Yet another fan housing

Post improvements made for UP, and share ideas.
Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Yet another fan housing

Post by Robin » Sat Dec 29, 2012 7:35 pm

I bought the Afinia and am having lots of fun but I think there is room for improvement.

Correct me if I am wrong, but the theory behind this fan housing is...

1: If you don't open the trap on the standard part the air has got nowhere to go, it's blowing into a closed box.
2: If you do open the trap then the nozzle takes forever to warm up
3: Opening the trap seems to make no difference to the print quality that I can see.

So my new fan housing tries simply to steer the air over the heat sink and vents vertically. I can't say it's an "improved" housing because I will have to use it lots then ask myself if I would like the original part back. I have been pointing a digi-thermometer at the heatsink and I don't think it is going to demolish my part.

The nozzle certainly heats up faster and I can feel a rush of cold air coming out the top rather than a gentle waft of warm air coming out of the vents behind the stepper

I have also been pointing my thermometer at the table. The software says I have 100 C, The thermometer says I have 90 C in the middle and 80 C everywhere else. I think it is losing heat out the bottom, if I put my hand under the table I can feel it radiating. So I am going to use the bolt holes around the rim to hold a sheet of cork on below, with handy cutouts for clips of course. May need to calibrate the thermocouple.

I also want a new cover for the feed motor, simply because I like to see what it's doing at the nozzle and the original cover gets in the way.

Love the idea of lagging the nozzle heater with HT silicone, this is a great site :D
fan.jpg
Fan housing
fan.jpg (144.31 KiB) Viewed 21703 times

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Thu Jan 03, 2013 10:43 pm

It's coming on. I needed something inconsequential to design while I learn to drive the Afinia and Alibre 3D design software. This is perfect.

After about 4 attempts I think I am nearly there, just the lid to go. How can I work in a helical spline? I figured out how to draw and print them over Christmas and now I want to use them in everything.

I split the air to get some flow close against two sides of the stepper motor.

I couldn't resist inserting a blister underneath to mount a couple of 5mm LED's. I angled them 30 degrees down so they point at the sharp end of the nozzle :mrgreen:

More when I have made a lid for it.
fan2.jpg
fan2.jpg (77.07 KiB) Viewed 21649 times

User avatar
wilsonj
Posts: 569
Joined: Tue Feb 28, 2012 10:26 pm
Location: Australia

Re: Yet another fan housing

Post by wilsonj » Fri Jan 04, 2013 3:26 am

Hi Robin,

Welcome to the forum.

Nice designs btw.

I just want to add a couple points.

Firstly, I think you will find that with the trap door closed it forces the air up and over the heat sink.

Second, with OEM material and stock temperatures there can be a tendency for slight burn marks in the print with the door closed. A bit of air also helps with overhangs and makes support easier to remove. Its also fairly essential when printing with PLA.

Keep up the great ideas.

Regards
Jamie
Regards
Jamie
3DPrinterGear.com.au

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Sat Jan 05, 2013 12:16 pm

Hi Jamie

Thanks for the information, I can never get enough information.

I didn't introduce myself. I am a cantankerous old git in the UK who makes things. I have a 25 ton Arburg injection moulder and want to try CNC cutting my own moulds in aluminium alloy. I bought the Afinia so I could test shapes before I attacked metal. It is a nightmare trying to perfect a 3D mould design on paper and you don't know if you have got it right until you get the test shots back from a very expensive piece of tooling.

I have seen the occasional burn mark. I thought it was old plastic stuck to the nozzle transferring to the model, but if it becomes a problem, or my overhangs start to sag I will try adding a cooling breeze.

I have now tried wiping the perfboard with acetone, as recommended here, and adhesion improved no end. I had tried alcohol and tricho but they made not a jot of difference. Good tip, hope I can give something back.

