Wrong temperature values returned for onboard NTC100K sensor

[lemmingDev]

Hi

When using the onboard NTC100K to check the temp with:

[temperature_sensor Warehouse]
sensor_type = ATC Semitec 104GT-2
sensor_pin = sb2040:gpio26

the value is too high.

When I set
pullup_resistor: 10000
then the value is better

Did you guys put a 10k resistor instead of 4.7k?

Thanks

[djessup]

I have the same issue. A board I bought a month or two earlier worked as expected with a default (omitted) pullup_resistor value.

[arielbernal]

Same issue here.... the NTC100K values are way too high, similar to the mcu temperature.
I would be nice if Mellow could answer the question

[pbutler]

I am having the same issue... but even 10k pullup doesn't seem to be 100% right for me. Room is ~16C but the thermistor in the unheated chamber reads as 22C

[lemmingDev]

I gave up and put a standard hotend thermistor in the z drag chain connected to the Octopus board directly to use for the enclosure temperature sensor

[arielbernal]

I gave up and put a standard hotend thermistor in the z drag chain connected to the Octopus board directly to use for the enclosure temperature sensor

I gave up as well, I used the z drag chain as too

[djessup]

@Mellow-3D Could you please give some input on what is going on with your SB2040 thermistors here?

[Villiem]

Same problem

[VooDooSi]

I have the same problem. I dont really care for onboard thermistor, problem is i also get negative temperature for extruder. So far i flashed the board, connected extruder thermistor, fans and leds. Even with everything disconnected all the temperature readings are negative.

Solved, at least for me. I looked at the board under the microscope and noticed a few blobs of solder just at adc pins. Some flux and soldering iron, board is now working ok.

[pbutler]

@VooDooSi: I came across this document someone posted either to the Voron forums or reddit that mentions negative temps: https://docs.google.com/document/d/1vqLZFZ6slT3QIw_dWkraFHuKx9kYC-jz_46KikXdVu4/edit

[VooDooSi]

@pbutler, thank you. I saw this already yesterday when i flashed the board without connecting anything else and i got negative temperature for pcb ntc. I didnt even come to connecting endstops :)

Solved, at least for me. I looked at the board under the microscope and noticed a few blobs of solder just at adc pins. Some flux and soldering iron, board is now working ok.

[muditguptacode]

I initially thought I was good after switching the pullup value to be 10000 but no, the values "look better" but still don't seem to reflect the actual temperature.
@Mellow3D do you actually intend to address this in any way or provide a potential fix? I'd take a scope to the board myself and diagnose but I really don't want to disassemble my printer at the moment.

[thiagolocatelli]

@jaysuk Do you mind helping the community and answering to this issue?

[jaysuk]

i would if i knew the answer

[thiagolocatelli]

@jaysuk I saw that you are the one made all the commits to this repo, so I apologize if you are not the correct person. If you work with Mellow, can you please reach out to Mellow and have someone help with this issue?

[jaysuk]

i don't "work" for Mellow, just look after their github etc. i'll nudge flymaker again to try and get a response