A3X Driven Support Shaft (TPU) by Aesth3tix

Aesth3tix

3D Printer>3D Printer Accessories

Print Profile(2)

All

P1P

P1S

X1 Carbon

X1E

A1 mini

TPU 95A_A3X Driven S Shaft r00_0.4mm Nozzle, 0.16mm Layer, NO BRIM

Designer

1.2 h

1 plate

4.9(27)

0.16mm layer, 6 walls, 20% infill, A1 mini

1.3 h

1 plate

Boost

454

942

476

306

Released

Description

After weeks of testing, I happily introduce the A3X Driven Support Shaft!

I was tired of seeing tangles in my AMS randomly at the worst possible times, so I set out to find the cause and hopefully, a solution. What I found was my AMS driven support shafts on slots 2 and 4 were not always “gripping” the spool as the machine began to roll the filament back up. The timing was, ever so slightly, off which would cause a bird nest of filament inside the AMS. Sometimes the machine could recover, however, other times, it could not. Therefore, I would be left with an error code and a paused print job. To remedy this, I installed a brand-new driven support shaft from Bambu Lab, but it did not fix my issue and slot 4 (I used it exclusively to determine the fix) continued to give me trouble. Thus, I set off to create my own driven shaft and to be quite honest, I was surprised when it actually worked out after having the issues for so long.

The A3X Driven Support Shaft was born. Featuring flexibility of TPU 95A, 693-2Z bearings, smaller outer diameter to better suit larger spools, and quick-change grab for easy removal. This thing has worked wonders for me. Bambu spools just glide during loading and unloading and other, smaller 1kg spools work just as well. The smaller OD allows for cardboard spools with printed rings to move much more fluidly. With the stock shaft, there is barely any clearance between the spool ring and the lid. Check out the pictures and you will see an example of such. I had ditched the rings because of this and was just buying extra spools to transfer to but now I can actually use them and be worry free.

I strongly suggest using TPU 95A. I tested both PLA and TPU 95A thoroughly and found the TPU versions were much more efficient and quieter. For me, I used it as an excuse to print with the new TPU 95A HF from Bambu Lab and it prints great and is much faster than standard TPU.

PARTS LIST

(2x per shaft) 693-2Z bearings [3mm x 8mm x 4mm]

https://www.amazon.com/uxcell-Groove-Bearing-2080093-Bearings/dp/B07FVYPMPX/ref=sr_1_1_sspa?crid=14CZY41AZN41I&keywords=693-2z%2Bbearing&qid=1701034044&sprefix=693-2z%2Bbearing%2Caps%2C87&sr=8-1-spons&sp_csd=d2lkZ2V0TmFtZT1zcF9hdGY&th=1

(2x per shaft) 3mm x 12mm Dowel Pins

https://www.amazon.com/MroMax-3x12mm-Stainless-Rust-Proof-Elements/dp/B07ZF3478S/ref=sr_1_1?crid=134428WTTJJEU&keywords=m3+x+12mm+dowel&qid=1701033993&sprefix=m3+x+12mm+dowel%2Caps%2C106&sr=8-1

Assembly Tips

Dowel pins should be near flush to the end of the bearing (see photo). Tolerance here is very tight so the dowel cannot wobble. The dowel pins will likely have some variations due to the manufacturing tolerances so if it is not tight, grab another dowel. I had a few undersized pins in my order (as well as some oversized).
Slide dowel into the shaft until you see a very small gap between the bearing and the shaft (see photo). This insures smooth operation. If the bearing is mashed up against the shaft, it will not be near as smooth. (Ask me how I know lol) Again, tight tolerance to reduce any wobble.

Special Notes and Findings

Nope, still cannot fit Atomic spools with the lid closed unfortunately, and believe me, I tried. It is still too wide to fit and rubs against the internal guides on the lower side of the AMS. So, while you can make it work, it is still just a bit too much resistance for me to be comfortable with.
After testing several brands, I never ran into an issue of the shaft being too small for your standard 1kg spools. (Looking at you Sunlu)
As briefly mentioned before, the cardboard spools with printed rings worked so much better with the added clearance between the lid or the AMS. During testing I noticed some would actually rub the inside of the lid causing a lot of friction and resistance. This alone was the selling point for me to share this A3X concept in hopes of saving others from future frustrations and unanswered questions.

Give it a try! See how you like it and get back to me!

HAPPY PRINTING!!

Check out some of my other projects below!