I had problems aligning the bed flat this week and took it apart. Tightening the screws on the linear rails improved rigidity (note: I have used a lot of Allen keys over the years and the only brands I have found worth having are Allen, Bahco and Snap-On). Replacing the springs that do the bed tilting fixed my bed alignment problem. The original springs only had 0.6mm of useful travel, I now have 2.6mm. Suggest you use the two outside holes on the front and the centre hole at the back. That way you can align the sideways tilt at the front then do the front-back tilt with the other. No point in having more than 3 screws, if you fit 4 one will always be irrelevant.

Slightly concerned to find the bed heater element wasn't plastered with thermal transfer compound. Couldn't find my tube so it will have to wait until the replacement arrives. I know the element is enclosed and the heat can only escape to the surrounding aluminium but without the compound the element will run a lot hotter than it needs to. It should not discolour and mine has.

Looking at the extruder assembly from a design point of view, the filament feed motor connects to the heatsink at the wrong end of it's thermal gradient. Why it cannot connect to the top is mystery to me, can anyone come up with a good reason for bolting it on at the bottom next to the nozzle heater. Why can't it bolt on at the top? Are they using the motor as a heat damper? Sounds like a terrible idea.

best

Robin

User avatar
wilsonj
Posts: 569
Joined: Tue Feb 28, 2012 10:26 pm
Location: Australia

Re: Yet another fan housing

Post by wilsonj » Sat Jan 05, 2013 8:11 pm

Hi Robin,

Wow, sounds like you have some big toys to play with! I understand about how expensive dies for molding can be. Part of my business is making sprinkler parts, which are injection molded. I certainly don't have my own machine though!!

The bed leveling springs have been an issue for a few in the past, and have sometimes been replaced with rubber washers.

The job of the heat sink is to cool the "cold" end of the extruding tube, so the filament stays solid enough to push out the already melted plastic in the hot end. I was also confused about the heat sink's use, thinking it was some how supposed to cool the stepper motor. It's not.

I'd like to see a fan design like this http://www.soliforum.com/3dpreview.php? ... 435892254f for the UP. Not sure if it would really be possible though, given the large heating block, it's proximity to the frame and the wires coming out the other side.

Cheers
Jamie
Regards
Jamie
3DPrinterGear.com.au

Blown454
Posts: 49
Joined: Fri Oct 12, 2012 6:13 pm

Re: Yet another fan housing

Post by Blown454 » Sun Jan 06, 2013 1:53 am

Robin wrote:
Thanks for the information, I can never get enough information.
Robin
Same here, thanks for the insight as you all have hit some points that I have tackled recently.
Robin wrote: I have seen the occasional burn mark. I thought it was old plastic stuck to the nozzle transferring to the model, but if it becomes a problem, or my overhangs start to sag I will try adding a cooling breeze.

Robin
I thought exactly the same thing, I never imagined those burn marks were possibly coming from always keeping the door closed when printing (especially the white OEM stuff). I now want to experiment with the door a bit. Thanks!
Robin wrote: I have now tried wiping the perfboard with acetone, as recommended here, and adhesion improved no end. I had tried alcohol and tricho but they made not a jot of difference. Good tip, hope I can give something back.
Robin
I had read when this was suggested and I pretty much ignored the possibility of it working, you say "adhesion improved no end". I think that means it worked good ;), so I am going to give this a shot as well, thank you very much Wilsonj and Robin!
Robin wrote: I had problems aligning the bed flat this week and took it apart. Tightening the screws on the linear rails improved rigidity (note: I have used a lot of Allen keys over the years and the only brands I have found worth having are Allen, Bahco and Snap-On). Replacing the springs that do the bed tilting fixed my bed alignment problem. The original springs only had 0.6mm of useful travel, I now have 2.6mm. Suggest you use the two outside holes on the front and the centre hole at the back. That way you can align the sideways tilt at the front then do the front-back tilt with the other. No point in having more than 3 screws, if you fit 4 one will always be irrelevant.
Robin
Totally agree about Allen keys, Bluepoint only for this guy (Snap-On). Definitely diving in to take a quick look at the linear rails and screws. I also had noticed the springs just the other night and you bringing this up confirmed what I had seen and I thank you for that.
Robin wrote: Slightly concerned to find the bed heater element wasn't plastered with thermal transfer compound. Couldn't find my tube so it will have to wait until the replacement arrives. I know the element is enclosed and the heat can only escape to the surrounding aluminium but without the compound the element will run a lot hotter than it needs to. It should not discolour and mine has.
Robin
Never even thought about this but it makes sense, using the thermal transfer compound will run more efficiently I imagine? Does anyone by chance know if thermal transfer compound is the same stuff we use between heatsinks and processors?? I think it is!?

