Hot end thermistor calibration

[texsc98]

I've tried searching the forum and can't find a mention of how to do this other than mhackney stated he would post one up in the future but I can't even find that original post now, lol... But does anyone have a link or quick instruction on how to calibrate the hot end thermistor temp in the firmware to match real world temp?? I assume this would involve modifying the thermistor table, but I do not have any experience in that realm.

Since I first built the MAX back in January I knew that I should've went ahead and measured my real world hot end temp against the software calculated temp, but figured since I was getting decent prints (or so I thought) that it must be close... BOY WAS I WRONG! I have been printing at 245C (software) basically since I got the printer, but when I measured it this past weekend I was floored... 245C in software on my machine equated to 192C with a thermocouple! So I've been printing ABS at down in the PLA temp range, lol! If I didn't have Steve's geared extruder and fought it into submission to push that filament through 30C under where it should be with such force I probably would've figured this out a long time ago. Currently I've determined to print good with ABS I have to print at 293C in repetier/gcode, but when measuring the hot end between the no22le & heat sink I get ~230C. This is with seemecnc filament.

So obviously either the thermistor is not to spec, or it has to do with my mounting of the thermistor. But either way it does at least read stable/solid and I can make better quality prints now. Before my prints were extremely stringy, and now I get very little stringing. But it would be nice to have the correct temp set in the gcode rather than a "fake" temperature to make it match what I want in real world temp, lol

[cambo3d]

http://forum.seemecnc.com/viewtopic.php ... 259#p10259

ACCURATE TEMP READINGS:

*** Works with the Repetier Firmware only ***
So, as many, many people have experienced, thermistors have a tolerance, and can even vary outside their tolerance ratings. Different lengths and routing of wires can even influence the results. Also, you have probably learned by now that the temp you print at can have a big impact on print quality as well. So, in order for us all to be able to share our tips and secrets of our successes, we need to be sharing ACCURATE temperature data so we have a more even comparison. In the past, we've kind of just used the method of measuring what our actual is, and comparing it to what the machine 'thinks' it is, and just knowing that, and compensating our gcode etc... for that difference. It's worked, and still does, but to get more people sharing accurate information, lets learn how to eliminate that variable altogether. Here's what you need to do. Using a contact type thermocouple, like what comes with some digital multimeters, heat up your hotend to 200c, and unscrew the nozzle just enough to slip the measuring point in between it and the aluminum. You can just snug the nozzle to allow it to hold the thermocouple in place. Now, take the actual temp reading. If your meter reads in fahrenheit, use the following formula = (Celcius X 9 / 5 +32 = Fahrenheit). Now use the actual and reported temps to figure out if you are reading higher than your actual temp, i.e. Repetier says 200, but your thermocouple says 184 (184 / 200 = 92%) and come up with what percentage you are off. In the example, we are 8% too low. Go ahead and open up Arduino, and open up your firmware. Go down to line ~454, and look for "#define GENERIC_THERM1_BETA 4267" Now, if you are reading 8% too low, then you need to INCREASE the value by 8%, in this case we would change it to 4608. Don't worry about decimals, and don't use them actually, just round up. It's a good idea, just in case you forget, to add comments describing what it origionally was, by adding to the end of the line something like
" #define GENERIC_THERM1_BETA 4608 // default was 4267 " Now, save and upload your firmware to the board. Turn the heats back on to 200c, and check it out. You may need to iterate this once or twice, but it's WELL worth the time spent to get really accurate reading temps!!!
(P.S. in the last 4 machines i've calibrated, they all have been under reporting, and i had to set the beta value to around 4550, so maybe start there)

[Eaglezsoar]

Thanks for the great writeup. That kind of information is invaluable to all of us.

[texsc98]

I knew it had to be somewhere in the forum, I just couldn't find it

Thanks a bunch!!

[Eaglezsoar]

Is there a line in Repetier that allows the same thing to be done with the heat bed?

[JohnStack]

Mine was 4550 after switching to heat cartridges. Back in business with a stock hot end!

[Jimustanguitar]

I would also recommend dabbing a bit of thermal grease (the heatsink kind) on the thermocouple when doing this. This keeps it from being cooled by the air in the room. I'm not sure how much of a difference it makes, but it helps the placebo effect

Is there a line in Repetier that allows the same thing to be done with the heat bed?

I've been looking for this as well.