Noctua fan PWM issue
I have 2 x Noctua 40x10mm 12V PWM fans on my hotend, one for the heatsink (always on) and one for the print cooling fan.
Both fans have their +12V and GND connected to the heatsink fan GND, which are connected to one of the always on fan headers. The print cooling fan has the PWM wire connected to the FAN- for the print fan, which is connected to FAN0.
The heatsink fan works fine - it's always on, full speed. The green and blue wires are not connected at all.
My issue is that the print cooling fan is not behaving...
The print cooling fan worked fine when I first installed it, all M106 commands worked fine, and it could be fully stopped at S0, which was great. However, once the hotend and heated bed were warmed up, the fan would sometimes not start. If I brushed it with my finger, it would whir into life. I figured maybe there was something funny about that fan, so I swapped them (they're identical). Right after the swap, the second fan worked fine, but then after the hotend and bed were heated, it died completely - it won't even start on 12V with the PWM disconnected now.
I did a bit of digging, and it looks like the fans are expecting to see a 5V PWM signal - have I damaged (both) fans by putting 12V on the PWM line? I'm sure a lot of others have used Noctua fans - did anyone put a voltage divider or similar?
I would like to avoid swapping the fan jumper to 5V, as I have a bed fan (for cooling large PLA prints, and for post-print cooling to get prints to detach quicker, which runs on 12V, and won't run on 5V. I would also (certainly) need to buy 2 x new Noctua fans to run on 5V rather than 12V...
Any thoughts / suggestions are welcome!
Are these actual 4 wire PWM fans? What is the Noctua model number?
Yep, they're 4-wire PWM fans. Noctua NF-A4x10 PWM.
Yes - this is the guide I followed to connect them. They're connected as specified in the guide, but not working as expected.
Which Duet are you using?
The FAN- connection is an open drain output with a flyback diode connected to VFAN. The leakage in the flyback diode is very small, so it should be safe to connect the FAN- pin directly to the PWM input. However, if you are worried about the leakage then you cold connect a 1N4148 or similar diode between the FAN- pin and the fan PWM lean (cathode to the FAN- pin). I am running two Noctua 4-wire fans on Duet 3 in this way.
Did you remember to set the fan PWM to be inverted, and to increase the pWM frequency to 25000 ?
I had my Smart Effector off to change out some other components yesterday, and took the opportunity to re-make all the fan connectors and test some things.
I tested both fans with a known good 12V supply - both were working. I reassembled everything, and haven't had an issue with it since...
It seems that there was a dry(ish) joint somewhere between the Duet and the fans, and the voltage wasn't quite high enough on the fans for them to start reliably (apparently Noctua fans are known to need a high voltage to start - i.e. if there is too much droop on the 12V, they won't work).
You didn't say in your original post that you were using a Smart Effector. The magnetic joints on the effector prevent some makes of fan from starting, especially if you haven't alternated the rod directions so that they go N-S-N-S-N-S around the effector. I suspect that may have been the problem.
Psylenceo last edited by Psylenceo
If this helps you or anyone else this is how I have my 2 Noctua 4pin fans:
; Fans M950 F0 C"!out4" Q25000 ; create fan 0 on pin out4 and set its frequency M106 P0 C"Extruder" S0.2 L0.2:1.0 X1.0 H1 T60:180 ; set fan 0 name and value. Thermostatic control is turned On, low low speed at 60c max speed at 180c+ M950 F1 C"!out5" Q25000 ; create fan 1 on pin out5 and set its frequency M106 P1 C"Part" S0.2 L0.2:1.0 X1.0 H2 T25:80 ; set fan 1 name and value. Thermostatic control is turned on low speed at 25c and max speed at 80c