Thermocouple K 2000º when extrude move

FBG

Hello,
i have DUET wifi with 1.20 version
I have thermocouple daughter(Version 1.1) with FAULT led version.
I have external drivers for X/YZ, for E0 and E1 (same heater) i use inside drivers of duet.

The problem is easy, when move X/Y/Z not problem, when move Extruder 0 or 1 the temperature show 2000º!!!. Also FAULT LED is ON.
When stop move, the temperature return again to normal values!

Ideas?

I check the wires and i test another thermocouples, but the fail happen always.
The thermocouple be electrically isolated from the hot end heater block.

dc42

The thermocouple leads are evidently picking up noise from the extruder motor cables. Do you have twisted pair thermocouple wire running all the way back from the thermocouple to the daughter board on the Duet? That's very important, to minimise inductive interference pickup and to ensure that the cold junction is at the daughter board.

If you are already doing that, you may need to move the thermocouple wires away from the extruder motor wires, or shield the thermocouple wires from the extruder motor wires, or perhaps use twisted pair (preferably shielded) cable to connect the extruder stepper motors to the Duet.

FBG

All my wires have shield, and all shields is connected to ground.
So where can be the problem?

dc42

OK; but Do you have twisted pair thermocouple wire running all the way back from the thermocouple to the daughter board on the Duet?

FBG

Yes absolutely, and also is shield. The shield is connected to ground.

FBG

Fixed!!!! i remove the shield of the steppers E0 and E1 and work perfectly!!

I do a comparation of temperatures between this thermocouple K and Thermistor of the bed with Duet, and i see than the Thermocouple K show 6ºC more than the bed thermistor.

This also happen with arduino and marlin firmware, in marlin i know like adjust with value.

//These defines help to calibrate the AD595 sensor in case you get wrong temperature measurements.
//The measured temperature is defined as 
//"actualTemp = (measuredTemp * TEMP_SENSOR_AD595_GAIN) + TEMP_SENSOR_AD595_OFFSET"
#define TEMP_SENSOR_AD595_OFFSET -4.0
#define TEMP_SENSOR_AD595_GAIN   1.0

But here, how?

dc42

There is no provision to adjust the thermocouple reading, because the digital interface chip (MAX31856)is guaranteed to be accurate to within better than 1C. It is more likely to be the thermistor that is inaccurate. Analog thermocouple solutions such as the AD495 are far less accurate.

FBG

oh, fantastic
A thing, you say: MAX31856, but i can read MAX31855 in the chip

dc42

If it's the version with the FAULT LED then it should be MAX31856. What X value are you using in your M305 command to select it?

FBG

M305 P1 X150

dc42

In that case it definitely uses MAX31856 chips. I expect part of the 6 has rubbed off to make it look like 5.

FBG

Yes is clarify
MAX3185
[h]6MUD[/h]