Another filament cooling mod

Post improvements made for UP, and share ideas.
ming
Posts: 194
Joined: Wed Mar 12, 2014 2:06 am
Location: Georgia USA

Another filament cooling mod

Post by ming » Wed May 28, 2014 12:21 pm

Objective: to cool the extruder wheel and filament area using an external air pump
Idea: To not have to modify the existing extruder and can be added on anytime without having to change major components
Equipment: 3 Way tube splitter (w oshort sleeve on the extruder) and a small oring

Steps:
1. Drill (enlarge the hold for the sleeve to insert approx 2-3mm into the extruderbody (ingress hole)
2. insert 6mm long sleeve partially into the hole (id 2.0mm)
3. thread small oring (id 1.8mm) onto filament (helps direct suction to chamber)
4. add t fitting
5 add assembly to system

ps. i currently do not use this option as im only printing ABS and do not have this requirement.
Attachments
coolfil_2.jpg
optional filament cooling
coolfil_2.jpg (86.15 KiB) Viewed 18825 times
Dream@night / Holdit@sameday

roller
Posts: 1622
Joined: Sat Oct 06, 2012 9:09 pm
Location: Brisbane, Australia

Re: Another filament cooling mod

Post by roller » Fri May 30, 2014 12:54 pm

I've tried similar before. It does seem to extend PLA print times a little but long term it doesn't work because the flow of air seems to be insufficient. I even tried chilling the air (through a peltier chilled aluminium heat exchanger) but that wasn't enough for a 2 hour plus print.

It seemed in my experiments that you don't really need to get air on the filament so much as on the extruder gear with as great a volume of air passing it as possible.

ming
Posts: 194
Joined: Wed Mar 12, 2014 2:06 am
Location: Georgia USA

Re: Another filament cooling mod

Post by ming » Sat May 31, 2014 12:10 am

i am not able to say if it works or not because i ve not used it and not needed to (yet)

But if more air is needed just a larger "T fitting" and sleeve thats all is needed
Dream@night / Holdit@sameday

roller
Posts: 1622
Joined: Sat Oct 06, 2012 9:09 pm
Location: Brisbane, Australia

Re: Another filament cooling mod

Post by roller » Sun Jun 01, 2014 1:49 pm

It always good to have new ideas brought to the forum but it's not good to write ideas as solutions when they are no well tested as it can lead new users in the wrong direction, wasting time and causing frustration at best and cause unnecessary damage to their printers in the worst case.

In this particular case I have tested such a solution and before doing so I trawled the forum for previous experiences to take on the knowledge of others experiences. The problem with your solution is there is insufficient airflow around the filament in the feeder tube and increasing the feeder hole sufficiently will:
- pretty much remove all the plastic around the feed shaft - the default design is too fine
- cause the feed port to be wide enough that the filament will be prone to wander which will sometimes result in misfeeds.

The idea is fundamentally sound (except for the focus on trying to cool the filament) but the practicalities of the default feeder design means it doesn't work . If you are going to go as far as a redesign you might as well go with Drew's great redesign with a cooling tube feed. :)

ming
Posts: 194
Joined: Wed Mar 12, 2014 2:06 am
Location: Georgia USA

Re: Another filament cooling mod

Post by ming » Sun Jun 01, 2014 4:49 pm

Glad I'm not in school for innovation
I've always shared that I may not be successful in many things does not mean you will not be unless you try.

Sure try the tested successful ways that's always best. But I'm confident if I need to get that to work I can

Maybe should ask the q

What's jamming
What's causing it to jam or slip
Oh so it's softening because it's reaching it's softening point
Oh so it's temperature
What temp is it softening at

So cooling the source that conducts the filament is increasing the temp seems to help basically reducing the conducting temp. Is the conduction time very long? That's thermal energy. So solving it is to prevent the filament from reaching softening temps.

