akashrpatel/-mhz-ercf-filament-rack-buffer-tube-assy

~MHz ERCF Filament Rack/Buffer Tube Array

Per user request, I went ahead and created STLs from gcode generated from the project file. Hopefully these will work with little to no issue. Let me know, please. TIA

(see bottom section for details on a recent mod to add filament guide 'ramps' to ensure filament doesn't miss the buffer wheel when the buffer empties)

Being inspired by having drama with the ERCP Filament Buffer assemblies, coupled with watching a RMU video on Chris Riley's youtube channel, I decided to undertake the endeavor of fabricobbling a KISS buffer solution that didn't rely on multiple loops of filament. With that, I present the following to y'all for your consideration of how one gent went about tackling same. This solution provides ~1750mm of filament buffer for each filament channel (12 channel solution pictured here).

The 'buffer tubes' are Betaduct 75mm*25mm solid wall raceways. I got a set of 8ea 2m long pieces for roughly $120 from Newark. The PVC flavor is what I bought. The Noryl flavor is roughly 5x more expensive - be careful. The buffer tubes pictured above are cut to 36" in length.

• Engineering Drawing
• Step File from Cablecraft (OEM}

I cobbled the endcap assy's together in PrusaSlicer (/me != 3D CAD skills).

Each assembled Endcap needs 2ea PC4-M10, 1ea 608RS Bearing, 1ea M3 Heat insert, 4ea M3x10 screws (to attach the hanger halves to the lid) and 1ea M3x16 screw for the bearing axis - to tie the stack of parts together.

The bowden comb is an extra that I added to the model - ensures that one has more than enough if multiple end caps are printed. Also, on the ERCPs, using a piece of filament as the axis that the PC4-M10 coupler/holder slides on is a design choice that I find substandard. I ended up using jobber length 1.75mm drill bits - works well...

For the frame, I ordered a set of 10ea 1000mm lengths of 2020, in addition to a couple additional pieces I had stock on - used for the bottom anti-tip supports - printed feet/end caps (sharp edges will cut a toe or three off). The 4 up rights are 1000mm, the lateral pieces are 650-ish mm.

I used a ton of roll-in M5 t-nuts and 2 sets of these to tie it all together: https://smile.amazon.com/KOOTANS-Bracket-Connection-Aluminum-Profile/dp/B07RPT9X8X , along with a set of these for the top of the verticals: https://smile.amazon.com/Befenybay-Connector-Aluminum-Extrusion-2020s-Black/dp/B092YS1SX3.

The 4 members holding the buffer tubes are just the left overs from lobbing the lateral pieces down to 650mm, trimmed to length to allow ample spacing between the buffers - I may actually squeeze the buffer wings tighter and trim the metal down more - tbd - this was a finalize-the-design-while-building evolution.

At the top of each buffer tube, I used this type of bit to plow a hole through on each side (used an extra printed cap as a guide) and then dyked the top of the material away to allow the cap to be slid on/off without having to unscrew anything - makes servicing a breeze.

M5x8 with roll-ins round it out and hold the tubes to the wings. The bowden tube I used is here: https://smile.amazon.com/MUYI-Chemical-Electrical-Medical-Equipment/dp/B019PZ5MY4. It's 2mm ID but my orbiter v1.5 has plenty of pulling power, so no issues for me, with #4/#5 tophats driven by a NMEA 17 gear stepper (Y axis servo from my E3v2 in the printer graveyard).

Note: Expect to have to do some post processing - I am printing ABS fast with a 0.9mm nozzle & I needed to enlarge some of the holes with a deburring tool. Also might want to make the bowden comb's cutouts (the negative volume modifiers) a bit larger - i.e. increase diameter by 0.2mm. I had to go back and redrill all of them so the bowden tubes would slide in.

EDIT: I've had one instance of the filament coming off the wheel with the initial design - stayed on the sides of the wheel, but the bottom was not riding the wheel radius - I was messing with the filament spools at the time and I think I imputed a twist.?. However, if it happens again, I will amend the design with a lower hemispherical 'ramp' to prevent same. Time will tell...

EDIT: I've decided to create and add another .3mf Prusa Slicer project file to add filament guide ramps that aid in preventing the filament from coming off of the buffer wheel - hasn't happened again, but could and wanted a robust solution for y'all. This should address the one-time issue I observed. The only BOM change is the buffer bearing axle screw - a M3x16 BHCS is now required. The M3 heat insert now gets installed in one of the guide halves (the one without the screw head recess) and serves to secure the stack (filament guide half, buffer hanger half, buffer with bearing, buffer hanger half, filament guide half (with heat insert)). Pay attention to the orientation of the filament guides WRT the buffer tube cap, as one side is tapered by design (goes on the cap side of the buffer tube).

Filament Guides Slicer Images:

Assembly Images Cap, PC4-M10s, Wheel Hanger, Buffer Wheel, and Filament Guides:

Before/After Filament Guide Assy

Post assembly closeup

Completed assembly with guides fitted into the buffer tube (buffer still spins easy enough)

Each buffer tube assy will need, 1ea length of buffer tube cut from the betaduct lengths (36" is good), 1ea Cap, 2ea Hanger Halves, 1ea Buffer Wheel, and 1ea of Guide A and Guide B, along with the bearing, 2ea PC4-M10 couplers, and other hardware spoken to above (4+1 screws and 1 M3 heat insert)(.

Endcap Printing takes a while...