Fans and stuff improvements…
-
And my axe! On a side note, if adding a variable voltage regulator is too much then maybe a small screw terminal for the fan power supply? I'm currently feeding 12V for the fan selector on a 24V system, but I had to wire the big ones directly to the PSU - I'm afraid a single pin wouldn't be able to reliably handle the current otherwise.
-
The main issue is that providing a 12V fan supply when the board is powered form 24V is the cost of the additional switching regulator. The Duet is already one of the more expensive 3D printer control boards, and we are reluctant to make it even more expensive.
On a more positive note, I have been in communication with Noctua regarding 24V 40mm fans and 12/24V 30mm fans for 3D printers. They told me that they intend to add them to their range - but perhaps not until near the end of 2018.
-
Having the 12v select on the Duex5 has been awesome, but I'm thrilled to hear that Noctua will be looking at increasing their lineup, too.
-
A suggestion for next gen Duet then - add on fan modules with different voltage configurations to be available as accessories.
EDIT. Come to that, you could take the existing voltage regulator off the board altogether making it cheaper for users who run everything at Vin, but have a connector or connectors that would accept an add on board (or boards) to step down the voltage for those that need it.
-
+1 for that, some sort of multifan add on module would be quite popular, something with Vin, 12v and 5v available, maybe even bus connected to the new duet system, so that you could attach maybe 6 fans at whichever voltage you wanted. Maybe native support for RPM monitoring for failed hotend fan etc..
-
+1 for that, some sort of multifan add on module would be quite popular, something with Vin, 12v and 5v available, maybe even bus connected to the new duet system, so that you could attach maybe 6 fans at whichever voltage you wanted. Maybe native support for RPM monitoring for failed hotend fan etc..
Some sort of multi fan module as you describe would need to also facilitate fan mapping to heaters (virtual and real) as well as PWM control. So I'd have thought bus connected might be essential (but that's my line of expertise).
-
Have you considered making the next hardware release split into two or more models? One basic and one maxed out with all the possible bells and whistles? Seems anecdotally that the Duet brand is becoming really strong in the market place though I have no evidence to base the feasibility of this approach of course.
-
Yes, we have. Expect an announcement next month - but we won't be announcing anything that supersedes the Duet Ethernet and Duet WiFi.
-
Yes, we have. Expect an announcement next month - but we won't be announcing anything that supersedes the Duet Ethernet and Duet WiFi.
Looking forward to it. It'd be nice to be able to use more than 10 steppers without using external drivers. It'd also be nice if I didn't have to run 20 conductors all the way from the Duet through cable chain to my 5 moving extruders but I' not holding my breath on that.
-
More than 12 steppers won't be in next month's announcement; but we are working on it.
-
Instead of creating a new thread, I'd rather just piggyback onto this. I have an M3D Promega which has spots for 2x 50 mm fans that I may use for heat intake for a heated chamber or exhaust to regulate temperate or a combination of that, not sure just yet. There is also this: https://www.thingiverse.com/thing:2812765
Which uses two 50 mm by 10 mm fans.
I am used to building PC's and have thought about PC fans for my 3D printer. The only thing I don't like is the voltage because all PC fans I've ever come across are 12v. The only problem with that (in my opinion) is that due to some of the fans being 0.3A, 0.4A, or even more per fan, it is quite easy to exceed an entire amp per header and the speed is lost with two pin.
I think 3 pin fans should be supported completely for larger machines two or three capable of 1.2-1.4 A would be perfect so you could chain them together but two or three would be ideal.
The easiest way I see this being able to be integrated is adding a third pin to the already existing fan headers, but is it possible to have two rows of headers that are PWM controlled, all with three pin for 3 pin case fans? I think this becomes more and more important as machines grow. With the Maestro, the voltage of the headers can be changed which is REALLY convenient, I think this is a good thing to keep and maybe have more of as far as flexibility goes.
This is more because it would be convenient more than anything else. I'm having a hard time finding high-quality 24v fans and probably have to buy new part cooling fans and/or my hot end heat sink fans, and/or my duet motherboard fans in order to get the 12v 50 mm ones by Fractal Design unless I can find sufficient fans that aren't the cheap standard terrible ones. It would just be nice to be able to hook them up directly without having to use a 3 pin to two pin connector or cutting the wire and crimp a connector for only two wires.
I understand this may not be particularly high on the priority list, but with so much heat generated by both the drivers at times, the power supply, and the always an issue oozing of the nozzle and other things, better fan super and more flexibility (along with just a bit more convenience) with fans is not what I would call a bad thing to have, even if PCB real estate is valuable.
-
@noskillzengineer we are looking at supporting 4 wire fans (rather than 3 wire) in Duet 3.
-
@t3p3tony How come? Why?
In my experience, 4 wire fans have physical switches, likely because of control on the fan itself, that is responsible for the speed control of the fan.
I choose not to use these in my Personal PC (or in any that I build) simply because I cannot control the speed of the fan with PWM on the motherboard so if the processor heats up due to higher load, the software follows a specifically designed temperature curve plotting points on a graph that is essentially the duty cycle on one axis and the desired temperature on the other so you have complete control over the fan duty cycle through software and it can then be completely automated.
I prefer software pwm control over hardware control for fans, pumps, and motors in general.
Why support 4? If your response means 4 pin headers like PC fan headers that will also run two and three fans/motors, then no issues here.
If you four pin headers without software PWM control for two and three pin motors, I would like to understand why, if it's not too much trouble.
-
@noskillzengineer said in Fans and stuff improvements…:
In my experience, 4 wire fans have physical switches, likely because of control on the fan itself, that is responsible for the speed control of the fan.
I have never used a 4-wire fan with a physical switch. I have 2 of them in the PC that I am using, and some more on the bench that I have tested with various Duet-based controller electronics.
PWM is PWM whether it is controlled by hardware alone (which would be very unusual these days) or by software.
Why support 4? If your response means 4 pin headers like PC fan headers that will also run two and three fans/motors, then no issues here.
There are two problems with controlling the a speed of 3-wire fans:
-
Most of them were never designed to have PWM applied to the power feed to them. You won't find any PCs using PWM to control the power feed to a 3-wire fan. It's remarkable that so many 2- and 3-wire fans tolerate PWM at all.
-
If you PWM the supply to a 3-wire fan then the output from the tacho wire is no longer reliable, because the tacho sensor depends on receiving continuous power. So the tacho reading would be wrong except at full speed.
4-wire fans are designed to work with PWM, so they are a much better choice if you want variable speed and RPM sensing. OTOH we could provide a 3-wire always-on (or just on/off) fan connector if users want that.
-