I am looking for a 3D printed re-spooler contraption that works reliably and spools filament onto a second spool as neatly as a manufacturer’s spooling machine. Is there such a model published that works better than a Pastamatic?
Please share your GOOD and particularly your BAD experiences, including re-mixing mods.
These models use a lot of filament to print, and I read countless posts by people disappointed by the results after their efforts to print a large device. The most downloads clearly does not imply the most efficient re-spooler.
The best I could find was the Jarvis Winder. And even that doesn’t seem to work as good as I would like it. The newer Jarvis Pro Winder looks better but isn’t open source.
So I fear there currently isn’t a really satisfying solution available for hobbyists.
Yes Jarvis Pro is not for personal use. But Jarvis non-pro is open source, so you can build one yourself at much lower cost.
I don’t think that there exists anything really useful that you can just print.
You need to fine tune the winder to the exact width of the spool and the diameter of the filament. You also want controlled constant tension. That is just not possible with a purely printed device but necessary to achieve the quality of winding you are asking for.
Even the Jarvis Pro is possibly far from what manufacturers use. The predecessor seems to work but to have problems with tension control. Building a Jarvis non-pro is probably the closest you will realistically get to neatly rewound spools.
This looks great but I was hoping for something simpler - kinda like a Pastamatic or V3 Spooler or Bambu Lab Filament Spool Switcher & Winder that spooled the filament more precisely on the target spool.
Every one of the mechanical spoolers has multiple remixes that claim to solve various problems, so it is tricky to find the most efficient model.
I tried a couple others and I don’t like the lead screw ones. They don’t wind tight enough or don’t work at all and bounce around. Overall I’m happy with this respooler.
I’m also looking for a respooler.
But most, if not all of the printable spoolers, are not adjustable to the spool width.
Most systems seem to be designed to work on a full revolution/sequence of the rod/gear before going back. So this just gives a fixed spoolwidth.
The Bambu winder seems a bit more adjustable. But all is fixed to this full gear rotation. Although it should be a bit customizable with adjustable teeth. But seems the top gear is holding everything firmly in place and it’s not designed for easy switching/replacement.
The V-spooler looks great, especially because it’s compact design. But it’s completely over designed, needs a lot of resources and I actually don’t like the rod either, which is also not customizable.
While it would be nice to have something as accurate as the machines the big filament suppliers use, I think it’s possible to get something as close as to make no difference, though doing it without electronics would be very difficult indeed.
So let’s consider a few things:
The exact width of the spool could be entered or supplied by a sensor, perhaps lidar.
The diameter of the filament, I think, can be safely assumed to be 1.75mm.
Constant tension could be supplied by monitoring the takeup reel motor current. Alternatively, rpm of the takeup reel motor could be used to adjust the tension by moving a metal disk closer to or farther from, magnets mounted in a printed disk .
As for positioning the filament guide to wind it onto the right place: If we know the width of the spool and know where the edge is, we can set that value for the number of steps for the following.
A stepper motor could be used to move the guide 1.75mm per revolution, reversing at the end of the travel, and auto-correcting at that time.
I offer these ideas in the hope that someone (or a group of someones) can come up with an implementation and produce it.
The guys from Jarvis winder did exactly what you describe and quite a bit more. And they even made it open source.
In the second version, instead of measuring tension by the motor current, they use a lever, which seems a far better idea to me. But I guess, the open source version is good enough.
So what you ask for is already available, you just have to follow their build plans.
I did print this one, I’m happy with the results so far. The only hiccup was when I broke the axle with my drill being too strong (while the original filament got stuck until i disentangled it). I printed another one at full strength, made sure to stay gentle on the drill, it went ok. The manual handle goes fast enough though.
I screwed both part on a wood board and it’s now there stored on a shelf.
