3D Printer Advice Please.

GremlinRC

FT_Nut
Hi Guys,

I have a Wanhao Duplicator i3 just over a month and a half. I've been extremely happy with it to date. However recently I've been experimenting with thin wall printing with the ambition of printing some 3dlabprint planes. I've had moderate success by keeping the layer height down but still not perfect.

What I'm wondering is would upgrading the original MK10 extruder/hotend be a worthwhile investment? For example, I can get the full flexion upgrade for $150. I've been reading some great reviews but they all seem to focus on the ability to print a wider variety of materials. Few focus on whether its a worthwhile upgrade for printing plain ol' PLA.

The other question is regarding slicer software. Currently I'm using CURA which came with the printer. It's been fine for standard printing. I note that Simplify 3d gets great reviews, but I'm struggling to see how it would make the killer difference.

So my question is which one would you go for 1st? Simplify3d or the Flexion Kit? Also are there are any other alternatives to the flexion that you think I should consider?

My problem is I just don't know enough about all of this yet to make a good decision. My instinct says go for the extruder but for the moment only one of these options is in budget.

Thanks in advance,

Dave.
 

CartCurt

Member
Dave,

I bought my Wanhao Duplicator i3 V2 about a year ago so I could build my MPCNC. I found that a few upgrades to the printer helped improve the print quality. I installed the gantry brace and added glass to the print bed.

Have you been to http://3dprinterwiki.info/wiki/wanhao-duplicator-i3/? There is lots of good information and some very smart people on the site. You can get information on mods and calibrations that can improve the print quality of the Wanhao.

My recommendation would be to look towards getting Simplifyed3d. Reviews I have seen say that if you are going to print PLA only then the Flexion Kit isn't necessary. It can print PLA but not enough better than the MK10 extruder to justify the cost. Check out MakersMuse on YouTube. He installed and is using the Flexion extruder on his Cocoon Create (Wanhoe clone) and he likes it very much.

Hope this helps.

Curtis
 

jhitesma

Some guy in the desert
Mentor
I have a Folger 2020 i3 not a Wanhao - but they're both i3 derived designs so very similar. My stock hotend was a Mk8 not a Mk10 but again the changes between them are minor.

I upgraded from the MK8 style extruder/hotend to a E3D v6 hot end with a 3D printed extruder with a few of my own modifications done to it. There are plusses and minuses to the upgrade. My primary reason for upgrading was I wanted an all metal hot end so I could use more exotic filaments that require higher temps. I do a LOT of PETG and found that it works best at the upper limits of what a teflon lined hot end can handle. I even melted my liner at one point pushing things - you don't want to deal with that! I also wanted to be able to print Nylon and other materials that are even stronger but require temps that teflon liners can't deal with.

I picked an extruder based on what would be easy to adapt to my printer, would be direct drive (I didn't want to deal with gears or belts), and would have a well constrained filament path so I could also do flexible filaments well.

Overall I was able to get just as good of print quality (and in some cases better) with the with MK8 than I am with the current setup. But the new setup has opened up a variety of material options which I greatly enjoy.

I did do an intermediate upgrade from the stock plastic Mk8 to an aluminum version I got off ebay for $12. That helped improve print quality a LOT because it provided more consistent and adjustable pressure on the feed idler:

20160312_152933.jpg

I would have been very happy to stay with that extruder (at least until I started doing flexible filament) and just swap to the e3d hotend...but there was no good way to mount an e3d to the metal extruder.

So I went with this extruder to match the e3d: http://www.thingiverse.com/thing:825133

20160320_141817.jpg

I added two of my own upgrades to it:
An improved idler that has a tab on it so it's easier to release when changing filament: http://www.thingiverse.com/thing:1427181
An inductive sensor and layer fan mount: http://www.thingiverse.com/thing:1449709

Note - while I printed most of the extruder in PETG (I eventually want to enclose my printer for doing ABS and worried about PLA holding up to the temps) I did do the idler in PLA because I wanted it to be stiffer. I also did the sensor mount and fan in ABS because without a fan ABS was easier to print than PETG and I was worried about PLA holding up that close to the heated bed and hot end.

I also upgraded the teflon tube on the extruder since that photo by slightly lengthening it and trimming it into an inverted V at the top so the filament is fully constrained as soon as it leaves the drive gear. I've done a good bit of TPU flexible filament now and only once did I have an issue with it getting out of hand when I turned up retraction too high.

