Heatbed temp is 15 deg lower than what printer says
When I measure surface of my heatbed with a high quality IR thermometer (fluke), it shows 96 degree, and printer says 110 degree.
in the corner areas its even less (90 degree).
Fluke device has 1.5 deg precision. Printer is installed in a housing (IKEA model).
Is there any way to compensate that offset in the printer settings so that heatbed will be more accurately heated?
RE: Heatbed temp is 15 deg lower than what printer says
I was concerned about low readings too, when I got a digital infrared thermometer. There's not as much difference in mine as there is in yours, but I was still concerned, since a low 90s measured vs. a 100C set and indicated seems like a lot of error.
When you think in terms of Kelvin degrees instead of Celsius, the error is far less significant. My error turns out to be less than 2% of the absolute (Kelvin) values, and I don't think that's enough to be concerned about.
You can indeed change the bed temperature in the filament settings on your slicer. If you want to goose it a few degrees, just type in what you think will compensate for the error.
RE: Heatbed temp is 15 deg lower than what printer says
...
When you think in terms of Kelvin degrees instead of Celsius, the error is far less significant. My error turns out to be less than 2% of the absolute (Kelvin) values, and I don't think that's enough to be concerned about.
You can indeed change the bed temperature in the filament settings on your slicer. If you want to goose it a few degrees, just type in what you think will compensate for the error.
question was if there is a general offset adjustment for heatbed temp, so that the readings will reflect true temperature, so that filament settings would work without all being modified to reflect that big temp offset in my printer (which is by the way brand new delivered by prusa).
to judge the impact of a 20 degree temp error, I would not look at kelvin range but to the fact that it is a 20 % error based on the temp range the heatbed is designed to provide, and that results of first layer adhesion and warping could suffer significantly due to such differences.
RE: Heatbed temp is 15 deg lower than what printer says
Markus,
did you measure the upper surface or the lower surface where the thermistor is?
Regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: Heatbed temp is 15 deg lower than what printer says
Markus,
did you measure the upper surface or the lower surface where the thermistor is?
Regards Joan
upper side on top of the print bed
RE: Heatbed temp is 15 deg lower than what printer says
And the sensor is on the underside of the bed, could there be a thermal gradient between the sides?
check the results as close to the thermistor position as possible so that your thermometer view is not picking up interference from any unheated parts.
what is the measured and reported temperature at room temperature? are they also significantly different?
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: Heatbed temp is 15 deg lower than what printer says
And the sensor is on the underside of the bed, could there be a thermal gradient between the sides?
This is one thing I certainly considered. I'm sure that the temperature at the thermistor is indeed within tolerance.
I also get the impression that there's a fair amount of wiggle-room regarding temperatures in 3d printing. It's not as critical as, say, temperature when developing color film.
RE: Heatbed temp is 15 deg lower than what printer says
When I measure surface of my heatbed with a high quality IR thermometer (fluke), it shows 96 degree, and printer says 110 degree.
in the corner areas its even less (90 degree).
Fluke device has 1.5 deg precision. Printer is installed in a housing (IKEA model).
Is there any way to compensate that offset in the printer settings so that heatbed will be more accurately heated?
As covered in one of the Tom Sanladerer episodes, the emissivity of the PEI surface may be throwing off your IR measurements. Try measuring using a separate thermocouple directly attached to the heatbed and see how it compares.
Also, according to Tom, the cooler temps at the printbed corners is a known problem with the prusa design. In addition, he found that an all aluminum headbed, like what the Ender 3 uses, has a far better temperature uniformity.
RE: Heatbed temp is 15 deg lower than what printer says
I also get the impression that there's a fair amount of wiggle-room regarding temperatures in 3d printing. It's not as critical as, say, temperature when developing color film.
Yes and no. It depends on how close the temperature is to the margin of the working range. At some point it's going to matter.
RE: Heatbed temp is 15 deg lower than what printer says
When I measure surface of my heatbed with a high quality IR thermometer (fluke), it shows 96 degree, and printer says 110 degree.
in the corner areas its even less (90 degree).
Fluke device has 1.5 deg precision. Printer is installed in a housing (IKEA model).
