Heat Bed temporature sensor is not working

[ccholas]

I have a Rostock Max. I accidentally scraped the thermistor off a couple weeks ago while I was adding my aluminum heat plate. The heat bed was not damaged in any way. I ordered a new thermistor from http://shop.seemecnc.com/100K-Thermistor-26141.htm. I soldered the new thermistor into the board. I turned on the system and for the first few minutes the thermistor read 20 degrees Celsius, because it was at room temp. But quickly drifted to 163 Celsius at room temp. It has been there ever since. I have unsoldered and resoldered two new thermistors to the board. I have verified the resistance of those resistors to be ~86k and I have verified that the resistance goes down as the temperature rises. Two questions: 1) I am concerned that for some reason the A/D input on the Rambo board was nuked through this. Is there a parameter in the code that I can change the thermistor to one of the other inputs? 2) I am concerned that I am not using the right profile in the code for the thermistor that I have purchased above? I am using profile 5 for this device. It was the default with the original thermistor. I cannot find a part number of this device nor can I correlate it to the code in any way.

Help????

[aehM_Key]

I have verified the resistance of those resistors to be ~86k

It should have ~100K @ Roomtemperature.

that the resistance goes down as the temperature rises.

I'm not sure, is it a NTC oder PTC?

[ccholas]

I don't know which one it is, because I don't have any information from the SeeMeCnc website. Do you have a recommendation on a better thermistor part that I should use?

[mhackney]

The RAMBo has 4 thermistor inputs so you can try one of the others. I have information in my build thread since I actually blew the original heated bed thermistor port by shorting it. It is in pins.h in Repetier firmware:

Code: Select all

#define HEATER_BED_PIN 3
//#define TEMP_BED_PIN   2 
// I popped the thermistor input on my RAMBo :(
#define TEMP_BED_PIN   7

I vaguely remember that I had to change something else too so it would be wise to look at my thread.
EDIT: no, that's all you have to do: My Thread

Here is the thermistor entry I use for SeeMeCNC's thermistor:

Code: Select all

#define EXT0_TEMPSENSOR_TYPE 97

[ccholas]

So I have been troubleshooting this for the past week. I suspected the Rambo board and the temperature sensor. I ordered a new Rambo board because I was seeing weird behavior on the other inputs. I replaced the thermistor (semitec brand, and the part number is 104GT2 They are 100k @25c ) and ensured that there wasn't a short. I installed the new Rambo board and reprogrammed it. The problem I have is at room temperature the thermistor reads 30 degrees. I turned on the heat bed and as expected it slowly rises in temperature. Until I get to ~40 degrees the read jumps to 131 degrees and continually ranges about ~20 degrees. I have tried all the different thermistor profiles in the Marlin code base for the 100k thermistors. I have double checked every connection between the thermistor and the Rambo board.But I get the same behavior. I am drawing an absolute blank now.

Any ideas on what to do next?
How can I profile a new thermistor profile that matches the thermistor behavior I am seeing?

[mhackney]

I suspect the Onyx might be the root of the problem, perhaps the wiring or the board itself. I would suggest removing the thermistor from the onyx. Solder 2 leads to it so you can hook it up to RAMBo but keep it separate from the Onyx. Get a glass of hot - not boiling - water and put the thermistor in it while you monitor the temp in your host - DO NOT turn on the heated bed power! Does the temperature rise and then stabilize at a reasonable temp? If so, it is not a problem with RAMBo or thermistor.

If that worked, now tape (Kapton is best but even blue painters will work) the thermistor to the top of the onyx, maybe 1 cm offset from the middle hole. Now set the Onyx temp to 50°C and start it up. What happens?

Report back and we'll take the next step.

Choosing the wrong thermistor calibration curve in the firmware is not going to give erratic behavior like you describe, so let's get to the root of the problem first.

[ccholas]

Sorry I have been traveling a lot lately and have not had time to get back to my 3D printing until last night. The Rambo board was fine, I did verify the inputs on the Rambo and thermistor. It was showing weird readings with the Onyx. So I wired a thermocouple directly to the Rambo without the Onyx in the picture and noticed it was giving stable readings. I measured the resistance of the Onyx board for the thermistor and the resistance(heat element) and noticed that it would change as the board flexed. I think that there was a short somewhere in the board itself that was causing the issue. As a note, I put a .5mm heat pad under the aluminum to help distribute the heat better and cover the vias that were exposed on the top of the board. The traces were not shorted by the heat plate at all. The board was faulty. I ordered another Onyx replacement board and the problem went away.

Thank you all for your suggestions. They really helped by to debug this issue.