[Solved] All PT1000 sensors showing 13 degrees higher
Happens to every PT1000 sensor, temperature is normal when I use 104NT-4 sensors
31 Replies
rare-sapphire•2y ago
same here , hope someone will help
did you set your dip switches on your control board for the PT1000? 1000 is not a direct plugin replacement without settings changes
I use Octopus Pro board, which supports plugging PT1000 into the thermistor slots directly.
yes but you need to set SW3 OFF (1,2 and 4 stay on ) read your manual
I too use an octopus pro 446
section 4.5 of the users guide
Like so?
yes
pt1000 is less accurate at lower temps but more at higher as well as able to read higher temps. These switches change the resistance values on the bridge that reads the sensor
after a reboot...did your reading change to something much closer to expected?
Do I need to reboot the whole pi or just Klipper?
not sure. You could try just klipper but the board might need to see a cycle to re-read those switches
I mean it didn't change after firmware restart. Let me try rebooting the pi
Rebooting the Pi also made no difference
power cycle is all that's left. I'm not sure how else to have the BBT board re-read the switches (not sure how Klipper firmware works o tneh board)
I did power cycle it
sensor on the connector center two pins?
pages 19-21
I use the connector shown on the RatOS wiring diagram
did your reading change? or still the same?
?? https://os.ratrig.com/assets/files/octopus-pro-wiring-full-3d9d989a41707c83aaebfd75f2ed3846.png doesn't even show the hotend ot therm. Go by the board manual for that
Many others are wiring the PT1000 this way, and they had no problems
I do it here
dumbass me! Sorry David...I dumped those last message.... I was mixing boards in my head even though I was looking at the docs. I just looked at my own board when I started questioning my own comments. Yes indeed your wiring is i nthe right place. And again.... did the reading change with the switch change?
And now you have me questioning the oem RatOS configuration... as the hot end thermo is configured on the Octopus board as PF4 which is where you have your sensor. The connector for the PT1000 using the switches uses the max31865 chip and uses those switches. So how is MS (author/maintainer) handling the PT1000?
I'm headed over to my printer now to flip SW3 and see what changes (if anything)
No, the switch did nothing, it is still showing wrong readings
Just did a whole bunch of digging.
Looking through the example config files, it's not obvious, RatOS is even using the MAX chip or that socket at all! The Hotend sensor (E_sensor_pin)is on PF4 which is that socket you are plugged into. The MAX chip uses PF. So the question arises.... how in firmware is Klipper connecting the correct resistance to read the PT1000 correctly (if it does at all). In that latter case, it would explain your errored reading. BUT, then I have to wonder why min reads correctly.
PT100 uses a 430 ohm and PT1000 would be a 4300 ohm resistor for the MAX chip to read and use the full range scale of the Pt1000 (vs PT100) . This 'scheme' is the same on a Duet3D board (changing values for the bridge reading the sensor) So I guess I need to dig into the firmware if I expect to understand what the connection of PF4 vs the MAX chip pin connections is. 
asked this over in Klipper
https://discord.com/channels/582187371529764864/750831417738395748/1064052285828775956
asked this over in Klipper
https://discord.com/channels/582187371529764864/750831417738395748/1064052285828775956I added pullup_resistor and set it to 4700, made no difference
Because that is the default
I just changed to another thermistor pin and it seems like it is around the correct temperature now.
I just cannot believe this.
That bad pin still works with 104NT. Thank you all for helping. Problem solved!
Appreciate this thread - this just helped me solve the same exact issue - in the same manner... PF5 shows temp about 15C lower than PF4 - bizarre