Mosquito hot end fan
-
@gperman said in Mosquito hot end fan:
I dont think the pin output on the Duet is voltage variable though.
No, not exactly. PWM control means the voltage is the same, but it's only applied in short durations. Pulse width modulation. So we are modulating the width of the pulses of voltage over a time slice, ie at the frequency chosen.
You could try adding a resistor to the + leg and run it at a reduced voltage that way.
Also try a different fan port to see if the behaviour is any different there.
-
The resistor thing would be a last ditch effort IMO. Figuring out why the fan cant change speeds via PWM is the main objective. I'm wondering if the people contributing config code have a mosquito fan or another.
I switched connector to fan2 output and the behavior was the same. I'll try another fan. I'll try and contact mosquito and see what they say about their fan, pwm and freqency.
-
As efficient as the mosquito cold end is I'm surprised they went with such a typhoon of a fan.
-
@gperman those fans have electronics inside . not all fans can accept pwm signal .
can it run from on board 5v or is it too low ?
you can also try adding small electrolytic capacitor in parallel with the fan and try is with S0.7 -
It seemed fine at any voltage when modulated that way. Now, I'm not sure what speed it will be best at for printing. Havent gotten to that point yet. I might be able to install a variable DC board with a pot that sticks out of the electrical enclosure to play with it manually since it doesnt seem the duet can do it electrically.
Then again, perhaps the new fan en route will be more cooperative.
-
@gperman said in Mosquito hot end fan:
Figuring out why the fan cant change speeds via PWM is the main objective.
There are 2 type of fans that does not play well with speed control.
The first set of fans will change the speed but will fail quickly, rattle, and have another type of problems as they are designed only to run at full speed. Typical example are fans with "air bearing", they are quiet and last a very long time but must run at 100% speed. If you drop the speed (by any mean, pwm or lower voltage or...) they will not last long (dropping speed to 50% usually make them last less than 10% of their life time in my experience). So while these fans will often behave ok when you try to pwm them, you should not and you should use other fans. They are good option only for hotend cooling, not for part cooling where you want to vary them.
Second set of fans will not work at all or will work "weird" (strange noises, blocks..) and for them the problem is just the type of bldc controller inside the fan that does not play well with PWM ... Sometimes you can solve it by providing lower voltage to them but that also reduce the lifetime of the fan so also this fan is ok only for running at 100%. Sometimes you can "force" this type of fans to work with PWM by varying the PWM frequency, sometimes going very low with PWM frequency will make them work ok, but usually lowering the pwm freq. will also produce audible noise from the fan that I'm sure you want to avoid
Most 3 wire fans are ok with PWM input, good brands will have in datasheet if they are pwm-able or not, 4 wire fans are also 99% working with pwm on the voltage input (they are actually expecting to receive speed control signal on the 4th wire, but will usually accept pwm on the voltage too). Small 25, 30, 40mm 2 wire fans very often come with driver that does not support pwm input.
-
good info, thanks.
so for the hot end cooling fan: good practice is to find what kind of cooling is necessary for filaments you plan to use and stick with that, hopefully it's quiet.
I'm confused as to why the duet2 board only provides 2 wire fan outputs and pwm control when primarily only 3 and 4 wire fans are PWM'able. That makes it difficult to modulate the part cooling fan properly.
-
@gperman said in Mosquito hot end fan:
so for the hot end cooling fan: good practice is to find what kind of cooling is necessary for filaments you plan to use and stick with that, hopefully it's quiet.
In my experience the more you cool the hotend the better. You cant cool it too much, only too little. Especially with mosquito, you can't go wrong with a lot of cooling of the cold part of the hotend.
I'm confused as to why the duet2 board only provides 2 wire fan outputs and pwm control when primarily only 3 and 4 wire fans are PWM'able. That makes it difficult to modulate the part cooling fan properly.
Because 3 wire fan uses 3rd wire (yellow commonly) to report back the RPM so you just cut that third wire and use the two (red/black) to power and pwm the motor. Since duet don't really care what real rpm of the fan is they don't waste cpu power to count the impulses from the third wire.
The 4 wire fan you could connect to duet but imho it's a waste of money as 4 wire fan's in small sizes are expensive as hell if you can even find them. You would connect the GND (black) and 12V (usually yellow or red) to GND and 12V directly, the third "sense" wire that reports RPM you again cut off and don't use (it is usually green or white) and fourth wire that's pwm control wire (usually blue) you connect to duet's pwm output (ground of the fan connector) with a 10k pullup resistor to 12V
So, 3 wire, you just dump the third wire and 4 wire you either use only first two and it works or you use the three + resistor so no need really for any changes on the duet
-
I have the very same problem with many fans, and my solution was converting PWM to an analog voltage. If you can (so it seems) use a high enough PWM frequency, an inductor-capacitor lowpass (inductor from V+ wire from Duet, capacitor parallel to fan) could be a solution. For cases where PWM frequency can't be adjusted (Duex5) or when needing more control, I designed a small board:
https://forum.duet3d.com/topic/21680/designing-a-pwm-to-analog-mini-board-for-fans
@hackinistrator said in Mosquito hot end fan:
you can also try adding small electrolytic capacitor in parallel with the fan and try is with S0.7
If you just use a capacitor, the PWM pulse will charge the capacitor completely, and you loose regulation capability. It'd probably run at 100%RPM from 100%PWM to... ¿50%? PWM and then almost stop. There might be a very small window of regulation when PWM period is small enough not to charge the capacitor completely and discharge through the fan, but that's a lottery and seldom works. You need to use a current control device (inductor, or even a resistor) and form a lowpass.
-
@Egon-Net you are right, its trial and error. I've done it before many times, but i use scope, which makes it much easier.
