Greg Saunier

I've done lots of testing on nozzle fan ducts and I have found a recipe that works for me. However, I've never tried CPAP or berd air (pump), so I will speak here for blowers as a source of airflow.

Image credit: Nozzle fan duct of BearExtruder V1 generated by Keith Bennett https://x.com/thisiskeithb

A small thread ⤵️

#3dprinting #cooling #shroud #duct

Keith Bennett (@thisiskeithb) on X

@MarlinFirmware team | 3D printing | serial modder | beta tester

X (formerly Twitter)
Aug 18, 2024 at 10:34amWeb
Show threadGreg SaunierAug 18, 2024
The 1st most important factor is the fan specs, it will impact a lot more the results than changes on the fan duct shape. Then comes the directions of the flow. To me the best solution was to use 3 sides (front or back, left and right). The 3rd side helps to flush away the heat compared to a 2 sides (left-right) design. Another important factor is to not loose too much energy. For this, you need large outlets, not too much changes of directions and smooth surfaces. ⤵️
Show threadGreg SaunierAug 18, 2024
The 4th factor is to get closer to the nozzle as possible, using a hotend with a thin hot block is therefore a good idea. It helps both the nozzle temp stability and the layer cooling. Finally, I would recommend to not loose too much time with CFD. It is powerful but it is difficult to setup correctly and much more difficult to interpret the results adequately. I also suspect working on such a small scale is harder to simulate. So in addition to CFD, I recommend to focus on real prints with a ⤵️
Show threadGreg SaunierAug 18, 2024

temperature stabilised enclosure. The prints should test different types of overhangs, bridges and short layer time. Checking the flow with water under the extruder is also useful to see where the duct blows.

After all this I wouldn't say I know what we must do with the flow directions, pressure and velocity. I only took the experimental and testing approach. Understanding what we really want is, I think, extremely hard. So I wouldn't be surprised if other designs work great.

(end)

Show threadDennis.Aug 18, 2024
@bear_lab Hey😊
A while ago I also tested berd air, with a self-adapted duct: https://www.printables.com/de/model/234017-bair-berd-air-prusa-bear-fan-shroud
Maybe that helps 😊
On my Voron I‘m using cpap with a disturbingly loud 4028 fan, but the power of that is just amazing.
Lower resonances on the toolhead, therefore more acceleration without impact on results
Printables

Show threadGreg SaunierAug 18, 2024
@brokenminded interestingly you did the 3 sides like I described! I have the feeling that berd air on a single side, very focused, at few mm of the nozzle would be better. But nozzle fan ducts are a mystery so I might be totally wrong. If not better I would try 3 sides but very close to the nozzle and straight out of the end of the tubes.
Show threadGreg SaunierAug 18, 2024
@brokenminded I am tired of the noise of my filtration system so I am not going to add CPAP soon, that would be even noisier :D
Show threadDennis.Aug 19, 2024
@bear_lab At first, I simply tried with a short pipe, directly pointing below the nozzle. The problem was the heat - my ASA -printed duct melted very fast. I think, because of the brass tube.
Since I got CPAP on the voron, I‘m having ANC headphones laying around😅
Show threadGreg SaunierAug 19, 2024
@brokenminded Oh I didn't imagine it was getting that warm but yeah slow airflow.
Good idea the ANC :D