Part Cooling Fan and Ducts



  • Two questions really both related to part cooling

    1 : Has anyone found a 24V DC PWM controlled radial fan ? - does such a thing exist ?

    All the fans I've found are two wire voltage based - tempted to use a 40mm axial fan but ducting that will foul the back of the printer / Z axis leadscrew.

    2 : Optimum distance from the part for the cooling ducts - is there any formulae for this ?

    On my Prusa I keep spare part cooling ducts as over time the heat radiated from the hot end distorts them, especially when printing above 220 without the cooling fan running.

    Has anyone ever considered making up these things in brass or copper plate ?

    I've designed my own duct for tests and although the part cooling works fine with minimal impact on the hot end the bottom of the duct is a little over 2mm above the print, the higher the discharge the more it impacts the hot end. I print almost exclusively PETG with occasional Nylon12 and Polycarb / Carbon Fibre so heat soak is a problem for me.

    I'm considering putting the duct high up and then using copper tubes such as brake line to discharge the air near the print but before I do I'm testing the water to see if there is any information out there that is hiding from Google.



  • @Garfield said in Part Cooling Fan and Ducts:

    Two questions really both related to part cooling

    1 : Has anyone found a 24V DC PWM controlled radial fan ? - does such a thing exist ?

    Depends if you truly need 4-wire PWM or not. Simple voltage control via the Duets PWM works quite well, but it is dependant on the fan on how low you can reliably go.

    https://www.digikey.com/products/en/fans-thermal-management/dc-brushless-fans-bldc/217?k=&pkeyword=&sv=0&sf=1&FV=14|155281%2C790|320427%2C1989|0%2C-8|217%2C589|203935&quantity=&ColumnSort=0&page=1&pageSize=25

    All the fans I've found are two wire voltage based - tempted to use a 40mm axial fan but ducting that will foul the back of the printer / Z axis leadscrew.

    Axial fans for ducted cooling is not ideal. Not enough static pressure.

    2 : Optimum distance from the part for the cooling ducts - is there any formulae for this ?

    There are a few schools of thought on this. As usual it probably depends on a lot of factors, how fast you're printing, how big the part is, how much cooling you need. There's no formulae I'm aware of. But consider if a small directed flow of air would be best or if a wider more dispersed flow would be best.

    On my Prusa I keep spare part cooling ducts as over time the heat radiated from the hot end distorts them, especially when printing above 220 without the cooling fan running.

    Whenever possible use something like PC-ABS or ASA for ducting. It will resist much higher temperatures before deforming. PLA simply isn't suitable for high temps. Annealing your part may help, but you'd need to compensate for warpage.

    Has anyone ever considered making up these things in brass or copper plate ?

    The weight may be prohibitive. At that point there are probably better alternatives like a byrd air pump with small pipes for ducts.

    I've designed my own duct for tests and although the part cooling works fine with minimal impact on the hot end the bottom of the duct is a little over 2mm above the print, the higher the discharge the more it impacts the hot end. I print almost exclusively PETG with occasional Nylon12 and Polycarb / Carbon Fibre so heat soak is a problem for me.

    Silicone block sock?

    I'm considering putting the duct high up and then using copper tubes such as brake line to discharge the air near the print but before I do I'm testing the water to see if there is any information out there that is hiding from Google.

    Byrd air + tusk duct

    https://www.sublimelayers.com/2017/10/tusk-part-cooling-for-titan-aero-and.html



  • Interesting that last one for sure - I'll do some digging as I hadn't seen that 'Tusk' before - probably more controllable than the brake pipe thing I had in mind. I hadn't considered putting the air pump off the machine completely - that may be worth exploring.

    Where that design has holes I'd have individual pipes - at least that was the plan and run them back to a manifold - but then you need a fan that can generate decent static pressures. I'm definitely going to experiment with something like the Tusk I think.

    The Prusa fan can go pretty slow but it is only 5V and has a tacho lead so you now when it isn't running as it should. I could run a Prusa set up on the Duet I guess as I've had no issues with them and I do have spares of both the 40mm Noctua and the radial fan.

    I've had issues with socks in the past (falling off and picking up filament) so I tend not to run with one these days although I probably should.

    Part of the trigger for this search is to find the option to run on a Duet3 6HC with the least messing about translating voltages etc, I'd also like to be able to flag an alarm / buzzer if the fan doesn't run when it should - so some sort of tacho / feedback - although I have never in 2 years printing had this scenario so perhaps I'm just over engineering ....



  • @Garfield said in Part Cooling Fan and Ducts:

    so perhaps I'm just over engineering

    Nothing wrong with that, but yes it's likely a non-issue if you use a quality fan.

    There are a few more examples of the byrd air system on the forum.
    https://forum.duet3d.com/search?term=berd air&in=titlesposts&matchWords=all&sortBy=relevance&sortDirection=desc&showAs=posts

    For silicone socks some people use a small piece of solid core wire to secure it.

    You can make use of 4 wire PWM fans. For the Duet 2 I think the limit is one, on the Duet 3 they've bumped that up a bit.

    https://duet3d.dozuki.com/Wiki/Connecting_and_configuring_fans#Section_Connecting_4_wire_fans




  • administrators

    @Garfield said in Part Cooling Fan and Ducts:

    Part of the trigger for this search is to find the option to run on a Duet3 6HC with the least messing about translating voltages etc, I'd also like to be able to flag an alarm / buzzer if the fan doesn't run when it should - so some sort of tacho / feedback - although I have never in 2 years printing had this scenario so perhaps I'm just over engineering ....

    @Garfield, Duet 3 6HC has one bank of three 4 wire fan connectors (which can also drive 2 wire fans), and one bank of three 2 wire fan connectors. With 24V power, each bank can be set to 12V or 24V. Maximum total 12V power is 0.8A.



  • The ideal for me would be to find 4 wire 24V fans sufficiently small but that is proving to be the challenge - getting the airflow, keeping it quiet and keeping it simple.

    I've even considered building 'option' boards that can fit to the fan and use whatever voltage I wish - make them totally autonamous - they look at the sensor and manage the fan with no need to do it in the Duet I don't think a teensy or similar with a FET would actually add that much weight. This would eliminate the need to run the sensor wires just power but you'd still need comm's ( hmm bluetooth / wifi ...) and then there's interference to deal with.

    But this doesn't eliminate the heat distorting the cooling ducts -



  • @dc42 - in your arsenal of designs over the years have you created any 'low pass filters' a DAC to convert PWM back to voltage ?

    i.e. source voltage in, PWM signal in, drive voltage out.

    Since DC fans are not PWM fans and this would eliminate my struggle to find small 4 wire fans.

    Would an electronics expert care to advise a total amateur.



  • Just found this - just in case people thought moving air was easy ...

    Why cooling fans are so hard, and some preliminary testing



  • @Garfield Haha, yes I saw that just the other day and thought your post was good timing. I almost posted a link to the video myself actually.


Log in to reply