How do you calibrate the temperature?
I have a temperature probe Laser Meter
and when I check the bed and nozzle temps compared to what is set in my slicer/printer
they are quite different
Printable View
How do you calibrate the temperature?
I have a temperature probe Laser Meter
and when I check the bed and nozzle temps compared to what is set in my slicer/printer
they are quite different
I havent do e thus yet bit i know there is something called auto pid calibration
I just did a PID auto tune not too long ago. I just followed this:
http://reprap.org/wiki/PID_Tuning
and this as well https://www.youtube.com/watch?v=APzJfYAgFkQ
The PID calibration won't affect the temperature obtained - just manipulate various parameters in how the control loop reaches the set point, with the goal of minimizing things like overshoot/undershoot on the temperature obtained. PIDs are actually quite fascinating if you're into scientific math - http://en.wikipedia.org/wiki/PID_controller. I've got a PID temperature controller on my BBQ, and thought my engineering-background brain was going to explode the first time I read the PID details in the manual for it. Jeez. On a BBQ.
adamflip, good or bad, there really is no built-in way to calibrate the temperature. In theory, you can add new tables or replace tables in thermistortables.h to better map to the temperatures you think you have, but IMO you're on a slippery slope in doing so. The bottom line is that practically no one calibrates the temperature. If you do, you'll end up with temperature data that's only good for you, and won't necessarily mean that you're better-able to use manufacturer filament temp specs, etc.
Simply pointing an IR thermometer at a hot end, for example, doesn't mean you're accurately reading what's going on inside the aluminum block where it really matters.
Yeah, thermistors are generally not that accurate, and countless people especially bemoan the quality of the far-east stuff ending up in repraps. If there's a concern for how accurately a temperature is being measured, one thought would be to consider replacing the crude thermistor with a thermocouple and compatible thermocouple amp (such as with the -1 and -2 type devices in the configuration.h temp sensor table). They'll reportedly get you to within a degree or two accuracy.
BTW - the simplest off-the-shelf solution for a thermocouple approach that I am aware of is the E3D K-type thermocouple and the E3D thermocouple amp board, at a total cost of about $35 USD. There are other thermocouple boards from places like Adafruit and Sparkfun, but I don't believe they're as drop-in compatible with the Marlin baseline.
Although intended for use with the E3D hot ends, there's no reason why the E3D thermocouple couldn't be used on other hot ends as well.
printbus, Thank you, your replies are always quick and in depth, its very much appreciated