super fan duct for ultimaker -- delicious clam

[clamking 0]

Say goodbye to curling edges and ugly overhangs！

If your prints knocked over by your nozzle, try this fan duct.

Up to 80 degrees overhang without any support structure!

guys who don't konw fluid mechanics at all, please stop to design any fan duct.

BUT EVERYTHING HAS ITS COSTS. Some sacrifices must be made.

The printing length of x axis will be 25-35mm shorter. Use the slider extender attached.

Recommend to print first one with few support, and use this fan duct to print another without support.

fan duct design is not easy, the fluid won't flow as you wish. remember fluid always choose the easy way.

boundary layer separation will always cause pressure loss

fan duct is published here http://www.thingiverse.com/thing:1492821

Edited April 16, 2016 by Guest

[gr5 2,005]

Did you print this out of PLA? doesn't it melt near the nozzle?

[IRobertI 516]

Looking at the pictures it looks like he has reinforced it with kapton and aluminium tape.

Very interesting, I'll see if I can get around to printing one tomorrow to test.

[neotko 1,417]

I assumed that this kind of designs constraints the air input since its using normal fans than can't generate enough force with an air output that small.

Also I assumed that if the air output it's so small it will work while printing on concentrated areas but might not cool down as well for bigger areas.

Anyway I would kill to have free software that could analyze air speeds through an object.

All that air that bounce back into the heater block doesn't create heat fluctuations?

Edit. Oh I see. You use a jennyprinter with 1.75

Edited April 17, 2016 by Guest

[avogra 68]

How did you model the fans in your simulation? Is it constant flow, constant pressure or did you somehow integrate the flow/pressure characteristics of the fans?

[clamking 0]

How did you model the fans in your simulation? Is it constant flow, constant pressure or did you somehow integrate the flow/pressure characteristics of the fans?

 

use the curve in the fan datasheet,

[IRobertI 516]

Ok, I gave it a try today. Not much difference in my eyes. Slightly better at the most extreme overhang but on the other hand it has defects on the edges that are not there on the stock fan shroud of the UM2+. From this initial test I wouldn't say it's worth the reduction in print area. The fans sound quite restricted as well.

[gr5 2,005]

Wait - you compared UM2+ versus the "delicious clam"? Or versus stock um2?

[IRobertI 516]

By "UM2+ Stock" I mean the stock fan shroud that comes with the UM2+. Maybe I could've been clearer there.

[ultiarjan 1,221]

How did you position the thing on the bed, my guess is on the stock 2+ the left side is better than the right side.

Here I made a comparison between the stock 2+ and labern's version, where the overhang part was on the right side, and labern is clearly better than stock 2+

https://ultimaker.com/en/community/20409-um2-upgrade-kit-first-impression-and-some-tips

[IRobertI 516]

Good point, should've thought about that. The overhang was on the left. Dumb of me. I'll re-run it on Friday.

[clamking 0]

Good point, should've thought about that. The overhang was on the left. Dumb of me. I'll re-run it on Friday.

 

make sure your nozzle is 3mm out of the duct. if you still couldn't get good quality, then give it up, maybe not suitable for your printer. a guy in our group uses the same nozzle like yours, he couldn't get better quality, too. i'm using a e3d nozzle.

Edited April 21, 2016 by Guest

[gr5 2,005]

Oh! I thought "UM2+" referred to the left print and "stock" referred to the right print!

[IRobertI 516]

Ok, new prints, now with the overhang on the right. I adjusted the nozzle so that it was sticking out 3mm (was at 4mm previously). There are some slight differences but I wouldn't say they are major.

If you look closely on the UM2+ version on the third "tooth" from the front you can see that it has caved in slightly while on the clam version it did not. The two most extreme overhangs are slightly better looking on the clam as well.

If you look at the left front corners of the first few teeth here you'll see that the UM2+ version has a crisper edge than the clam. The UM2+ also has less curling on the tips. On the last two, both shrouds seem to produce similar results but I'd give the clam the edge here.

[MartijnvG 16]

Hi Clamking

I really like your fan-duct design. especially the flow simulations you provided next to the design.

Having a close to symmetrical design would always be a preference. And the advantage of 3d printing is perfect in being able to perfectly guide the air to the nozzle as smooth as possible.

But as the metal also doubles as a heat shield/ anti finger burner we made some choices in that area.

One question i do have is why you have the airflow enter on the nozzle instead of underneath it? in my opinion you would be cooling down the nozzle here maybe a bit to much, especially when actually printing and printing other materials then PLA.

The simulation shows how the air would flow when the bed is moved all the way down. but this is rarely the case in actual printing. there is almost always either a complete or partial layer underneath, making the air flow differently.

Discussions have always been going on with 3d-printers about what would be the perfect way to cool. Very concentrated around the nozzle (https://www.youmagine.com/designs/mkiii-clone-r1-active-cooling-duct) or general layer cooling (think of the big laminar coolers on the side of the machine https://ultimaker.com/en/community/4890-the-crossflow-fan-approach).

We try to do something in between, cooling the model as well as cooling the material just after printing. but with the head moving all the time while printing and the height of the print affecting the flow and backflow of the air when out of the fan-duct this is always a very interesting balancing act.

Keep up the good work! i would love to see more improvements/idea's.

[clamking 0]

thank you for your testings. many people just take it and don't give me any feedback.

another possible reason is i modify the angle of outlet after i test it. the outlet is 120 degrees before and 80 now to reduce the volume. according to my simulation, the temperature of nozzle decreased by 10 degrees at least. but people in our group is still not satisfied ,they want more powerful cooling, so the turbo fan edition is already under working by an old man in our group.

[MartijnvG 16]

cool! i would love to see a turbo fan version.

I would be very careful with cooling the nozzle more then 10C this can possibly make it harder for the feeder to push material through and grind the filament faster, or prevent the layers from melting together properly and then loosing structural integrity or worse de-lamination.

but i'm curious to see what you end up with!