Faulty thermistor/MCU on v4.2.7

Basic Information:

Printer Model: Ender 3 pro
MCU / Printerboard: Creality V4.2.7
Host / SBC
klippy.log

klippy (6).log (708.6 KB)

Describe your issue:

My ender 3 pro shutdown mid print with the error “MCU ‘mcu’ shutdown: ADC out of range" This generally occurs when a heater temperature exceeds its configured min_temp or max_temp” I’ve tried to replacing the hotend thermistor (both read with multimeter at ~120 k ohm) with no success. I have probed both R66 and R67, of which R66 is normal at 4.7k ohm. However R67 reads 100 ohm, the printer have run for hundreds of hours, im just wondering how this happened and if i can simply solder another SMD resistor to bring back functionality

  1. It seems your bed’s thermistor is causing the issue.
  2. Both thermistors report a startup temperature above 180 °C, which seems strange (could be a startup glitch).
  3. Your Klipper version is too old to report which thermistor is acting up, but according to the log, it should be the bed.
1 Like

I just tried probing the bed thermistor aswell, it’s at 100k ohm which as far as i can read should be fine. I’ll try to update Klipper later. I found a schematic of the 4.2.7 board that you had posted some years ago Creality Board 4.2.2 and 4.2.7 Schematics and from what i can tell, the voltage divider of R66 and R67 is likely the culprit if the values you provied are correct

This would be the first time that I see a defect here. I consider this as pretty unlikely. Much more likely is a fault in the thermistor, its cable or connector.

I just reposted the information from a GH issue. The source is given in the post.

Allright, thanks, you seem to have far more experience, ill try to replace the bed thermistor and see if it works out ^^

I’ve tried to replace the bed thermistor and the error prevails

Just for diagnostic purposes, set some insane min and max temperatures and provide a new klippy.log please.

Have you checked for the correct sensor_type?

The new log is here
klippy_newpobe.log (1.2 MB)
While the printer was running i tried to remove and reinsert both probes, but the temperature remained the same.
Further I tried to probe the resistance between TH + and - aswell as TB + and -. The Resistance on TB was only 2 ohm while the one on TH was 4.7 kohm. Which then makes sense why the temperature remains high regardless of the thermistor as there must be a shortened connection on the TB.
Aslo i didn’t get around to update the klipper version

I forgot to adjust to the new thermistor, I can try that later, but both thermistors have the same resistance at room temperature so for startup diagnostics I believe it should be fine

There is something very fishy going on. As you can see, the bed shows around 587 °C, while the extruder shows 184 °C. Both without a set target and both without any PWM activity.

The bed could really be a short, but the extruder is strange. It could be a wrong thermistor setting as @EddyMI3D pointed out.

I’d recommend to:

  • Closely inspect wiring
  • Closely inspect the connectors on the board and the board itself
  • Verify the type of NTC you are using and cross check against settings

As for the schematics, this is no voltage divider. It’s just two pull ups for TB and TH each.

2 Likes

You are correct that was a mistake on my part

The thermistor forms the rest of the voltage divider.

Assuming the thermistor is 100k Ohms the voltage measured from pin 1 to pin 2 should be
3.3V * (100/147) or 2.24V. If your meter has a low impedance (rare in 2025) it will load the circuit and pull the volts lower.

With the thermistor unplugged you should read 3.3V

1 Like

I doubt that it’s part of the settings with the hot end, as stated the printer was running mid print while the error occurred, the hot end temperature had a normal rise during that print. And the hot end thermistor is identical.

Without being entirely certain I suspect that the short of the bed is causing a parallel to ground greatly reducing the current/ voltage drop in the hot end leading to a bad reading, mostly based on the constant temperature no matter if the hotend thermistor is connected or not. Feel free to correct me on this though :slight_smile:
I will try to borrow a friends microscope/camera with zoom to check for faults on the board as I can’t see any when examining it
If i still can’t find the error at that point i suspect that it is c44 or the STM32f103 that somehow shorted

I have an old printer running a V4.2.7. If it gave me any trouble I’d bin it in a heartbeat and get a board with 5 motor drivers and UART support. The 2 Z motors on a single driver sucks.

Just saying.

I unsoldered both the C44 capacitor and R67 resistor and a short remained on the board, i couldn’t see any other connections on the physical board. I belive the Stm32f103 might have shorted, and im not gonna attempt to replace that.
Thank you all for the help

So what board would you recommend ?

Personally I prefer boards with soldered on drivers.

The SKR Mini is the same size as your 4.2.7 but has some quality control issues.

I don’t have any hands on experience with one though. Do your own research.

Lastly I personally would never EVER buy a board that integrates the SBC. That limits your upgrade options and makes diagnosing communication problems a nightmare.

1 Like

SKR Mini or SKR Pico are good choices. Pico is a lot smaller and a bit cheaper with most of the same features as the SKR mini.

I have both a SKR Mini E3 V2 and V3 and they work fine. Well…I did fry my V2 after feeding 12V to the wrong port, but that’s user error. My V3 is still running one of my printers just fine.

1 Like