Cooling with an air pump?

Hey Mike,

Actually, we could make the Nimble out of temp resistant material for you. That is the beauty of SLS manufacturing, tune the material. We could make the gears from the same material and polish it. It will last long enough, but POM is better of course, under normal circumstances.

Funny thing is, I was thinking about it when you said you were doing the vacuum bed printer. I thought the issue would be the drive cable sleeve, but have since found a (heavier) solution where temp is less relevant. I will have to modify the design a little, for the added girth, but hey, that is what CAD models are for, right?

So the net result is, yep I can do it for you, if you want. Just pull the trigger. :-)

Lykle

Any chance I could get a quote on your high temp version? Feel free to message me if you don't want to give a quote on this page.

Professionals often use fine sand to fill thin walled tubes before bending.
That way you can heat it up and bend at the same time. Less risk of tearing. The sand rinse out afterwards and you can be sure, the tube isn't clogged by a blob of solder.

Hi Lykle,

I will hold off ordering until phase 2 i.e. "We know that the vacuum plate experiment works/doesn't work so now make the printer useful" but put me down as very interested.
If I come across one of those long periods where I seem to be banging my head on a wall and want a break I may get a nimble for my existing Delta.

Mike

Here is hoping it all is a quick success and you start banging the wall.

I will do some proper research on this and get back to you guys as to how realistic it is and at what price.
I know the materials are available, just want to check if the long high temp will not impact durability too much.

lykle

Hi guys,
I just stumbled on this topic...I did this about 2yrs ago with very limited success. The aquarium pump is mostly unnecessary IMO, and here's why below. I'm sure it can work well, but costs more than a 30mm fan. The size of your pump will probably not help you much. I used a tiny aquarium pump at the heat break, and it cooled the heat break exactly to the air temp, but I still couldn't print PLA with that machine.

I'm assuming you're doing this to print PLA reliably. If you have access to a thermal camera, this will show you what's going on way better than I could explain, and better than you can guess. PLA melts at 50-55C, and most often, reprap machines have trouble feeding it because the gear melts the filament and thus strips it easier. The filament might be stripped without melting, but definitely strips much easier when your motor is warming the filament via the gear. Most extruder motors run at over 50C. The PTFE liner is necessary. Throughout your print jobs, melted PLA will rise into the heatbreak, with the height depending on how much you've "overloaded" the hot end with cool PLA.

So, for PLA you mainly need these 2 things:
-ambient temps around the extruder drive gear need to be under 40C (means a cool, or no enclosure)
-must have 2" or so of PTFE heat break liner (to help you recover from overfeeding your hot end)

...the e3D v5 and v6's print PLA so well because they take the heat from the hot end away from the extruder gear and have the PTFE liner. You'll still have to be sure temps at the gear are low.

Regardless of it it's truly necessary or not, I think you could come up with a great alternative to the 30mm fan.

Pumped air delivered through a pipe is has been used for different purposes..

• What I call "Near Field Cooling", that is cooling the filament immediately after it has been laid. The purpose of this is to cool the freshly laid plastic before plastic flow smooths out fine detail. The air can be delivered by a fabricated shroud or an air ring having multiple nozzles.
• Bridges. This is a special case of near field cooling but extending over a longer distance. Air rings are less helpful here and a small local fan may be more useful.
• Creating a heat break so that semi-molten plastic does not stick to the inside of the filament feed tube. This heat break should be as short as is practical but there is nothing precluding air from being used to do this as Srek has shown. Too much heat at the extruder gear may be indicative of poor extruder design. (or it may be that the gods of thermodynamics just don't like you)
• Bulk cooling of the print. The necessity and extent of this seem to vary between prints, slicer settings, printers and even users. My personal take on it is that it is at its most useful in minimizing hot spots from the heated build stage - but I may be wrong.

Mike

Instead of a ring around the nozzle, I can see a line of air vent holes following "Mama nozzles" path like ducklings.
This would be servo driven and the angle information would come from gcode or firmware.

Quote
o_lampe
Instead of a ring around the nozzle, I can see a line of air vent holes following "Mama nozzles" path like ducklings.
This would be servo driven and the angle information would come from gcode or firmware.

When you see tiny stepper motors at about a pound each even the most arcane ideas become practical. I can imagine a bunch of nozzles moving in perfect synchrony like Riverdance.

Having said that, I am finding difficulty getting a ring of nozzles on my hotend so that they fit in the available space and put air where it is wanted. I am planning on using two lines air nozzles with each pointed close to the extruder nozzle.

I will try your earlier suggestion about using sand fill to bend tube when I get the chance. I have tried it in the past with builder sand but with poor success. This time I will try it with kiln dried sand commonly sold for dressing block paving.

Mike

Quote
leadinglights
Having said that, I am finding difficulty getting a ring of nozzles on my hotend so that they fit in the available space and put air where it is wanted.