Maybe just maybe reducing the start temp can also reduce the end temp. Just saying. What do I know I'm not a thermal engineer just one that don't believe in status quo. I believe in failures and fail early.
Dream@night / Holdit@sameday

roller
Posts: 1622
Joined: Sat Oct 06, 2012 9:09 pm
Location: Brisbane, Australia

Re: Another filament cooling mod

Post by roller » Mon Jun 02, 2014 4:58 am

There is a difference between being innovative and being stubborn. In my daily work I am an innovation facilitator. I help designers realise their the works of their immagination so trust me that I understand the principals and I also have a lot of experience thinking laterally to make innovation happen. But sometimes an idea is in contradiction to the laws of physics and on these occasions its back to the drawing board for the designer.

As part of my work I see lots of designers who keep coming back with a new angle on the same design approach ... unwilling to accept the science. I see this a lot with you. I am not trying to put down your ideas. I am trying to help you to see that they have either been tried thoroughly and scientifically or that the science fundamentally says no.

In this case the issue with cooling the filament is that plastic is a thermal insulator so it will be hard to cool. It will also not dissipate the contact heat from the extruder along it's length for this same reason and it's cross sectional is minimal so it takes very little energy to get it to soften at a point. So you would literally need to refrigerate if not freeze the filament to make gradient sufficient for the filament cooling method to be reliable. Now that the theory.

On top of that I did all the experimentation because I love using PLA - it's my preferred material and I have tens of kgs of the stuff. I used a range of filaments and measured the before and after temps of passing through the extruder drive, I tried alternate feeding systems and I spent a lot of time cleaning PLA powder out of my extruder. So I did the science. I played with your idea ... I tried it with higher volumes, peltier chilled air and redrilling parts of the feeder included using a secondary duct system. It did not work well.

So if you want to work through alternates I am happy to help if you want it.

To learn to be a good designer we also have to learn when to shelve a design that is a dead end. I am a designer too ... so I know it's hard to give up on our brain children and it's much easier to see the weakness of others designs with clarity. It's good to learn to take criticism not as being personal but as a critical examination of what your work offers then take these on board and work them through. Stubbornly denying the science and the experience of those who have tested your work practically gets us no closer to inventing a better solution. It just eats up everyone's time and worst of all misleads those which don't have such experience to spend a lot of frustrating time on a dead end.

I am not saying give up on you filament cooling ideas ... but address the issues raised or strike out in a new direction.

ming
Posts: 194
Joined: Wed Mar 12, 2014 2:06 am
Location: Georgia USA

Re: Another filament cooling mod

Post by ming » Mon Jun 02, 2014 12:36 pm

Nothing personal bout it because At this time I have no use of a cooling mod so again you can take it and run or you can "shelve it". I live in reality. And no point talking bout it and I'm sure you have lots of experience and lots of good ones too.

I live by making mistakes and that's just who I am. Thanks for pointing out the errors and hope someone can learn from.it. If we just sit still we will grow old fast. Example many years ago LASIK was done with a few kinds of keratomes and everyone just rides the boat. Look up the first. Commercial epi-LASIK keratomes see whose patent it was ..... Just maybe there are alternate options...

Like you said the filament is an insulator and contacts the gear for a split second. Somehow it melts it so there must be some heat transfer that's more weighted ........
Dream@night / Holdit@sameday

roller
Posts: 1622
Joined: Sat Oct 06, 2012 9:09 pm
Location: Brisbane, Australia

Re: Another filament cooling mod

Post by roller » Tue Jun 03, 2014 3:47 am

ming wrote: Like you said the filament is an insulator and contacts the gear for a split second. Somehow it melts it so there must be some heat transfer that's more weighted ........
Filament is an insulator but that means it stops heat spreading readily so it is hard to heat or cool across an area. It does not mean the substance is impervious to heat. Conduction by direct contact with a good heat conductor such as the extruder gear is very efficient and will quickly transfer heat to the filament. Cooling the filament (an insulator) by conduction with air which is also an insulator and low density is very inefficient. It's not magic ... there is no somehow to it ... simple physics.