I've since printed a wades style geared extruder as printing with maker geeks PLA I was having issues that seemed to indicate I needed more drive force on my filament. I haven't installed that extruder yet though as I've yet to come up with a good way to mount a layer fan and my inductive sensor to it. And I've since solved the issue the maker geeks PLA was giving me by using a tiny clothespin to hold a bit of paper towel with a few drops of vegetable oil on it over the filament just before it enters the extruder. I've only had to do that with the MG PLA, no other filament I've used has had that issue. Not sure if it's because of the all metal extruder or just something about MG's PLA but I've heard it's fairly common with their PLA. (Note: their PLA is kind of odd and likes much higher temps than most....the roll of white I just got they recommend 235c! But at $15 a roll delivered when you buy their grab bag and don't care about color, or in their monthly maker box it's a killer deal.)

So - the "upgraded" hot end and extruder has widened my printing ability...but I wouldn't say it improved my quality any. Dialing in my slicer did far more to improve quality. And that's where I have an issue with 3dlab prints.

I really want to print one of their planes. But I've yet to get a usable print of any of their demo files.

The issue is I use slic3r and their designs don't slice in slic3r for some reason. All the internal structure gets lost.

I've tried CURA several times (with both hot end/ extruder setups) but can't get decent prints with it and get frustrated with it as a result.

The big issue is I use single wall prints to dial in my slicer. The logic is it's the best way to make sure things are going to do exactly what you expect and be dimensionally accurate. Most of my printing is functional parts not decorative stuff so I want the most accurate parts I can get. I use a 20x20 hollow topless/bottomless test cube designed with a 0.48mm wall thickness to test. I print with a 0.4mm nozzle which means 0.48 is the ideal width for a single wall extrusion. The plastic expands as it comes out of the nozzle, so you usually want to print with an extrusion width that's 120% larger than your nozzle size, if you want a different thickness you should run a different sized nozzle. And if you run a different sized nozzle you'd want to aim for a different sized thickness.

The basic writeup on the calibrations I do are here:
http://northwoods3d.weebly.com/blog/filament-calibration-part-1
http://northwoods3d.weebly.com/blog/-flow-rate-how-much-plastic-is-coming-out

I don't agree with everything on her blog, but her filament calibrations are one of the best write-ups I've seen of the method.

The problem is this doesn't work in CURA. If I try to slice that test cube in cura - I get an empty file. It ignores the single wall thickness!

This is why 3dlabprints confuses me...with a part designed for single wall printing cura ignores it but slic3r does great. But with 3dlabprints stl's the opposite is the case and cura will slice them but slic3r chokes on them. So I have a really hard time getting CURA to do accurate prints because I can't get it to slide single wall test pieces to confirm that I'm extruding the amount of filament I'm asking for.

As for simplify3D...I can't justify spending more on slicer software than I did on my printer when I can get such great results on everything except 3dlabprints designs with slic3r which is free and open.

With the calibration method I used I can print the Make 2015 shootout tests pretty much perfectly: http://www.thingiverse.com/thing:533472 Specifically on the negative space tolerance test I can remove all of the pins (though the tightest one does leave a dent in my finger from the pressure required to get it out.) I can print all kinds of print in place designs with no issues and get extremely accurate dimensions on my prints. The error from my $15 harbor freight calibers is larger than the error from my printer at this point :D

I plan to spend more time with cura wrestling it into submission so I can do one of the 3dlabs planes at some point...but my project list is pretty long already ;)

Back to the original topic. I don't think a new extruder would give an increase in quality like you're hoping for. More time spent calibrating and dialing in your slicing software is more likely to achieve that goal in my experience. Though I'm still struggling with just how to achieve that in Cura.

(FWIW the biggest issue I have with cura is that the same designs that print fine for me in slic3r refuse to stick to the bed when I slice them with cura?! Which makes no sense to me as the settings are pretty much the same and I can't see anything about the gcode that would indicate the first layer should be any worse. Honestly I think cura is just cursed for me :D )
 

GremlinRC

FT_Nut
Jason and Curtis. Thanks to you both for your generous input.

Curtis: I've already done a couple of the mods, the Z brace being the most worthwhile. Also the spring cups are an excellent addition to help with bed levelling consistency. Thanks for the wiki link. Looks like there is loads more there to consider. I did some reading on S3D this evening and from what I can gather it adds a lot more features than CURA but I'm still not convinced its gonna be the big difference.

