Part cooling fan help needed
My kit came with the protechnic 4028 fan and I have been unable to get the fan to not run at full blast. I can't find any documentation on how to install this fan. Based on what I have found, I have the 12V power coming from the FAN7 connector on the Octopus v1.1, and the PWM signal connected to what I believe is the pin tied to PA6 in the connector for FAN0. The tach is not connected. I've looked through https://os.ratrig.com/docs/guides/4028/ and have tried the various combinations in there. I also tried uncommenting the 4-pin 25kHz fan config that's in printer.cfg, which didn't work because the board doesn't have any aliases for "4p_fan_blahblahblah" in it.
I've tried using the pin definition both with and without the !, I've tried using the enable_pin setting, etc. Nothing has allowed the fan to be turned off (other than unplugging it of course).
Can anyone supply a link to documentation on how to set up this specific fan with an octopus board? Preferably including tach support, with wiring info.
Connecting a 4028 fan | RatOS
4028 server fans are becoming a popular option for part cooling because of their light weight and great pressure and flow rates. In this guide we'll go over how to connect one and use it for part cooling. 4-pin fans usually aren't used in 3d printers, and many boards only provide 2-pin ports, but fear not! There's always a way.
47 Replies
if you use the enable pin you need to connect your FAN GND cable to the heater output GND you have configured in the enable pin
this definitly works if correctly installed and configured
Heater output. Not the big screw terminals where the power comes in, but somewhere else?
sunny-green•2y ago
Correct, those big terminals
One problem is that the pwm pin was set to 24V. I did see a 25-ish kHz square wave on that pin at least. Duty cycle seems to match the fan setting, within reason. But the fan speed never changes. Bad fan? Does it maybe only work with the tach connected?
Maybe the 24V pwm blew it out?
You can't set the PWM pin to 24V. It's a mosfetted gnd connection.
Please post a pic of your wiring
This, only instead of feeding FAN0 into the tach output like I was doing, it is now going into the pwm, and it works 🤦♂️. Regarding the voltage, the instructions currently say to set FAN0's jumpers to 24V, but as you can see in the pic, I changed mine to 12V. Yes, it was definitely 24V then 12V as I had hooked up my oscilloscope to confirm I was getting the expected signal. So if I want to get the tach signal into the octopus and on mainsail, where can I find instructions? Currently searching the interwebs.
Yes, it was definitely 24V then 12V as I had hooked up my oscilloscope to confirm I was getting the expected signal.And how did you hook that up? It's the positive pin that provides the connection to the respective power rails, and thats what you're switching with the jumper. You don't use the positive pin. But good you got it working 🙂 Btw, the PWM pin needs a 5V signal. (which it's getting.. regardless of the jumper setting).
Pwm always connected to pin 1 on fan0. And the data sheet for the fan says pwm is minimum 3.3v max 14v.
Oh really, they usually want 5v, but cool.
It's still getting 5v 😄
But the pwm was on that pin, but definitely affected by the jumpers
Try removing the jumper, you'll see your fan still works
Should it be on the right pin or left pin, assuming all the header connectors are on the right? Pics indicate right like in mine and that's how I've had it
Ok
in the orientation on your picture, it's the right pin. Left pin is positive.
It would be very bad to connect it to that.
50%, no jumpers. Ik. Pardon the lazy screen shot
what's the positive connected to here?
50%, jumpers on 12V
You're not measuring correctly
Positive on the fan is connected to positive on FAN7
Positive input to the oscilloscope
Pwm
and gnd?
Fan7 gnd
yeah that's why
You want to measure between +12v (like your fan is getting) and the negative pin (the one you use for PWM) on the fan port.
the "pwm pin" is a mosfetted gnd connection, like i mentioned
So gnd to pwm, positive to fan positive?
that would give you the correct signal yes
If the fan positive is connected to 12v
which it should be
Again, all the jumper is doing is switching which voltage rail the positive pin is connected to
you can look up the schematics
Again on jumper but with reversed probes
No jumpers. Fan isn't turning on now.
"no jumpers"? Plural?
english grammar 🙂 There was never more than one jumper on there.
Just wanted to make sure you didn't remove it from the fans power source (FAN7) as well
Because your results make no sense
ah. 🤷♂️
at least it's been made to work 🙂
anyway, put the jumper on 12V, and move on i guess 😄
indeed 🙂
if i could sit down for 5 seconds without my cat jumping in my lap I'd get this finished a lot faster 😆
Home stretch, and I appreciate everyone's help so far. I have the tach pin connected to PG10 (just above the thermistor connections, I believe). Mainsail shows the tach reading, and it works up to 66%. After that, the tach reading drops down to roughly 7k rpm (down from over 11k), and quickly falls to 0. I'd cranked down the poll interval to improve the situation, but after doing this a few times and getting down to 10 us, I'm starting to wonder. I did confirm that the ppr was 2 for this fan (or at least seemed to be, at 50% the tach was showing 22 kHz). Here's the config I'm using at the moment.
i guess the fundamental question here is, what's a reasonable poll interval for a 22000 rpm fan... Now that I think about it, maybe I'm wrong about ppr,.
then again, probably not because at 50% the rpm shows as 11000rpm
(I did start with the recommended formula, 30/(rpm*ppr) = .00068, but that wasn't a short enough poll interval to get reasonable results at lower fan speeds, so I kept shrinking it down)
You want a
^ in front of the tach pin, however i'm not sure that'll fix your problem.It didn't completely fix the problem but at least it gives reasonable results up to 88% now, instead of only 66%.
try a lower poll_interval, something closer to the max RPM
Wish there was an easy answer, but i think it requires additional circuitry to get a good reading from some fans. Some seem more consistent than others, the tach output from a sanyo is much more stable than from a delta for example.
I saw where one guide (I forget which) mentioned needing a pull-up resistor but it didn't get very specific, so I was hoping I could get away without it
That's what the
^ does
That enables an internal MCU pulluplowering the poll interval to the suggested value (.000681) makes the measurement diverge at 85% instead of 88%, fwiw
🤷♂️
it was worth a shot
indeed
at least it's not a critical item. I should be able to start printing stuff without it. Just a nice to have item.
This hasn't been touched in a few days as I'd kind of moved on, but I thought I'd follow up as I've been doing exploratory calibration prints and I noticed that the fan is currently showing non-zero RPM at 100% now. I didn't do anything more to make this happen, unless one of the recent software updates did it. Or a power cycle. Who knows.
could also be that I finally put a proper connector housing on the RPM sensor wire instead of just a bare socket crimped on the end of the wire 🙂