I've been having the same problem with a printed fan duct that I wanted to get close in to the nozzle, there just isn't much space between the heater block and the print. I was toying with the idea of making some long nozzles with a couple of mm more clearance, but then the heat gradient along the nozzle would be bigger, so you'd probably want to add insulation... It all starts to seem a little overly complex. Who would have thought it's so difficult to get cool air to go where you want it I think the best results I ever had was with two 50mm blower fans, one on each side of a single nozzle blowing air from about 10mm away. When I went to dual extruders that wasn't practical and I've struggled to get even cooling since then - it always works better on one side of the model or the other.

I would think that two nozzles would be all that's needed. If you point them in opposite directions along side the nozzle - not pointing directly at each other or the nozzle but, on the circumference of a circle surrounding the nozzle- they will set up an approximately circular air flow all around the nozzle. Something like this, maybe:



Of course, who knows what it will do when you factor in a close proximity print...

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I just printed this turbine blower. As most comments suggest, it does blow like hell. I used it with a 24V 10000 RPM DC motor. However there are two main disadvantages. The blower ist incredibly loud at 100% PWM. However, it is much worse that the blower cannot produce air pressure. It is good just for moving air around. Once I connect a hose adapter for two small 6mm hoses there is about 25% of the initial air flow left.

Maybe I would try this: Two small turbine blowers with very small 24V motors that blow air directly to the hot end nozzle instead of using my 40mm radial fans.

---

Self-sourced Mendelmax 2.0-based Reprap Machine -- Ramps 1.4 & Mega 2560 -- DRV8825 (Z@1A, [email protected], [email protected], E@1A) -- genuine E3D v6 direct setup -- 350W custom silicone heated bed -- ABS 1,75mm -- Marlin 1.1.0-RC7 -- Cura 15.04.6

Has anyone thought about using compressed air? A mini compressor might seem over the top but they're cheap, could be located away from the printer to keep the noise down, only needs a hose and a remote on/off switch.

You'd have massive quantities of air when needed.

You'll need a regulator and I presume some sort of servo operated valve to control the air flow.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
the_digital_dentist
Of course, who knows what it will do when you factor in a close proximity print...

That seems to be the biggest issue with the designs I've been trying recently. They seem quite promising when you run them in clear space, but once they are operating in the confined space between the nozzle and the print it becomes much more difficult to get even coverage. My first attempts at rings were useless, but now that I have a better idea of what's going on, I might give that another go. The current iteration is a fork, it seems to do well on three sides, but the side facing away from the duct isn't well cooled.



The air doesn't come out of this in the direction I would have expected, there's about a 30 to 40 degree slope in the direction that the body of the duct is pointing, so I set the tips of the fork level with the nozzle axis - hence the difference between the side of the model facing the duct and the side that faces away.

Hmm, I am going to use 2.9 mm tubing. Also, I made a little bending jig for Brian and he managed to get a reasonable good ring in 2.3 mm tubing. The jig still needs a little work, but I am beginning to think that the easiest way to do it is simply bend it by hand, keep tweaking it till it is perfect. Or breaks. :-)

The thing that makes it difficult is that I want a really small diameter ring around the nozzle. Anyway, onwards to other things.

Quote
DjDemonD
Has anyone thought about using compressed air? A mini compressor might seem over the top but they're cheap, could be located away from the printer to keep the noise down, only needs a hose and a remote on/off switch.

You'd have massive quantities of air when needed.

You'll need a regulator and I presume some sort of servo operated valve to control the air flow.

Not only thought of but build it and posted about earlier in this thread
[b.bonkers.de]
No need for a real compressor an aquarium bubble pump is sufficient.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

That's interesting and nicely done, I've tried an aquarium pump, and it wasn't very impressive. If the issues here are that its hard to pump air, that volume is lacking, that its difficult to direct it at the print etc.... Even compressed air at 30psi would give you the control you wanted. Remotely locate the compressor to the garage and the noise issue is gone. A tank full of compressed air is going to last a good while at the quantities were talking about, plus its instant. Sure getting a valve system to deliver the amount of air you want, controlled by the print is a bit of a challenge. I'd do this myself but I am really not unhappy with my twin 30mm fan shroud I use on two of my printers, its small, light, neat and cools the prints enough.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

You don't need a compressor and you certainly don't need anthing near 30PSI, 8PSI are sufficient for the setup shown on my website.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

I have two aquarium pumps, one is almost silent and the other ifs fairly quiet. The almost silent one is on the right and is a Hailea ACO-9602 which has an adjustable output of 0.8 to 7.2 litres per minute. The pump on the left is a Hailea ACO-318 which has an output of 60 liters per minute.


