Apparently the culprit was the wiring. I wired the positive fan wires to the buck converter output (12V) and the negative fan wires to Duet fan grounds according to the documentation, specifically Mixed-voltage setups are not directly supported, but the Duet switches the fans' connections to ground, so you may be able to connect each fan's positive side to the appropriate voltage and its negative side to the appropriate pin on the Duet. This results in an average of 8V across the fans when PWM is set to 0%.

Scrolling down I found the section Using 12V fans when VIN is 24V and it mentions wiring the output of the buck converter to the center pin of V_FAN. This makes the voltage average go down to about 0.4V on the fan output when PWM is set to 0% and the fans behave as expected.
Also mentioned in that section is leave the VFAN pin of that connector not connected. While I assume that this refers to the positive output for the fan it gave me the idea to remove the V_FAN jumper and this also made the fans behave as expected.

Overall this mild frustration could have been avoided by reading the docs more carefully
For those of us too eager to get results fast may I suggest adding the part about disconnecting the V_FAN jumper to the section I quoted above since that is what people first see when navigating to that page? Either that or a link to the section below it which explains in more detail how to wire mixed voltage fans.

One last thing, how do I mark a thread as solved? I can't seem to find that option anywhere

I would like to have the firmware save the last print timestamp for all files and send it to Duet Web Control to be displayed in the Jobs list. Displaying it on the PanelDue would also be a nice touch if the available panel space allows for it.

This would make it easier to keep track of which files were printed or not out of a large assembly, such as a list of components for a 3D printer, without requiring the use of external tools.
It would also be helpful to save the last print state. That is whether the print was successful or failed/cancelled.

Maybe this could be added as a comment in the gcode file itself, similar to how the "Generated by" info is parsed and displayed.

I would like to have the firmware save the last print timestamp for all files and send it to Duet Web Control to be displayed in the Jobs list. Displaying it on the PanelDue would also be a nice touch if the available panel space allows for it.

This would make it easier to keep track of which files were printed or not out of a large assembly, such as a list of components for a 3D printer, without requiring the use of external tools.
It would also be helpful to save the last print state. That is whether the print was successful or failed/cancelled.

Maybe this could be added as a comment in the gcode file itself, similar to how the "Generated by" info is parsed and displayed.

@Phaedrux said in Fans on buck converter not turning completely off:

Thanks for the feedback on the fan docs. Glad you got it sorted out.

No problem. Hopefully it'll help other people who might end up in my situation

Topic Tools button, choose ask as question, then topic tools again, and this time, mark as solved.

Thanks

Apparently the culprit was the wiring. I wired the positive fan wires to the buck converter output (12V) and the negative fan wires to Duet fan grounds according to the documentation, specifically Mixed-voltage setups are not directly supported, but the Duet switches the fans' connections to ground, so you may be able to connect each fan's positive side to the appropriate voltage and its negative side to the appropriate pin on the Duet. This results in an average of 8V across the fans when PWM is set to 0%.

Scrolling down I found the section Using 12V fans when VIN is 24V and it mentions wiring the output of the buck converter to the center pin of V_FAN. This makes the voltage average go down to about 0.4V on the fan output when PWM is set to 0% and the fans behave as expected.
Also mentioned in that section is leave the VFAN pin of that connector not connected. While I assume that this refers to the positive output for the fan it gave me the idea to remove the V_FAN jumper and this also made the fans behave as expected.

Overall this mild frustration could have been avoided by reading the docs more carefully
For those of us too eager to get results fast may I suggest adding the part about disconnecting the V_FAN jumper to the section I quoted above since that is what people first see when navigating to that page? Either that or a link to the section below it which explains in more detail how to wire mixed voltage fans.

One last thing, how do I mark a thread as solved? I can't seem to find that option anywhere

@Phaedrux Visually, they look brand new. I have seen a MOSFET turn into a solid wire without visual evidence in the past, though.
I'll know more about them once I get it up on the scope. I would be surprised if both failed at the same time I changed fans, though, bit too much of a coincidence. I'll have a look at their datasheet, Sunon fans do draw more current than Noctua according to their labels.

Hi,

The converter is connected to VIN so its input voltage is always around 24V and its output is always around 12V when the printer is powered on.
As for the fan outputs on the Duet, I didn't measure it but I have to assume that the average voltage is above the minimum required by the fans to start spinning.

After it finishes its current print job I'll hook it up to my oscilloscope and check the PWM signal.

I replaced the 5V Noctua and "Prusa" fans I was using with 12V Sunon fans using a buck converter. I would have preferred 24V fans so I wouldn't have to use extra electronics but I couldn't find any easily.

Both fans work fine, including PWM for the part cooling fan except that the new fans don't turn off completely when set to 0 and just continue to run at a low speed. This doesn't concern me for the hotend fan but I do want the ability to stop the part cooling fan.
Could it be that the firmware isn't setting the PWM duty cycle to 0 when turning off the fans and it was just pure coincidence that the old fans stopped because whatever duty cycle was being used was below their minimum?

The system is running on a Duet 2 Wifi with RepRap Firmware v3.2.2.
The buck converter is connected directly to the printer's PSU, the fans' positive wires are connected to the 12V output of the converter and their negative wires are connected to fan0 and fan1 ports respectively as per documentation.
Fan configuration:

; Fans
; Part cooling fan
M950 F0 C"fan0"                                ; create fan 0 on pin fan0
M106 P0 S0 H-1                                 ; set fan 0 value. Thermostatic control is turned off
; Hotend fan
M950 F1 C"fan1"                                ; create fan 1 on pin fan1
M106 P1 S1 H1 T45                              ; set fan 1 value. Thermostatic control is turned on

Thanks in advance for any insight or solution.