Thanks again for all the great info and thanks for giving back!! :)

mr6k
Posts: 327
Joined: Tue Feb 14, 2012 11:47 am

Re: Yet another fan housing

Post by mr6k » Sun Jan 06, 2013 11:39 am

Thermal compound is indeed the same material as that used between many electronic components, power transistors/FETs/Processors etc and their heatsinks. Usually a white gooy material.

Alcohol and many solvents of organic materials will not help "clean" perfboard. As noticed you need Acetone or MEK, both are good solvents of ABS and when you wipe the perf board with them they are actually dissolving the ABS residue on the perfboard, or in the perfs and spreading a very fine film of ABS on the perf board, this acts as a great anchor to the print. I use Kapton tape, which on its own is an excellent "grabber" of the print. If I need extra grab I wet a cloth /tissue with acetone or MEK then wipe it momentarily across an old block of ABS then across the Kapton tape. Again this places a very fine film of ABS on the tape. Note, none of this will work with PLA, as PLA will not dissolve in Actetone or MEK. I guess it might in Alcohol, but I rarely use PLA so cannot comment.

cheers

Peter

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Sun Jan 06, 2013 8:38 pm

I added the thermal transfer guck to the heater, the sensor and the table joint. I also added 3mm of cork to the bottom of the table to try and lose that hot spot over the element.

Time taken to heat the table from room temperature to 100 C on the screen went up from 12 to 13 minutes. Digi thermometer pointed at the perf board gave me 83-91 C stabalising at 90-95 C after another 4 minutes. Major improvement heatwise, but will extra heat help it hold those rafts down?

I wasn't actually expecting it to take longer to heat up, I think it is because the heater block is now having to carry the table with it as it warms up. It cannot get ahead of it.

Neal
Posts: 32
Joined: Sat Oct 06, 2012 2:26 pm

Re: Yet another fan housing

Post by Neal » Sun Jan 06, 2013 11:25 pm

I coated the sides of my heater block, and the bottom of the bed, with high temperature silicone. I was hoping for a faster bed heat up. I think this will also create a more evenly distrubited heat profile acorss the bed. Because the heater is so far from the edges of the bed I think the silicone coating will reduce the amount of heat that is lost, at least from the bottom surface of the bed. This should help the outer edges of the bed attain a higher temperature.

Attached is a photo of the bottom of the bed with the black silicone applied.

Unfortunately I did not time the "heat up" time before adding the silicone but I have recorded it after the silicone was applied. It took 14 minues and 32 seconds to get the bed from a cold shutdown (not used overnight) temp of 14.9C up to 100.3C. This time was measured while a piece of standard window glass was attached to the top of the bed with the bulldog clips. These temp readings are taken from the display on the maintance window. Maybe someone with an original setup could record and post the time required to climb to 100C.

I haven't used any thermal paste on the heater or on the heater to heating block surface but I plan to do this.
Attachments
003.JPG
Silicone under bed
003.JPG (215.8 KiB) Viewed 21530 times

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Sun Jan 06, 2013 11:54 pm

Test 1: Octave ABS, 0.2mm, fine, 3 surfacing layers, tightest infill pattern, perf board with a 2mm raft

The piece I couldn't make. The piece that started me redesigning this Afinia and looking for help. I don't like to fail :mrgreen:

The base is 4.5" x 1.25", it's 3.2" tall and chunky. Every time I try to print it, it lifts at the corners and bends like a banana.