I have tried using two air jets but it was soon evident that it was leaving uncooled areas and was distinctly unhelpful with bridges. The air ring was a great improvement but mostly for preserving fine detail. Both types are shown in the composite picture below.


The greatest success though was from the 25mm fan on the left of the first picture. The positioning of this gives a lot of turbulence - this is a very good thing when you want even cooling including all of the nooks and crannies.
Long streamlined ducts give a laminar flow - good for aircraft but not too great for cooling,

Mike

I found that due to the much lower air volume the pumps deliver you have to compensate by focusing the air stream very tightly. The exhaust opening of my part cooler is about 8mm2 in size. Yours looks a lot bigger.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

With 3D printing the old saying YMMV, or "Your Mileage May Vary" becomes "Your Mileage WILL Vary" but the problems I got with two jets would typically be the definition on the feathers of the Owl on a Log which everybody prints out. With 2 nozzles I got fine definition on one side but poorer on the other. The air ring to give uniform(ish) and vertical(ish) air flow solved this. Tall thin wall towers did not have truly vertical walls with a simple fan but were solved completely with the turbulent fan arrangement.

Mike

I just realised that my success with one sided part cooling might be due to the Merlins airbrush nozzle being much thinner than most other nozzles, allowing a much better airflow near the point of etrusion.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

One way to create an easier airflow path and a better way to position air nozzles would be a different shape of the heater block. ( Assuming most people use E3D-like heater blocks )
The volcano style heater block is a step in the right direction. It's slim and leaves plenty of headroom to the heatsink.

Maybe there is a way to design a heater block that fits the short heat brake of classic E3D and is slim around the nozzle?
Maybe we can guide the air through the block with PTFE tube?

Edited 1 time(s). Last edit at 02/06/2017 03:09AM by o_lampe.

Quote
o_lampe
One way to create an easier airflow path and a better way to position air nozzles would be a different shape of the heater block. ( Assuming most people use E3D-like heater blocks )
The volcano style heater block is a step in the right direction. It's slim and leaves plenty of headroom to the heatsink.

Maybe there is a way to design a heater block that fits the short heat brake of classic E3D and is slim around the nozzle?
Maybe we can guide the air through the block with PTFE tube?

I don't think the PTFE tube will work, but Deltaprintr is using a super compact ceramic heater for their hotend. I would love to see something like that in more widespread use

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Now that is a heater worth having. Somebody please make those and put them on the market - I will pay real money

Mike

Quote
leadinglights
Now that is a heater worth having. Somebody please make those and put them on the market - I will pay real money

Mike

It is on the market, but sadly it's not available in 24V, otherwise I would have bought one to try.

---

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].

At HackFFM we tried to build something equaly small for the Merlin a few years back, but it turned out to be seriously difficult to build a reliable heater that small. Deltaprinter get theirs from a manufacturer in China, but i haven't yet been able to source something like it.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

I have one of these mini hotends on my microdelta and it produces stunning prints:

That is with no directed print cooling at all, just a 60mm fan at the side for bridges and extreme overhangs.
So what if it isn't available in 24v get a buck converter and run off that.
Heats up in under a minute too (probably 45 seconds), overshoots a fair bit even well tuned though and the heater block's thermal capacity isn't very high, so at higher flow rates/speeds a sudden long extrude can lower the temperature a bit.
Deltaprintr only make 0.4mm nozzles right now but have been promising other sizes for a while.

Apparently jamming with PLA is a well known problem, but I print mainly in ABS and it has only threatened to jam once, yesterday trying to print that 50% scale marvin at 60microns, the flow rate is just not enough to prevent heat creep.

Oh and the 25mm fan supplied sounds like a jet engine, in fact its so noisy I manually calibrate by lowering the nozzle until I can hear it buzzing against the build plate - no paper required, its like an acoustic sensor. I was wondering if this might another bed contact detection system, a vibrating hotend which triggers a mic to give a change in state on contact. I'd use a piezo but they don't like the heat do they?

Edited 2 time(s). Last edit at 02/06/2017 05:25AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

This DeltaPrintr heater is cute. Isn't it just a wire wound resistor in cement?

*goes to kitchen and rips the toaster apart *

Quote
DjDemonD
..............
Oh and the 25mm fan supplied sounds like a jet engine, in fact its so noisy I manually calibrate by lowering the nozzle until I can hear it buzzing against the build plate - no paper required, its like an acoustic sensor. I was wondering if this might another bed contact detection system, a vibrating hotend which triggers a mic to give a change in state on contact. I'd use a piezo but they don't like the heat do they?

I tried that with a pager vibrator motor and a single piezo disk. While it worked fine when touching some parts, it either did not work or was insensitive on other areas of the bed. Somebody elst with the same thought at [hackaday.com]
Back on the subject of the heater, is the heater alone (without the hotend) available and if so, from where?

Mike