Thread Rating:
  • 1 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
E3D V6 Thermistor Table Gen 1
#1
I've just installed an e3d hot end on the davinci, and the print quality difference is just massive. However I spotted the temperature is out. All the references I could find to a custom thermistor table for the davinci seemed to be for gen0 boards, and the solution was to remove a resistor and use the standard table 8.

I've had a look at the gen1 board, and the thermistor seems to be connected like this --

VREF ---- 1KR ---+--- Termistor ---+-- GND

With some kind of capacitor / resistor based filter in there with a 100R resistor. So quite different to the gen0 setup.

I fed the numbers from the 104GT-2 datasheet through some appropriate equations to generate ADC values and then graphed against the Excel sheet that comes with the firmware and got this:

   

It sorts the 4-5 degree temperature drift I see at room temperature and otherwise follows the standard table pretty closely. Which is precisely what you want, as its shown the standard table matches the 104GT-2 pretty well.

The table is:


Code:
#define NUM_TEMPS_USERTHERMISTOR2 36

#define USER_THERMISTORTABLE2 {{306, 2400}, {349, 2320}, {399, 2240}, {458, 2160}, {527, 2080}, {607, 2000}, {701, 1920}, {811, 1840}, {938, 1760}, {1084, 1680}, {1250, 1600}, {1438, 1520}, {1645, 1440}, {1872, 1360}, {2112, 1280}, {2360, 1200}, {2611, 1120}, {2854, 1040}, {3083, 960}, {3290, 880}, {3471, 800}, {3624, 720}, {3749, 640}, {3847, 560}, {3922, 480}, {3977, 400}, {4017, 320}, {4045, 240}, {4064, 160}, {4076, 80}, {4084, 0}, {4089, -80}, {4092, -160}, {4094, -240}, {4095, -320}, {4096, -400}}

Then use table 7. I'll give it a go tomorrow as its late now, but that looks like it should work.
Reply
#2
Its reading low at ambient temp now compared with the bed (13 vs 20). I'll check the ambient against my meter with thermocouple tomorrow and see what the score is, as I know the accuracy of the temp tables is not as good low down. I didnt feel like 13 degrees, though, so its not spot on I dont think. But the difference in the low range on the ADC is pretty miniscule with each integer ADC increment not being far off 1 degree C.

I know its all academic really, as you fine tune your temps, but I would like to know the reading is actually reasonably close to reality. Smile

In fact the curve is pretty much an exact match to table 5 using an 850R resistor in the equation for the voltage divider with the thermistor. It doesnt very more than a couple of degrees over the range, but interestingly does drop a couple of degrees around 20 deg C.

Its certainly interesting the thermal response of whatever thermister is in the DaVinci is pretty much identical to the E3D's.
Reply
#3
Hey Gizmo,

did you find time to do the thermocouple checks on the hot-end? I've just installed an E3D v6 with the supplied cartridge thermistor and I left the default temp table (table 5) in place. On settling down after power-on, the hot-end reads around 30 degrees C. I think this is similar to the reading, at rest, of the stock thermistor, The readings don't seem to be far off but don't have the tools to measure the actual temperature.

To verify my understanding of how this works, I generated a table with my own math and came up with something very similar to your own table though it drifts a little at the low end of the temperature range. Thanks for posting your work above, it was great to find my math wasn't out in left field!

Initial print tests were fine, but I decided to strip and clean the printer mechanics and have yet to re-assemble and re-calibrate so no definitive results yet. One issue I have to deal with is the warping of the heated bed. The centre is ~0.03 mm lower than the edges causing problems with bed adhesion.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)