that the ADC errors are permanent, and from my observation on the t3p3 duet, it looks like they are, and if the resistors they plug in cover the range from 100k (25.3°C) to 220ohm (260.8°C) with a 6800(97°C) and maybe a 470(213.2°C) in between it should cover all eventualities (and even if they're 1% resistors, it should improve the response tremendously) - a linear curve fit between expected ADC reading and actual (which would have to be back-calculated from the temperature readout) should give the offset (as C) and the gain correction (as M) in a y=mx+c style fit that excel can do easily.

Ray

Yes there are ADC pins brought out onto the expansion connector. I suggest pins 37 and 38 which are AD10 and AD11. VSSA is not brought out on to the expansion connector, so you will have to use the pin 2 GND. There may be a slight offset in this ground voltage, or noise on it, because the ground it is connected to is at the wrong end of the PCB.

---

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].

Groneorm /KevM - dc42 has released a new firmware version which allow you to calibrate the temperatures using a 100k resistor for low temperatures and a 220ohm for high temperatures (though you could use other values along with the spreadsheet above) - please see this thread [forums.reprap.org]

Ray

Quote
rayhicks
Groneorm /KevM - dc42 has released a new firmware version which allow you to calibrate the temperatures using a 100k resistor for low temperatures and a 220ohm for high temperatures (though you could use other values along with the spreadsheet above) - please see this thread [forums.reprap.org]

Ray

Great stuff, will try it out! Haven't had much time for Ormerodding lately.. :-(

Quote
groneorm
Great stuff, will try it out! Haven't had much time for Ormerodding lately.. :-(

Is that legal?

Dave
(#106)

Quote
dmould

Quote
groneorm
Great stuff, will try it out! Haven't had much time for Ormerodding lately.. :-(

Is that legal?

Dave
(#106)

Think so!?

Anyway, the last testprints is starting to look good. Will try some different slic3r settings and do some more testprinting before final verdict. It sure looks promising so far, AND the filament slipping seems to be completely gone!

Thanks for all help and especially big thanks to DC42 for a great firmware! I settled for:

M301 L-26.0 H10.1

..using the the resistor method to calibrate the hotend temperature reading described here (Method 1b)