Jason: I'm not sure why CURA wouldn't slice your thin walls, maybe it was down to an older version, my version which funnily enough doesn't seem to have a version number that I can find, slices the thin walls no problem. From what both of you are saying, I'm thinking the right move, right now is to keep my loot in the pocket until I've done more exploring.

I did a couple of test prints this evening. This is part of a wing section @ 0.2 layer height.

Img_1463.jpg

You can see the problem. From what I've read the problem here is either under extrusion or bad adhesion. Tell you what though even with this bad print, I'm impressed with the strength of the part. I haven't tested it to destruction but I did put far more force than you would get in normal flight and It stood up to it.

Anyway I tried a couple of wingtips again one at 0.2mm layer height and another at 0.1mm. Clearly the 0.1mm one came out MUCH better. It's too late now to try a larger part as its past midnight as I type, I'll give the same wing section a spin at 0.1mm tomorrow. If it works ok, I may just have to settle for this taking twice as long to print.

Img_1468.jpg

As for the slight gaps even in the 0.1mm print, I think the slicer layer view explains that. I'm sure if I printed these flat curves at 0.5mm the gaps would go away.

slicer.JPG

Again thanks to you both for taking the time to help.
Cheers
Dave.
 

jhitesma

Some guy in the desert
Mentor
The thin wall is a known limitation of CURA that the lead developer has repeatedly said he won't fix:

https://github.com/daid/LegacyCura/issues/606
https://ultimaker.com/en/community/3787-cura-13065-mac-fails-to-slice-thin-wall-calibration-object
https://ultimaker.com/en/community/...slice-this-simple-calibration-piece-correctly

Just tried again with the lastest Cura for Mac 2.3.1 and it still fails to slice a thin wall object:

Screen Shot 2017-02-09 at 8.45.41 PM.png

It looks ok in the preview...but note the specs on the lower right. It's not slicing it at all.

The developers suggest using a solid cube and then telling CURA to only print it as a single wall...which to me is a big limit on what you can design and print. I probably will eventually resort to that to get a quick easy single wall print I can use to dial in extrusion with cura...but it's just an ugly hack IMHO. Of course the cura dev feels that adjusting your extrusion multiplier based on the results you get is a hack - I do do the calibration he suggests of measuring the amount of filament going in. But there are a lot of things that can't account for and it just can't give the accuracy I'm after. But if I want to print a 3dlab prints plane then I'm going to have to make it work :D

Looking at the STL files from 3dlab in CAD it appears that they're just shells - with no thickness defined. Which is always something tricky to slice. One of the YT channels I follow about 3D printing just had a good video about shells this week and the issues they can result in:


I learned about the difference layer height can make on curved surfaces when I designed a replacement button for my accucell charger:
http://www.thingiverse.com/thing:1360018

It has a very slight curve on the top. But if you print it with anything more than 0.1mm layer height it doesn't come out since it's smaller than the resolution can resolve.

The issue on the bigger flat section definitely looks like under extrusion. If you can accurately print a small 20x20 single wall test cube and get the wall thickness you're asking for from your slicer then I would look into extruder issues. If you can't print a 20x20 single wall test and get the wall thickness you set then I'd work on calibration first.

Note - slic3r will by default try to auto set the extrusion width but will usually get it a bit wrong. It has an option to set it and setting it to 120% of your nozzle size will almost always give more consistent better results than letting it autocalculate the extrusion width. Cura...is just kind of wonky about all of this :D

If you can do the small single wall test (I emphasise small since that way you can do a lot of quick tests over and over to try new settings instead of spending an hour or two doing a full piece.) and still get the gaps then I'd be looking for something on the extruder like:

1 - Filament binding. I originally just used the length of 2020 extrusion that came with my printer as a spool holder. Every now and then I'd see that kind of thing happen. I ended up printing a set of rollers that used small bearings (one of the nice things about the Folger kit was it came with TONS of extra parts like screws and bearings which I'm still using for projects) and that helped eliminate a lot of little intermittent issues.

2 - drive gear. It's not uncommon on some printers for the extruder drive gear to come loose and occasionally slip. There's a lot of pressure going on in the nozzle and the gear slipping can result in bad extrusion.

3 - Hot end - I also had issues when I accidentally melted the PTFE liner in my hot end. What would happen is the filament would start to bond with the melted PTFE and that would cause intermittent issues with extrusion.