Is there any way to compensate that offset in the printer settings so that heatbed will be more accurately heated?
As covered in one of the Tom Sanladerer episodes, the emissivity of the PEI surface may be throwing off your IR measurements. Try measuring using a separate thermocouple directly attached to the heatbed and see how it compares.
Also, according to Tom, the cooler temps at the printbed corners is a known problem with the prusa design. In addition, he found that an all aluminum headbed, like what the Ender 3 uses, has a far better temperature uniformity.
Here it is:
He sprays his PEI build plates with black paint to avoid the emissivity IR measurement issue. Otherwise, as in your case, they would measure cooler than they really are.
RE: Heatbed temp is 15 deg lower than what printer says
What did you set the emissivity to on your fluke IR?
The Filament Whisperer
RE: Heatbed temp is 15 deg lower than what printer says
I'm seeing a similar problem although the temperature delta in my case (21-25%) is much worse and I measured directly on the heatbed using the thermocouple that came with my multimeter.
After reading this thread I repeated the temp measurements after attaching the multimeter thermocouple to the heatbed underside close to the thermistor and the readings were similar.
Below is what I e-mailed to Prusa recently. While I wait for a reply perhaps someone here has some suggestions.
Something I should have mentioned in my e-mail is that I was able to confirm the multimeter temp readings using a digital cooking thermometer with a metal probe.
------------- e-mail ----------
Hi
I noticed that the bed temperature in my printer is off by several degrees. I’m not really sure when it started but it’s probably been a while. I noticed because I was troubleshooting an adhesion issue with a particular PLA spool.
For example when the bed is supposed to be at 60oC I measured 48oC using the temp probe of an ExTech multimeter by touching the hot plate without the metal sheet and leaving the probe touching the hot plate until the temp reading stabilized.
When set to a 100oC I measured 82oC using the same method.
I followed your troubleshooting guide and the HW looks OK
- Heat element resistance was 3.2 Ohms
- Heating element is supplied with 24V
- Thermistor resistance at 25oC was 100k Ohms
- I set the bed to 60oC and after the printer reported reaching the temp I left it there for several minutes. Then I turned off the printer and immediately after removed the thermistor connection from the board to measure the resistance and it was ~32k Ohms which maps to ~50oC for an NTC 4250 thermistor (according to its datasheet assuming this is the one being used in this printer) which is inline with what I measured with the multimeter.
- There’s no bed damage
- The thermistor is touching the bed without any sag
FW version 3.9.3-3556
What do you suggest to do? Do I have to replace the thermistor? FW issue?
Thanks
RE: Heatbed temp is 15 deg lower than what printer says
I'll be interested in seeing the response, also what model & year printer did you test?
The Filament Whisperer
RE: Heatbed temp is 15 deg lower than what printer says
I'll be interested in seeing the response, also what model & year printer did you test?
MK3S. Shipped on early May/2019
I've been looking at the FW and I think at this point the issue could be that my thermistor is out of spec (for whatever reason) or there's a FW bug. Some details:
- Removed the temperature offset in the heat bed conversion. The PINDA one still uses it though.
- Regenerated the lookup table using a different value for the Beta coefficient using the specs for thermistor TDK NTCG104LH104JT1. The TDK thermistor is a chip thermistor with a narrower temp band so I don't think is the one used in these printers but its Beta value will generate a mapping that matches very well with the temps measured by the thermocouples for the [25,100] range so there's that (used this script from the Reprap repo):
python3 ./createTemperatureLookup.py --r0=100000 --t0=25 --r1=0 --r2=4700 --be
ta=4485 --num-temps=60 --max-adc=1023
I've asked Prusa to share the raw thermistor ADC values from one of their printers to compare with what I'm seeing. Also asked for the datasheet/thermistor part# that was used in my printer to find out if my thermistor is out of spec or the lookup table is the incorrect one.
I guess that for now I'll use the custom FW since the issue is essentially solved with the changes I've mentioned (I've yet to print anything though) although now I want to check whether something similar is happening with the hotend thermistor which I think is the same part or a similar one.
Cheers