I nearly got it once with 2 surfacing layers, a wide open fill and Bulldog clips holding the raft down at either end.

But things are different this time...

It is glued down to the perf board with an acetone ABS wash.
The table is hotter with a more even heat distribution.
No clips, I've moved it over the hottest part of the table and they don't fit.
Zero downdraught from the fan to shrink it too soon.

So far it has reached layer 57 of 317 and all four corners are still down on the perf board. I have hopes but won't know if I've won until it's cold.

If it does stay straight I won't know why because I have changed too many variables. If it works I shall probably become superstitious and keep a whole mess of unneccesary modifications rather than take a chance. Life's like that.

Over 5 hours to go, I was up early for a muzzle loading clayshoot, it's nearly midnight over here and I'm off to bed.

Tomorrow in the cold light of dawn, I will know :shock:
test1.jpg
test1.jpg (65.08 KiB) Viewed 21534 times

User avatar
wackojacko
Posts: 667
Joined: Fri Nov 12, 2010 5:46 am
Contact:

Re: Yet another fan housing

Post by wackojacko » Mon Jan 07, 2013 12:58 am

Hi Robin

I would like to try your object with our plastic. If you wouldn't mind please share the file.

Thanks
Bruce
Bruce
http://www.3DPrintingSystems.com

Need help with your printer, check out our Youtube channel.
https://www.youtube.com/user/3dprinting ... /playlists

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Mon Jan 07, 2013 12:34 pm

wackojacko wrote: I would like to try your object with our plastic. If you wouldn't mind please share the file.
You think your filament can do better? http://www.robinhewitt.net/PlasmaBase.stl

The raft held but when I undid the clips the part bent the perf board, when I removed the perf board the part bent some more.

It's not terrible, in fact it is freakin' amazing how close it gets. If you eject ABS from the mould too soon it bends, if your wall thickness goes above 1.5mm it can bend lots. The Afinia doesn't have the luxury of a mould to constrain the part while it cools so what chance does it have?

The bottom bends up 1.2 degrees on the left and 1.6 degrees on the right. The ends are about 0.4 degrees off vertical.

It has certainly improved, but could it improve some more?

Pics: The Inclinometer is set to the machine bed the part is resting on and show 0.0 degrees both times
test2.jpg
test2.jpg (104.12 KiB) Viewed 21498 times
Attachments
test3.jpg
test3.jpg (102.43 KiB) Viewed 21498 times

User avatar
scubamatt41
Posts: 303
Joined: Thu Oct 11, 2012 7:47 am
Location: The Netherlands

Re: Yet another fan housing

Post by scubamatt41 » Mon Jan 07, 2013 2:01 pm

Hi Robin,
Maybe you could try Polycarbonate or PMMA printboard vs perfboard just for experiment.
Always breath out on the way UP

Blown454
Posts: 49
Joined: Fri Oct 12, 2012 6:13 pm

Re: Yet another fan housing

Post by Blown454 » Tue Jan 08, 2013 11:28 am

Robin,

I know exactly where you are here. Have been there hundreds of times and have boxes full of failed prints to show for it. When I had a very large piece that was over 2x the size of the Up!'s bed (using another 3d printer obviously), the only way I could get close was to use an enclosure. You can use a cardbox box, big tupperware type thingy, etc. (try to keep as much of the electronics outside of box, good thing most important stuff is at bottom on Up!). I know how discouraging it is to think you have a perfectly flat print to find when you take the clips off and the whole perf board flexes and then you see the part is warped.

The other way I got decent flat prints was changing from ABS to PLA for those parts. It warps much less and may be an option for the larger prints. The enclosure is worth the effort if you have the time! Good luck! -john

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Tue Jan 08, 2013 3:41 pm

Blown454 wrote: the only way I could get close was to use an enclosure.
We think the same way :mrgreen: It isn't just a matter of getting it to stick, although that does help, it's a matter of keeping as much heat as possible in the part and in the table until it is all finished. Not quite sure I see the logic in this theory, but it is something to try.