4 - Lubrication - some filaments (like this MG PLA I've been using lately) really do better with a tiny bit of oil applied to them. Sometimes you can get by with just rubbing a bit on the end of the filament before putting it in the hot end - but usually people use a bit of towel or cotton ball around it to clean it and apply a tiny bit of oil. Apparently this is more common with all metal hot ends where the molten filament can start to stick to the inner walls of the hot end like it did with my molten PTFE lined hot end.

5 - Heat creep - This is another issue that can arise in some hot ends. If the heat break between the hot side and cold side isn't effective then in longer prints the filament can start to get soft higher up than intended. Sometimes even to the point where it can soften enough that the drive gear will start to chew it up and cause more slipping.


You could also have an issue with your layer height not being optimal for your Z axis mechanics. Prusa has a calculator on his site to determine optimal layer heights: http://www.prusaprinters.org/calculator/#optimallayer It could be the mechanics of your printer are such that at the layer height you're using error builds up over a number of layers and eventually results in issues. The calculator will give alternative layer heights that match the mechanics so you don't have any errors. Instead of 0.1mm you may have better results at 0.12 or 0.098 instead of going all the way to 0.05 if you can find the specs of your steppers and z drive then the calculator will let you figure it out.

There are also some good calculators here that help explain extrusion width and figuring out the optimal extrusion width for a given nozzle size and pick a layer height that's optimal for strength: https://nathan7.eu/stuff/RepRapCalculator/RepRapCalculator.html
 

jhitesma

Some guy in the desert
Mentor
@jhitesma,
I am obviously doing my math wrong if your printer costs less than Simplify3D! The Folgertech web site shows your printer at $269 https://folgertech.com/collections/3d-printer-full-kits/products/folger-tech-reprap-2020-prusa-i3-full-aluminum-3d-printer-kit and the Simplify3D site shows $149.... https://www.simplify3d.com/buy-now/

Cheers!
LitterBug

Yeah, I meant to say a good portion of what my printer cost not more ;)

I'm just a huge fan of open source. And can't justify $150 on a closed no-demo commercial package when there are excelent open source options available.
 

LitterBug

Techno Nut
Moderator
Yeah, I meant to say a good portion of what my printer cost not more ;)

I'm just a huge fan of open source. And can't justify $150 on a closed no-demo commercial package when there are excelent open source options available.

I kinda backed myself into a corner... The printer I have (xyzPrinting Da Vinci) is very proprietary in nature, and the only slicer that would easily work with it (ie without hacking the firmware) is Simplify3D. Hindsight being 20/20, would not buy this printer again. Gives me a good starting point, but support is a nightmare. (broke on first day after only 10 hours of printing)

Cheers!
LitterBug
 

GremlinRC

FT_Nut
1 - Filament binding.

Yes this is something to look at, I have noticed for example that sometimes the filament does not always uncoil smoothly and sometimes uncoils suddenly maybe introducing a jerk.

2 - drive gear.

I have checked several times and ensured the drive gear is running free, clear from debris and the lock nut is tight against the flat spot.

3 - Hot end

Check that too. All seems ok. Maybe though I wonder it there is some room in there for the filament to buckle as its being pushed through the hot end.

4 - Lubrication

Def gonna give this a try.

5 - Heat creep
I think this is ok, as when I remove the filament I can see tiny marks where the teeth marks engaged the filament, these are sharp and crisp. I would expect if heat creep were the problem these would be softer or non existant.

You could also have an issue with your layer height not being optimal for your Z axis mechanics. Prusa has a calculator on his site to determine optimal layer heights

Def gonna check those calculators as it makes complete sense.

Jason thanks a lot you have given me a lot to consider. I will go through all of this carefully.
 

GremlinRC

FT_Nut
I'm just a huge fan of open source. And can't justify $150 on a closed no-demo commercial package when there are excelent open source options available.

Yep me too. I just don't see why they don't provide a demo. Even if they limit the number of gcode lines that the demo can procude to a very small number or limit to a fixed number of outputs. Can't be too hard. I also read a couple of bad stories about customer service and activation problems.

The real kicker is I'm still struggling to establish what advantages S3D will give me over my plain old CURA. I really can't explain why your cura won't slice thin walls and mine does. The difference is that mine is the CURA Wanhao edition which came with the printer.
 

rockyboy

Skill Collector
Mentor
Yes this is something to look at, I have noticed for example that sometimes the filament does not always uncoil smoothly and sometimes uncoils suddenly maybe introducing a jerk.

