Setting precise temperature for chocolate printer

benecito

Hey,

For printing chocolate I am trying to get temperature as precise as possible. In a perfect world the actual temperature would be within 0.2° of the temperature shown and the temperature could be set in steps of 0.1° for better control.

With some time spent on tuning and tweaking the PID values I managed to get the temperature to vary by less than 0.5°. Measuring with a calibrated thermometer I had a deviation of something like 1.5° depending on the temperature sensor I use. Is there some way to correct / offset the temperature reading to the actual value?
Also is there a way to set temperatures in steps of less than 1°?
Both things I did not find anywhere so far.

Phaedrux

For best measuring accuracy a PT100 sensor would be appropriate. As for resolution of setting the temp I think it would need to be changed to get finer control since setting print temps of plastic is not nearly as picky.

I've changed the title and moved to firmware wishlist.

think you may be able to correct offsets with the H and L parameters of https://duet3d.dozuki.com/Wiki/Gcode#Section_M305_Set_temperature_sensor_parameters, no idea of the resolution.

dc42

@benecito said in Setting precise temperature for chocolate printer:

Is there some way to correct / offset the temperature reading to the actual value?

If you use a PT100 sensor with a 4-wire connection then the readings should be accurate to better than 1C. Of course, the temperature at the nozzle may not be quite the same as the temperature at the sensor.

Also is there a way to set temperatures in steps of less than 1°?

Yes, in RRF the G10, M104 and M109 commands accept floating point values for the temperature parameter.

benecito

Thanks a lot to all of you!
PT100 Sensors have the problem of all of them being quite "big", but I am in contact with a manufacturer who might be able to supply a small enough Version of it.

Right now I am using the H and L parameters to finetune my thermistors, but this is a little frustrating to do as you can't just measure the temperature difference and enter it but need to work your way there. At least my thermistors all vary so I'd need to tune all of them.

If the sensor temperature is not the same as at the nozzle it does not matter as the deviation should be the same on every printer if the sensors (probably PT100) don't vary.

@dc42 said in Setting precise temperature for chocolate printer:

Yes, in RRF the G10, M104 and M109 commands accept floating point values for the temperature parameter.

You made my Day! I did not find any documentation this is possible and stupidly did not try.
But this leaves me with one Question: Is there a way to set it at steps 0f .1 of a degree? I dont want my customers to have to use the m104 command all the time.

Do you happen to know if there is a way to set them using a repetier server? (or using PanelDue)