Afinia do warn against cold and draughts which is why I was surprised to see the fan blowing downwards. Seemed a bit self defeating.

I will try a cardboard box and if it works do something nifty in perspex.

I am also going to try the 12 ohm nozzle heat reducer for Octave filament. This will add less heat, completely contrary to the more heat theory above, but the only way to progress is to try different things and hope I hit on some combination that works.

Also thinking about adding mirrors to redirect radiant heat from the nozzle element sides down on to the workpiece.

marto
Posts: 87
Joined: Wed Nov 14, 2012 2:01 pm

Re: Yet another fan housing

Post by marto » Tue Jan 08, 2013 6:08 pm

You can modify the part to make it easier to print. If you can eliminate the really big long strips it will make it easier.

This vid is a good description of the idea. https://www.youtube.com/watch?v=YhfrHo53K18

I was printing some robot shells which were ~ 100mmx 70mm and adding a heap of holes around the problem areas improved the warping a lot.

Also I would retry with UP! filament. The one lot of after market filament I have used was terrible. It literally could not print anything without huge amounts of warping.

Steve

mr6k
Posts: 327
Joined: Tue Feb 14, 2012 11:47 am

Re: Yet another fan housing

Post by mr6k » Wed Jan 09, 2013 12:08 am

Just a thought, and I have never tried it, but where many are suggesting an enclosure to keep the object and table temperature up, why not alternatively try a fan heater blowing warm air onto the model and table, but of course not into the electronics. I know it would be directional but it might be easier than encapsulating the entire print area.

cheers

Peter

Blown454
Posts: 49
Joined: Fri Oct 12, 2012 6:13 pm

Re: Yet another fan housing

Post by Blown454 » Wed Jan 09, 2013 12:58 pm

I'm using that switch from Octave.com as well with Octave filament at the moment. It lets you extrude at a more reasonable temp. with the aftermarket filament so definitely worth the shot. Lots of good ideas to try and some seem like they will never work when thinking about them, you are close though! :)

Robin
Posts: 113
Joined: Sat Dec 29, 2012 6:59 pm

Re: Yet another fan housing

Post by Robin » Thu Jan 10, 2013 7:38 pm

Blown454 wrote:I'm using that switch from Octave.com as well with Octave filament at the moment. It lets you extrude at a more reasonable temp. with the aftermarket filament so definitely worth the shot. Lots of good ideas to try and some seem like they will never work when thinking about them, you are close though! :)
I have to lay some circuit boards this month and if i have some spare room on the panel I will recreate the little PCB at the top of the print head, dump the unnecessary connectors and see if I can pull some Volts to drive my vanity LED's :mrgreen: The headers and sockets are by JST and available from Farnell. I can then add the 12R resistor with a link so I can short it out by jumper or switch.

I made a sizeable two part enclosure today and it came out fine and dandy, no corner lifting, no warping when I separated it, the two halves went together perfectly. One difference was I was printing on 120mm tape over perf board with an ABS Acetone wash and no raft. Those first 3 layers of the raft are really open weave and cannot do anything to help the bed keep the part warm. It seems to improve every time I do anything to keep it hot.

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

Re: Yet another fan housing

Post by roller » Sat Jan 12, 2013 12:05 am

Just a thought, and I have never tried it, but where many are suggesting an enclosure to keep the object and table temperature up, why not alternatively try a fan heater blowing warm air onto the model and table, but of course not into the electronics. I know it would be directional but it might be easier than encapsulating the entire print area.
We tried this with repraps - unfortunately it result in another hot side and cool side (so as well as bottom being hotter than top, front is hotter than back for example) and so you get similar warping issue but in a new plane and often increased breakaway issues during printing.. I tried with a heated vortex blowing all around the part - it sometimes improved print and other times made them worse as heating was inconsistent depending on print shape.

Interestingly the new Cube printer has supposedly done away with heating completely having produced a new ABS formula to reduce warping and using an adhering surface for prints.

Post Reply