Temperature Bias?
Is there any way to manually set temperature calibration points? I'm seeing a 5 to 10 degree C bias. Both the heated bed and the extruder report room temp as 30 C, while the thermostat (and my body) tells me that the room is at 20 C.
Just assembled my the MK2S. I upgraded the firmware to 3.0.11. I've done Temperature calibration, but that just appears to use the hotend to measure the heated bed's temps.
I'm seeing some part warping and some bed separation when printing objects larger than roughly 4 sq cm footprint. I've had successes with 2 cm cubes. I'm wondering if the temp bias is causing the heated bed to operate at too low of a temp during printing.
Any suggestions?
Re: Temperature Bias?
Dont worry about what reading ambiently. Remember the temp sensors are inside enclosure type devices such as the heat block and under keflon tape for the bed. As long as you get steady readings, you're fine.
Re: Temperature Bias?
I'm seeing also seeing a bias in temperature readings at room temperature, similar to Andrew's problem, though mine is only six degrees off. My room is steady at 24 degrees, and the printer, after being turned off for quite a while, reads 30 degrees right after turning on, for both hot-end and heatbed. With the printer turned on, both temps rise to 31 and stay stable there, which I'm not concerned about; it could be self-heating of the thermistors.
Is there a way to change the sensor-to-temperature curve in software? (This would be independent, I'd think, of the PID tuning -- PID assumes that sensor readings are accurate, which these are not.)
Cory, I'm not sure what you're saying about the sensors being inside enclosures -- if the printer has been powered down for hours in a room where the temperature is stable, the temperature *should* be the same inside as out.
Re: Temperature Bias?
To answer my own question 🙂 it looks like in the firmware there is some kind of temperature reading offset for the bed thermistor, but the values are hard-coded. And there is nothing like that for the hot-end temperature.
😐 Based on the few minutes of code reading I've done, it looks like generating a new table in thermistortables.h and recompiling the firmware would be the only way to fix this.
I might do that later, but I guess for now I'll set temperatures six degrees higher than the recommendations as a starting point.
Re: Temperature Bias?
Looks like my worry was not needed. I found this post from Jeff Jordan: http://shop.prusa3d.com/forum/original-prusa-i3-mk2-f23/idle-temperature-too-high-gt-30-c--t3420.html#p29683 and checked it out with an IR thermometer. Yep, the error is too low to be sure there is any error at higher temperatures, so I can relax now. 
-Ticc