So to set the backdrop of this post, I was cleaning the hotend of a printer (Powerspec i3 Mini v2, which is a Wanhao i3 Mini rebadged and heated bed added) I picked up for cheap. There was some stubborn residue so I decided to warm the end. However, while it was still plugged in and warming I saw a spark while using the brass brush on the nozzle. Not sure where it actually sparked (thermistor or heater cartridge wires) but now the board constantly reads 193 °C for the hotend whether or not it is hot (or even plugged in for that matter.)

Looking around this seems to indicate either a bad thermistor or a messed up mainboard. The thermistor will read ~120k resistance at room temp and fluctuate based on temp changes (ice will cause it to go down, heat will cause it to rise), so can I take that to mean the thermistor is most likely still okay?

Doing some probing on the board, it seems the resistor that makes up the voltage divider on the thermistor circuit is out of spec (when probing it with a multimeter it reads only 470 ohms instead of 4.7k). Is this reading while in-circuit valid? Could this have been caused by a short in the thermistor? Can I build a small adapter to splice in a new 4.7k resistor? or will this just throw everything off more? If I were to build a new version of Marlin could I adjust something in there to get the temps back to something recognizable?

I figure since I got the printer for 40\$ I may as well use it as a learning experience.

Also, if anyone has the original firmware from this printer I would appreciate a copy.

  • $\begingroup$ Assuming the value of the 4.7k resistor is OK, it is in parallel with 522 ohms. The question is what does a good board measure in that location? Does the measurement change if you switch the leads? A difference indicates current through semiconductor junctions. That will also vary depending on the current the meter uses to measure resistance. $\endgroup$ – Perry Webb Jan 25 at 13:44
  • $\begingroup$ The heater is more likely to spark since it draws more current. $\endgroup$ – Perry Webb Jan 25 at 14:06
  • $\begingroup$ You can measure the heater voltage using a battery powered (floating) meter with the board powered up since it has significant current. The current on the thermistor is probably to low to get an accurate voltage with the board powered. Probably you would notice the sensed temperature change when you attempted to measure the thermistor voltage. $\endgroup$ – Perry Webb Jan 25 at 15:44
  • $\begingroup$ You could get a spark between a heater lead and thermistor lead. On some printer models this will damage an IC on the circuit board. $\endgroup$ – Perry Webb Jan 25 at 15:49
  • $\begingroup$ Indeed, that last tought is reflected in my answer, and it would be the pessimistic case. I look forward for feedback from @Samlii to exclude that. In case this board was damaged, I would probably go for some newer and more advanced board instead, but this means an extra investment in external LCD and stepper drivers. Fingers crossed. $\endgroup$ – octopus8 Jan 25 at 20:41

Yes, in newest Marlin (v2.0) there are configurable thermistor parameters. You need to set custom thermistor:

#define TEMP_SENSOR_0 1000

for your only extruder (suffix _0) in Configuration.h, and in Configuration_adv.h modify related section:

// Custom Thermistor 1000 parameters
#if TEMP_SENSOR_0 == 1000
  #define HOTEND0_PULLUP_RESISTOR_OHMS 4700    // Pullup resistor
  #define HOTEND0_RESISTANCE_25C_OHMS  100000  // Resistance at 25C
  #define HOTEND0_BETA                 3950    // Beta value

But in Marlin 1.x I couldn't find this section, so change would need to be done somewhere in thermistor definition source files. I mention this older Marlin following the discussion on Wanhao Duplicator i3 Mini & Rebrands Support, which refers to some Wanhao source code, but I cannot judge about value of it.

For resistors in the circuit I have similar observations: sometimes they do not give exact readings using multimeter, especially next to capacitor. Isn't heatbed temp probe's circuit the same? - can you compare? But instead...

I would suggest doing following things:

  • open circuit test: disconnect the thermistor and check readings (I could then see -19 °C)
  • closed circuit test: connect the heatbed's thermistor instead (in place of hotend's) and check if these readings are reliable (I would try to heat the bed for few seconds to check if changes are noticed, but not too long because it will be uncontrolled process)
  • thermistor test: connecting the hotend's thermistor as heatbed's probe and check on-screen readings for heated bed; are they again 193 °C? do they react to changes? (powering hotend for few secs or warming in fingers?)
  • you may consider to setup USB connection with serial terminal to directly read measurements using G-Code, just to exclude errors in post-processing of the temperature readings before putting it to screen (I know, it's a bit paranoic)

The worse thing I can imagine is that you could short wires of hotend heating circuit with the temperature probe circuit, which should be galvanically isolated. Then it could damage readings of the mainboard for this probe when received 12V/24V. I found that the mainboard is most probably all-in-one i3mini 0ne+ motherboard, some revision still available in Wanhao store for another $40. I googled for few photos, there are problably few versions of this board. If you are lucky and skilled then you may find some other pin (connected to ATMega) to reuse. Otherwise you would need to replace steering electronics with new board, which could be of great benefit for future improvements: more ports for extra fans, probes, more memory, better stepper drivers, etc.

  • 1
    $\begingroup$ Note: a capacitor will draw more current when it is charging (lower resistance measurement) then stabilize to its leakage current when reaching the voltage of the meter measuring resistance. Thus, if the capacitor isn't charged, the resistance will increase until it stabilizes, or if charged, resistance will be irregular until stabilized. However the 4.7k resistor should discharge capacitors if given enough time. $\endgroup$ – Perry Webb Jan 25 at 14:24
  • 1
    $\begingroup$ Note: We are assuming you know not to measure resistance with power on the board. $\endgroup$ – Perry Webb Jan 25 at 14:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.