This can be a really killer for print quality - it may not seem like much, but the increase in pressure on the filament when it's starting to stick does impact the extrusion amounts, and can also pull your Z axis up enough to impact adhesion if there is any backlash in your z axis lead screws. This was also a complete killer on my old PrintrBot with the long cantilevered extruder head, and has even been a problem on my Prusa i3 with two rolls of filament that got tangled on the roll by the factory (%!@%$ MakerBot filament - never again!).

Setting up a spool holder that runs on ball bearings can help a whole lot - as can giving a sticky roll of filament a light wipe down with a paper towel dampened with mineral oil so it comes off the coil without tripping over itself as much. For my Makerbot rolls, I ended up spending a good amount of time on each roll carefully tracing the filament path across the roll and trying to determine where the end went under an earlier loop. Total pain in the butt - however it did fix the problem combined with the mineral oil and I've been printing OK with those rolls again in the last week or so.
 
Last edited:

jhitesma

Some guy in the desert
Mentor
The real kicker is I'm still struggling to establish what advantages S3D will give me over my plain old CURA. I really can't explain why your cura won't slice thin walls and mine does. The difference is that mine is the CURA Wanhao edition which came with the printer.

Well, S3D does have great supports - though the prusa team is actively working on improving that in slic3r (they're actually completely re-writing the support code according to their comments in their github repo.) S3D also has the multiple process stuff that is pretty slick - but slic3r can do that as well just not quite as easily since you have to create additional STL's to define the parts of a print you want to have treated differently. I used this when I did my 3D printed mini quad so I could have dense infill in the center were it needed strength - but very sparse infill towards the outside where less strength was needed. Here's the slic3r blog post about it: http://slic3r.org/blog/modifier-meshes

The real "nice" thing about S3D is that it's more beginner friendly and can give acceptable results right out of the box with most printers quicker and easier than slic3r or cura. But...for a power user who isn't scared of piles of settings to adjust and tweak (and actually welcomes them since they help achieve my goals) slic3r is amazing.

I'd love to demo S3D - but they sell enough without offering that option so I'm in no rush to try them. Yeah, they have a return window but some of the support stories I've heard make me not trust that option. I could also just grab a hacked version to try...but if that's what it takes it's not worth my hassle.

As for cura. Have you tried slicing the demo thin wall test cube I linked? This one? It seems that Cura is ok with zero thickness shells like 3dlabprints uses...but it chokes on legitimately thin wall items that have thickness but less thickness than Cura wants to deal with.

I'd be surprised if Wanhao actually modified cura extensively enough that their version can do it since it goes right to the heart of the core algorithms that Cura is designed around according to their lead developer. Hence why they've been saying they won't fix it for 4 years.
 

jhitesma

Some guy in the desert
Mentor
This can be a really killer for print quality - it may not seem like much, but the increase in pressure on the filament when it's starting to stick does impact the extrusion amounts, and can also pull your Z axis up enough to impact adhesion if there is any backlash in your z axis lead screws. This was also a complete killer on my old PrintrBot with the long cantilevered extruder head, and has even been a problem on my Prusa i3 with two rolls of filament that got tangled on the roll by the factory (%!@%$ MakerBot filament - never again!).

Yeah, I had a bad roll of MakerGeeks this summer:

20160730_143121.jpg

20160804_183243.jpg

It's hard to see...but the winding on that filament was terrible. Looped under and through itself. No clue how they managed that. Sent them an e-mail about it and the apologized extensively and said they'd make it up to me on my next order...but I started their geekbox and now have more filament than I can use so haven't done a real order again. (in fact just got an e-mail today that the next geekbox is on it's way and I still have rolls from the last two that I haven't even tried yet!)

I had to babysit that roll while it printed and manually untangle it as I went. Otherwise it would bind enough it would actually cause my X axis to skip steps and throw off the print big time :mad:
 

GremlinRC

FT_Nut
As for cura. Have you tried slicing the demo thin wall test cube I linked? This one? It seems that Cura is ok with zero thickness shells like 3dlabprints uses...but it chokes on legitimately thin wall items that have thickness but less thickness than Cura wants to deal with.

Ah yes I see now. It won't slice it at 0.4mm Shell thickness, but if I set to 0.3mm it slices ok and seems to print ok too.