User avatar
DrewPetitclerc
Posts: 581
Joined: Sat Feb 12, 2011 7:39 am
Location: Silicon Valley California
Contact:

Re: Another filament cooling mod

Post by DrewPetitclerc » Wed Jun 04, 2014 12:21 am

roller wrote:If you are going to go as far as a redesign you might as well go with Drew's great redesign with a cooling tube feed. :)
Just posted the mod for the newer "Afinia H-Series H479" extruder with air and air+pressure adjustment.
http://www.thingiverse.com/thing:352168

In case anyone is interested.

Regards
Drew
Drew Petitclerc
Petitclerc Designs
Owner/Senior Principal Designer, prototype, tooling and test equipment design and 3D printing
http://flash-graphics.deviantart.com/
http://www.thingiverse.com/DrewPetitclerc

lukeott
Posts: 53
Joined: Wed Jun 26, 2013 6:43 am

Re: Another filament cooling mod

Post by lukeott » Wed Jun 04, 2014 7:38 am

In case anyone is interested
Yes, thank you Drew.

ming
Posts: 194
Joined: Wed Mar 12, 2014 2:06 am
Location: Georgia USA

Re: Another filament cooling mod

Post by ming » Wed Jun 04, 2014 6:48 pm

Dont forget the heat transfer equation
Has the following variables
Time / Area / Volume / Start temp / End temp

You are intelligent im sure you can figure out the spilt second its in contact and the minor area it touches for conduction
Dont forget the Start temp and the time constant


Ill leave it at that as its not something i would be interested in now
Dream@night / Holdit@sameday

pleppik
Posts: 336
Joined: Sun Dec 11, 2011 2:14 pm

Re: Another filament cooling mod

Post by pleppik » Sat Jun 07, 2014 4:51 pm

One approach I've wanted to play with is actually moving the stepper back by about 1cm and extending the feed gear somehow to reach the filament path.

Like many others here, I've been gnawing on the heat problem with PLA for years, and it seems to me that the fundamental problem is that the stepper motor will always heat up, and when the filament spends 30-60 seconds or so in direct contact with the motor it's just gonna get warm.

The cooling approaches seem to work, but a more elegant solution would be to just move the motor so there's enough of a gap to keep the motor's heat from being conducted directly into the filament.

I haven't tried this, though, because I haven't been able to come up with a way of extending the feed gear short of buying a whole new stepper motor with a longer shaft.

If anyone has some thoughts on this approach, I'd be interested to hear them.

User avatar
JuliaDee
Posts: 959
Joined: Sun May 06, 2012 10:00 pm
Location: Connecticut
Contact:

Re: Another filament cooling mod

Post by JuliaDee » Sun Jun 08, 2014 4:16 am

DrewPetitclerc wrote: Just posted the mod for the newer "Afinia H-Series H479" extruder with air and air+pressure adjustment.
http://www.thingiverse.com/thing:352168

In case anyone is interested.

Regards
Drew
Thanks, Drew, appreciated!

roller
Posts: 1622
Joined: Sat Oct 06, 2012 9:09 pm
Location: Brisbane, Australia

Re: Another filament cooling mod

Post by roller » Mon Jun 09, 2014 2:06 pm

pleppik wrote:One approach I've wanted to play with is actually moving the stepper back by about 1cm and extending the feed gear somehow to reach the filament path.

Like many others here, I've been gnawing on the heat problem with PLA for years, and it seems to me that the fundamental problem is that the stepper motor will always heat up, and when the filament spends 30-60 seconds or so in direct contact with the motor it's just gonna get warm.

The cooling approaches seem to work, but a more elegant solution would be to just move the motor so there's enough of a gap to keep the motor's heat from being conducted directly into the filament.

I haven't tried this, though, because I haven't been able to come up with a way of extending the feed gear short of buying a whole new stepper motor with a longer shaft.

