Thermistor compensation by temp messurement

Basic Information:

Printer Model:
MCU / Printerboard:
Host / SBC
klippy.log

Fill out above information and in all cases attach your klippy.log file (use zip to compress it, if too big). Pasting your printer.cfg is not needed
Be sure to check our “Knowledge Base” Category first. Most relevant items, e.g. error messages, are covered there

Describe your issue:

Trying to get my T-D500 thrmistor to behave, and are trying to find a way to compensate for errors when comparing to external thermometer (thermo couple connected to multimeter) .
It would me nice to have a secondary look up table for custom thermistors were you could simply map external calibrated sensor reading to klipper temp reading for compensation. Is this allredy a feature?

…

Did you even try to look in the docs? Custom thermistor definitions are easy.

You can define them in terms of resistance@temp OR ADCvolts@temperature.

I know that but that insn’t what I asked for :wink:

Maybe you could still use it but somehow add an offset to the custom temperatures to move the thermistor curve into the “right” range.

@stefan63 What kind of errors are you seeing?

You can correct the reading of a thermistor to mach an external sensor by modifying the “pullup_resistor” value, if thats what you are asking… I do that with a k themocuple glued in a drilled out nozzle..

Will try that, thank you!

Sensor readings in Fluidd UI is about 10-20 degree abowe the external calibrated thermo coupler test deice.

Calibrated how?

Thermocouples do not measure temperature. They measure the temperature difference between 2 points.

Is your thermocouple grounded or ungrounded? Does your thermocouple “reader” use a simulated ice junction or use a thermistor to measure the cold junction and the apply the delta temperature to the thermistor reading?

What hot end are you using? Is the heater on the same side as the thermistor or are they on opposite sides? Is your "test thermocouple closer or further from the heater than the thermistor?

Most important, using filament of known quality how well does a temperature tower corelate with the manufacturer’s specifications?

and filially if you read the ADC volts at several known temperatures defining a custom thermistor WILL shift your readings to your desired scale.

I don’t agree,

A thermocouple is a pair of electrical conductors made of different metals, connected at one end and suitable for temperature measurement due to the thermoelectric effect. In principle, the thermocouple generates electrical energy from heat when there is a temperature difference along the electrical conductor. The resulting electrical voltage at the other ends of the metallic conductors is called the thermoelectric voltage. It is relatively small, on the order of a few tens of microvolts per 1 °C temperature difference. Even at high temperature differences above 1000 °C and with selected materials offering high sensitivity, the achievable electrical voltages are below or on the order of 0.1 V.

Source: wiki Thermocouple

This is due to Thermoelectric effect - Wikipedia

I agree there are some (many) different factors involved.
I built a small chamber of a 12mm ID copper tube isulateded with a lot of glass fibre insulation where I could put a heater and two different T500 sensors + my multimeter thermocouple.
I use a lab power supply and cand by this way control the temperature and rate of change.
This way I have a clear view of what’s going on.
The T500 seems to be faster than my thermocouple but still when stabilized the T500 always show a higher temperature, and I have to compensate for it in my slicer.
Thats why I would like to have a lookup table to simply map (thermistor reading = 340 equals 320 in real world) etc…

FWIW,
The only way to define a custom thermistor is already mentioned: Thermistor compensation by temp messurement - #2 by cardoc

But there is no klippy log. I’m not sure if custom curve with:

value1 ~= temperature

is possible to define if some amplifier is used.
as for example, the MAX31856 computes the temperature directly according to the datasheet.
So, for example, in this case, there is no way to redefine the thermistor parameters in any way.

-Timofey

I guess you are not the first one having problems with T-D500. I think it is a thermistor not a thermocouple. Different principles compared to thermocouples.

Very often I read questions or problems here and I have a feeling they are old. I think, I read about it or remember I read something about it.
If you have a problem regarding 3D printers, always ask www first. The community is huge and might had that problem before. In this case g**gle might be your friend :wink:
You might read Dyze 500 thermistor · Issue #3182 · Klipper3d/klipper · GitHub.
Good luck.

How do you measure the output of the thermocouple? every time 2 different alloys touch a thermocouple is created. To get the thermocouple output measured, instruments with conductors of a different alloys are introduced into the circuit.

Don’t believe me attach a voltmeter to a thermocouple in boiling water. record the output in volts. Put the voltmeter in the freezer for 20 minutes and you get a different volts reading. Which voltage do you put into the equation to convert to 100 °C?

If you’ll read the Wikipedia page you linked you will see that (T) temperature never appears in Seebeck’s equations the term is ALWAYS (∇T). When you solve the equation the number you get is DELTA T.

JFI:

But no datasheet

Some info here: Hotends | DYZE DESIGN Docs

Because I see that I vas not clear on what I use to measure my nozzle temp: I use this:

Its the same principle used in checking the temp of the tip of the soldering iron

I dont use this because I really want my printers nozzle to be 200C when it shows in mainsail 200C, but because I have a toolchanger and I want all nozzles to heat the same.

Solving it by reading, thank you!