Mini stealthburner gave me enough woes to start looking for alternatives. This is a Tap like kinematic coupling using ball bearings. Still early but promising so far. github.com/viesturz/tapchange...
Thanks for the update! This is a clever design, to use the bearings and rails in this way with the MiniSB. It reminds me of the original rollers used by the Kossel, which had a similar 45-degree angle to them with alternating directions - but of course, for a different application (delta tower sliders). Similar Q to @rainmotorsports - can you say more about what problems drove you to move past the original design? I like the Tap/Boop optical sensor compatibility you kept.
The issues list is long. To be clear using a rail has a major accuracy and thermal stability advantage. I'm hoping rev2 is good enough in accuracy and vastly better at everything else. Here it goes: High Insertion accuracy - it got really tedious to tune the XY after every small printer modification. Messy failure mode - hunting for balls after each failure sucks. And messy servicing in general. Gunk getting inside the carriage. The nozzle has to travel over the rail plug and retraction is not perfect, depositing hairs of filament onto the rail that find their way inside the carriage. 50mm vertical movement for undocking - that is all lost print volume under the tools. The rail and plug in center blocks from installing a nozzle wipe. The plug system actually works surprisingly well for what it is. Having less moving pieces is still a plus.
Hello dude! A pleasant result. Such a sound is satisfying. I don't know how plastic will behave over time (won't it lose the accuracy of positioning bearings?). I apologize for the impudence, but I have been making my own version of a quick tool change for a long time. In fact, I'm making a whole printer. SL_ST 3D. I think you can definitely find him. TH-cam removes links. For this reason, I will leave only the name.
Nice design. Is the magnet the only thing providing the vertical constraint? Or are the metal rails at a slight angle so that they squeeze the bearings together as it is engaged? I am working on a magnet ring type toolchanger for a Ratrig, and I am debating whether or not to rely on printed part's elasticity for engaging force or leave a gap between the magnets to maintain force after the kinematic coupling has made contact.
The metal rails are all parallel. The vertical contact is plastic to plastic with the magnet in the middle just providing extra sticking when fully inserted. Pretty much cargo culted from Tap. The plastic flex is present but very minimal and is something I want to remove as much as possible. Making the parts as beefy as the limited space allows, printing with 80% infill. Also will print them out of CF nylon once the design is more or less settled.
@@viesturzig fan of what you are doing! Just wanted to give my two cents. Don't use CF Nylon for this application. Nylons printed in FDM tend to creep quite a bit, which would be detrimental for this application. Personally I'd recommend using pure PC CF or one of the PC CF blends that are out there. While I am a fan of Prusament PC CF, I've heard that lately they've changed their formula, making it weaker. Many people, whose opinions I respect, swear by 3dxtech's offerings, so I'd recommend checking those out. Keep up the good work!
I was looking into using the THK rails with the caged balls to reduce catastrophic failures. But I was also looking into building a smaller toolhead so you having trouble is a bit concerning. Can you say where the problem was? Is the lighter toolhead popping off the rail during tap or is it not docking well or ?
See the other comment for the problems. It's a long list of small issues. The main issue with mini is that I hoped it would work without the anchoring flanges at the bottom, which failed spectacularly, destroying the plastic end cap of the carriage. I actually did have spares at hand but it gave me pause to think. I had the bearing kinematic mount idea sitting for a while and I wanted to see how a clean start would compare. The rail approach is not out of the window yet. For now this is purely for Mini. Oh and rev2 is half the weight from 81 grams to 41 grams, so makes the mini even lighter.
I'm loving the direction you're taking this, it's such a fascinating project to follow. Thanks for sharing! Keep innovating 💜
Thanks for the update! This is a clever design, to use the bearings and rails in this way with the MiniSB. It reminds me of the original rollers used by the Kossel, which had a similar 45-degree angle to them with alternating directions - but of course, for a different application (delta tower sliders).
Similar Q to @rainmotorsports - can you say more about what problems drove you to move past the original design?
I like the Tap/Boop optical sensor compatibility you kept.
The issues list is long. To be clear using a rail has a major accuracy and thermal stability advantage. I'm hoping rev2 is good enough in accuracy and vastly better at everything else.
Here it goes:
High Insertion accuracy - it got really tedious to tune the XY after every small printer modification.
Messy failure mode - hunting for balls after each failure sucks. And messy servicing in general.
Gunk getting inside the carriage. The nozzle has to travel over the rail plug and retraction is not perfect, depositing hairs of filament onto the rail that find their way inside the carriage.
50mm vertical movement for undocking - that is all lost print volume under the tools.
The rail and plug in center blocks from installing a nozzle wipe.
The plug system actually works surprisingly well for what it is. Having less moving pieces is still a plus.
Hello dude!
A pleasant result. Such a sound is satisfying.
I don't know how plastic will behave over time (won't it lose the accuracy of positioning bearings?).
I apologize for the impudence, but I have been making my own version of a quick tool change for a long time. In fact, I'm making a whole printer.
SL_ST 3D.
I think you can definitely find him. TH-cam removes links.
For this reason, I will leave only the name.
Nice design. Is the magnet the only thing providing the vertical constraint? Or are the metal rails at a slight angle so that they squeeze the bearings together as it is engaged?
I am working on a magnet ring type toolchanger for a Ratrig, and I am debating whether or not to rely on printed part's elasticity for engaging force or leave a gap between the magnets to maintain force after the kinematic coupling has made contact.
The metal rails are all parallel. The vertical contact is plastic to plastic with the magnet in the middle just providing extra sticking when fully inserted. Pretty much cargo culted from Tap.
The plastic flex is present but very minimal and is something I want to remove as much as possible.
Making the parts as beefy as the limited space allows, printing with 80% infill. Also will print them out of CF nylon once the design is more or less settled.
@@viesturzig fan of what you are doing! Just wanted to give my two cents. Don't use CF Nylon for this application. Nylons printed in FDM tend to creep quite a bit, which would be detrimental for this application. Personally I'd recommend using pure PC CF or one of the PC CF blends that are out there. While I am a fan of Prusament PC CF, I've heard that lately they've changed their formula, making it weaker. Many people, whose opinions I respect, swear by 3dxtech's offerings, so I'd recommend checking those out. Keep up the good work!
@@ExcitingCreations Thanks for the info. Super useful. Also a redesigned version is in the works that seems to be quite rigid in pure ABS.
Does this design change still allow the use of tap as a nozzle probe?
I would expect less precise and repeatable results with this
Yes, it's still the same. About to find out about the precision.
Can you post a .step of the mating components for the carriage and rail? very curious to see how it works. Thank you.
See the repo in description. It has STL and Fusion 360 files. Let me know if that is not sufficient.
@@viesturz Ahh, thanks. Fusion files are no good for me but the .stl will work fine. Thanks!
I was looking into using the THK rails with the caged balls to reduce catastrophic failures. But I was also looking into building a smaller toolhead so you having trouble is a bit concerning. Can you say where the problem was? Is the lighter toolhead popping off the rail during tap or is it not docking well or ?
See the other comment for the problems.
It's a long list of small issues. The main issue with mini is that I hoped it would work without the anchoring flanges at the bottom, which failed spectacularly, destroying the plastic end cap of the carriage. I actually did have spares at hand but it gave me pause to think.
I had the bearing kinematic mount idea sitting for a while and I wanted to see how a clean start would compare. The rail approach is not out of the window yet.
For now this is purely for Mini.
Oh and rev2 is half the weight from 81 grams to 41 grams, so makes the mini even lighter.