If anyone has some thoughts on this approach, I'd be interested to hear them.
It's a fundamentally sound idea but I'm not sure 1cm would be enough. 1cm and using another fan ducting air across that 1cm gap would be I suspect. Certainly 2-3CM would be plenty especially with some active cooling. Maybe someone an make an extension shaft mod out of stainless or some other low conductivity metal.

walter
Posts: 19
Joined: Sat Sep 08, 2012 4:43 am

Re: Another filament cooling mod

Post by walter » Tue Jul 01, 2014 7:09 pm

I tried putting several layers of thin adhesive PTFE tape over the extruder block where it is in contact with the filament to try to reduce the heat transfer and it seems to help. You can also pull a teflon pad off the bottom of an old mouse and try using that if it's the right thickness. It won't fix the PLA feeding issue entirely, but it can be an easy way to print longer or expand the range of printable PLAs. I haven't tested it extensively though as I rarely print in PLA.

D3rax
Posts: 139
Joined: Thu Feb 07, 2013 7:54 pm
Location: Netherlands

Re: Another filament cooling mod

Post by D3rax » Thu Jul 03, 2014 1:27 pm

walter wrote:I tried putting several layers of thin adhesive PTFE tape over the extruder block where it is in contact with the filament to try to reduce the heat transfer and it seems to help. You can also pull a teflon pad off the bottom of an old mouse and try using that if it's the right thickness. It won't fix the PLA feeding issue entirely, but it can be an easy way to print longer or expand the range of printable PLAs. I haven't tested it extensively though as I rarely print in PLA.
Okay this gave me a little brainfart. I don't know how much space there is to put a teflon layer over the extruder block, I'll have to check that. But as far as I know nylon is a good heat insulator to right? So I was think about printing a thin (at the max possible thickness) nylon sheet to separate the PLA from direct contact with the extruder block.

Maybe combining this with placing some heatsinks at the back of the steppermotor and using this fan mod (http://www.thingiverse.com/thing:200072), might be a sufficient solution for me.
Up! Plus

User avatar
printmo
Posts: 14
Joined: Fri Jul 11, 2014 1:45 pm

Re: Another filament cooling mod

Post by printmo » Fri Jul 11, 2014 3:29 pm

Hey, check out my stepper motor cooler. This mod enabled flawless 7 hrs (probably more, simply had not enough parts :D ) PLA printing in a pretty hot environment @ 27°C ambient 9 K above the recommended printing temperature! ==> http://www.thingiverse.com/thing:390993

User avatar
JuliaDee
Posts: 959
Joined: Sun May 06, 2012 10:00 pm
Location: Connecticut
Contact:

Re: Another filament cooling mod

Post by JuliaDee » Sat Jul 12, 2014 2:19 am

Nice, I've wanted to put a fan on the end of the stepper for a long time, now you've saved me the trouble of designing the mount :) Two questions:

1) What's up with the big scoop on the extruder fan intake?

2) You thank pp3dp for including part files, but IIRC the supplied parts are all in the proprietary .upp (or whatever it is) format. Did you somehow reverse-engineer the proprietary format, or are they now making the part files available in STL?

Thanks,
Julia

User avatar
printmo
Posts: 14
Joined: Fri Jul 11, 2014 1:45 pm

Re: Another filament cooling mod

Post by printmo » Sat Jul 12, 2014 9:37 am

Hey Julia, thank you for your comment
1.) I think you are referring to this part: Thingiverse ==> /thing:253694
2.) No they don't offer STLs. I basically took measurements on the supplied cover (fortunately metric and a good guess), then build it in CAD exported it as STL and overlapped it with the original part in the UP software. A long process with fixing up dimensions, exporting and overlapping again, but it definitely works, of course with a parametric CAD software...
overlapping_up_cover.PNG
Pink one is the PP3DP cover
overlapping_up_cover.PNG (102.46 KiB) Viewed 18043 times
Greetings printmo

ming
Posts: 194
Joined: Wed Mar 12, 2014 2:06 am
Location: Georgia USA

Re: Another filament cooling mod

Post by ming » Sun Jul 13, 2014 5:21 pm

Congratulations for trying something different and resolved the PLA heat issue.
Dream@night / Holdit@sameday

Post Reply