Suck vs blow: E3D heatsink fan

Hi all,
For some time, I've cogitated on the way axial fans are used to blow air to cool filament and onto the heatsinks of hot ends. My understanding is that axial fans are good at extracting air out of a duct, but poor at blowing it. To back this up, I notice that air is blowing away from the fans on the "input" side. And there has been some research showing that radial fans are better than axial at filament cooling.

So, I decided to change my filament cooling fan to a radial 50x50x10, and I'm very pleased with the result... a significant improvement in print quality. The mount I used was designed for a 40mm axial hot-end cooling fan. In the spirit of experimentation, I tried putting the fan on the other way round (i.e. sucking instead of blowing), and found that it worked just fine. When printing, I can feel a plume of warm air coming out of the fan. I've done several hours of printing, with no jams or clogs.

I'll admit that this is a less than scientific test, because I replaced a 30mm blower axial fan and its associated duct (which worked fine) on my Chinese "E3DV6" hot-end with a 40mm sucker and a different duct. But it's at least a data point that says that 40mm axial fans *can* work as suckers on the heatsink. Whether they work better than blowers, I still don't know.

I'd like to encourage other RepRappers of an experimental nature to try turning their axial heatsink fans around. In theory, they should work better this way, and it would be nice to know if that's true in practice. Please post results of any experiments here.

Frank

Hi Frank, I've run a shrouded fan in suck mode and I agree, it seems to work ok. I can't think of any reason why a fan would work any better in suck vs blow configuration though - as far as I can see they are functionally equivalent, the fan can't tell the difference. Any fan has two basic characteristics, how much air it can move at 0 pressure differential, and the pressure developed at 0 flow. Axial fans (especially the thin ones we tend to use) generally aren't very good at creating pressure which is why you will hear people say they aren't very good for pushing air into a narrow nozzle. But it wouldn't be any better trying to suck out of a narrow nozzle either.

Hi James, I think it's to do with aerodynamics and turbulence. I'm not an expert, but here's how I see it:

Sucking air out of a narrow nozzle will create a low pressure area ahead of the fan, in the duct. Air can't flow into this low pressure area from the sides (because the duct walls are there) so you get a smooth, fast flow directly up the duct to the fan. Fast is good... more cold air going past the heatsink fins will cool it better.

OTOH, blowing air into the duct means that the air in the duct is turbulent because it's been all chopped up by the fan blades so it won't flow well. Secondly, on the inlet side of the fan, air can be sucked in from the sides, which leads to turbulence ahead of the fan. As I said, I've noticed air blowing *away* from the fan on the "input" side, which I believe is due to this, and which is obviously not helping cool the heatsink.

Interesting ideas Frank. What we need is some experimental data - a plot of temperature at a fixed point above the heat sink against time for a given hotend temperature, comparing each direction. We could use a thermistor connected to the bed input and use standard host software to draw the graph. It would be a bit of a pain pulling the printer apart to set it up, but sure would be interesting to see the results.

When using a fan to blow air over a heatsink that gives significant back pressure, an inlet duct makes the fan more efficient and eliminates the backwash. This is the one I use for 40mm fans [www.thingiverse.com].

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].