Img_1470.jpg
 

jhitesma

Some guy in the desert
Mentor
Ah yes I see now. It won't slice it at 0.4mm Shell thickness, but if I set to 0.3mm it slices ok and seems to print ok too.

View attachment 82034

The problem is with a .4mm nozzle printing a .3mm wall thickness isn't going to work well. What does the wall thickness actually measure if you measure it with calipers? In fact printing a .4mm wall thickness won't work well since as I pointed out earlier the plastic expands as it comes out of the nozzle so the thinnest you can extrude is actually about 120% your nozzle size or 0.48mm

You can actually confirm this by free-extruding into the air and measuring the thickness of what you wind up with. It will be thicker than the actual hole in your nozzle. Instead of just using 120% to set your extrusion diameter you can actually measure it this way to account for variations in nozzle - but it can be tricky since gravity pulling it down will tend to cause it to come out thinner. So it's better to extrude a test bead at 0.5mm layer height or so and measure that to find out what your actual minimal diameter is.

Any smaller and you'd run into serious issues since you'd be pushing less material than needed to fill the nozzle.

So the calibration system I depend on basically comes down to:

There is a physical limit on how small your extrusion width can be based on your nozzle. Usually this is 120% larger than your nozzle diameter so for a 0.4mm nozzle your thinnest extrusion will be 0.48mm

You then tell your slicer that that's what you expect for your extrusion width and single walls.

Then print something with single walls and measure the resulting width to confirm you're actually getting 0.48mm walls.

If they're thinner then you need to increase your extrusion multiplier or decrease the filament diameter you're telling your slicer to use. If they're thicker then you need to decrease the multiplier or increase the filament diameter setting.

This is of course assuming that you've already done gross calibration and can confirm that your extruder is delivering the amount of filament you tell it to (the old measure 120mm on your filament, tell it to extrude 100mm, then confirm that there's 20mm to the mark test.)

The idea of the single wall calibration is to fine tune for filament variations. Measuring the filament diameter and setting that in your slicer accurately helps - but it can be tricky to get an accurate measurement on filament. Hence the single wall calibration prints to empirically confirm that you're getting what you expect from your slicer and adjust for the difficulty in accurately measuring filament diameter.
 

LitterBug

Techno Nut
Moderator
And all of this calibration is subject to change if you switch to a different type of filament. Different colors can even change the extrusion results. It's kinda like tuning PIDs to get the perfect tune on your printout. Change one little thing and it can all go to @#$@ LOL

Cheers!
LitterBug
 

GremlinRC

FT_Nut
The problem is with a .4mm nozzle printing a .3mm wall thickness isn't going to work well.


..but that's not the slicers fault surely? If you want to print *really* thin you gotta get a really thin nozzle right?
Also I guess the layer height is going to impact on the width you can actually print insofar as at smaller layer heights your going to end up squashing the material thus making it wider?


I added a filament guide and oiler and for now I'm just babysitting the spool to make sure it doesn't snag. I'm reprinting the same wing section at 0.2mm lh, so far so good.
 

GremlinRC

FT_Nut
I think I might have hit the jackpot. Just as I finished my last post I went back to the print, right at that moment, I saw the filament looped under on the roll and immediately the whole thing juddered as it released and guess what. There's one of those gaps (not a huge one) right at that point in the print. I guess there's not much you can do about the filament getting looped under like that except maybe take as much as you need for a thin wall print, off the roll so it doesn't snag. I have a couple of nice bearings from an old spindle, so I think I'll conscript those to make a nice filament roll holder.
 

jhitesma

Some guy in the desert
Mentor
I thought I took a photo of it...but can't find it.

But what I did to prevent the filament unspooling and getting tangled was that I taped a popsicle stick to the side of my printer (the spool hangs on the side) so it just barely gently rubs against the bottom of the spool. That way any filament that tries to get loose can't escape.

Ideally the spool should be over the printer since that way you have the lease resistance against it...but I've yet to get around to setting mine up like that. If yours is like that you could rig a popsicle stick under the spool.

The only real issue is it doesn't work on on various sized spools. Like the tiny spool my Taulman 910 nylon came on. But that filament doesn't try to come off by itself as bad so it's not a huge issue.
 

dkj4linux

Elite member
Ignorance is bliss!

I'm sure glad I got a 3d printer and built a machine or two with it before I ran into you guys. This is all so neat and great... but really overwhelming and intimidating to me. I'd never have gotten started doing this stuff if I'd been aware of all this stuff first... :(