Bambu Tough Case by Aesth3tix - MakerWorld

Swatch Box MAX by Aesth3tix - MakerWorld

Bambu Lab Coupler Adapter (Splitter Mod) by Aesth3tix - MakerWorld

The Original AMS PTFE Guide by Aesth3tix - MakerWorld

Comment & Rating (49)

@theurbanarmorer

Print Profile

TPU 95A_A3X Driven S Shaft r00_0.4mm Nozzle, 0.16mm Layer, NO BRIM

I printed these rollers with black Overture 95A TPU with no issues, and it was honestly the best print I've gotten out of the stuff. I assembled these with the instructed stainless steel dowels, though since I had some issues keeping space between the bearing and the TPU roller I actually added an M3 washer between the two and that did a brilliant job keeping the bearing from getting caught against the model. I have the Hydra mod on my AMS and have seen no problems so far. I've changed the back rollers on all 4 of my units, which is awesome. I'm keeping my fingers crossed that everything continues to go swimmingly, but overall I'm really happy with this model. The back rollers are a lot easier to remove with the thinner center, and the overall design feels pretty sturdy.

The designer has replied

@Aesth3tix

awesome! thanks for the well thought out input. myself and others will find it very useful.

@Ssinger804

Love this print and I cannot wait to let it roll 👍- thanks again

The designer has replied

@Aesth3tix

looks great! I’ve been using these exclusively and I’m more than happy with the results. I’m so glad many have had the same experience. happy printing!

@Unforgivenll

Print Profile

TPU 95A_A3X Driven S Shaft r00_0.4mm Nozzle, 0.16mm Layer, NO BRIM

Sno Labs TPU 85A. Thank you! Edit: I'm using these on all 4 of my AMS units without any issues for months now.

(Edited)

@adrianhajduk

Print Profile

TPU 95A_A3X Driven S Shaft r00_0.4mm Nozzle, 0.16mm Layer, NO BRIM

perfect print. Working perfectly with AMs

@AU3D

I understand your perspective on the issue, but I have a few clarifying questions, if you don't mind. You mentioned that spools struggle to maintain traction during rollback. Following that, you noted an issue with timing discrepancies. Could you explain how these two are related? Additionally, how does altering the roller assist in improving both grip and timing accuracy? If the roller in question is a support type, meaning it operates without motors, then its smaller diameter would primarily affect the revolution count as the spool rewinds. In this context, how does a passive roller impact the spool's movement? How much smaller is your mod compared to the original? What is the minimum spool size with this mod?

(Edited)

@et.hbrt

To have printed and tested the model, the diameter is pretty similar to original (around -0,5mm compared to original). The only thing that changes is that the roller bends during rotation and the spool is indeed lower, due to the weight bending the roller... Just made and printed solid PLA rollers with 2mm less on the diameter. Works perfect !

@RoadShow

Did you end up swapping out all four slots, or only the two that were causing problems? I'm thinking about giving this a try because I have noticed that flash forge ASA spools will often rewind too loosely.

The designer has replied

@Aesth3tix

i tried it on my 2 that were giving me the issues at first. focusing on the one with the most issues. since then, I’ve switched all of mine out with these. So far, I haven’t had one single issue of the spool slipping and causing the rewind timing to be out of sync. I’ve been using these for about 3 weeks now and some large multi color projects during that time. my suggestion is to try it in one of your slots first so you can see how it performs and to save some print time. then if you decide you want to ditch it, you won’t have wasted much filament. But I have a feeling, you’ll dig it.

@Aesth3tix

also, I forgot to mention, I did test a roll of flash forge pla with this roller and had no issues. assuming they don’t have multiple spools circulating, should do you just fine. if you do try it, send me some feedback if you don’t mind. i went through probably 50 variations or so until I landed on this one. the goal was to make it as universal as possible without losing function.

@jsleo58

These look like a good solution to the same problem I'm having. I have no tpu left and want to try the new tpu 95a hf also. Any suggestions on the profile since BBL hasn't put one out yet for the new TPU. Great design..thanks for your efforts!

The designer has replied

@Aesth3tix

i used the standard Bambu profile for tpu but then upped the volumetric speed to what it lists on the product page. worked out well that way. I’m sure it can be tuned a bit more but I haven’t gotten around to it yet and it’s still been working out just fine.

@jsleo58

Great..sounds like a plan

@TheNiels

Can I just take the bearings and metal pin from the existing bambu rollers? When googling for 693-2z bearings, it's only the bearing. but on your images it appears a small metal pin is also required. where do we get that? (hence question above)

The designer has replied

@Aesth3tix

it’s a 3mm X 12mm dowell pin. there is an Amazon link available on the models main page if you’re having trouble finding them. the bearing from the original shaft will work just fine. it’s the same specs as the bearing for the tpu shaft. link also available for the bearings.

@stephanethery19

Hello, we should make an explacation video because it’s a very good idea.

(Edited)

The designer has replied

@Aesth3tix

like an assembly video? i don’t care a bit to do that