Welcome! Log In Create A New Profile

Advanced

Design questions.....

Posted by aamcle 
Design questions.....
October 02, 2018 06:56AM
I'm not entirely sure that this is the right forum to ask about this but here at least there are members who may actually know the answer.

I'm upgrading an Old CTC Replicator Dual it's the Makerbot Clone in a plywood box and I've come across something that puzzles me.

At the top rear of the printer that is a rod with toothed wheels at each end, it transfers the rotation of the Y axis stepper to the two Y axis belts.

At the front top there is another rod with toothed wheels fixed to it (the wheels do not rotate relative to the rod or each other) and the Y axis belts go around the wheels.

Why is this better than freely rotating idler wheels?

The timing is fixed by the rod at the back and the fixed wheels at the front just make it harder to refit the belts.


Many Thanks. aamcle
Re: Design questions.....
October 02, 2018 10:50AM
I've got a plywood box printer, too, with a similar design.

With both axles being rigid like that, it takes some of the load off of the rear axle. The rear axle is driven by a closed loop belt, and drives both Y belts, but with the front axle being driven by the Y belts as well, reinforces keeping them in sync, reducing some of the torque load on the rear axle.

Does it actually make a difference? Probably not, and a free-wheeling axle in the front would probably be just fine. I'd bet that the design works that way because it's the simplest solution parts-wise, with the least chance that the builder can screw it up when assembling. Also, the idler pulleys with a set screw will be the same as the ones used on the back, as well as the ones on the motors, meaning that it's just more of the exact same part. They also become the set collars which keep the axle from moving laterally, instead of needing a separate part which would be needed with a free-wheeling bearing based pulley.

For what it's worth, the way that I change the belts is to loosen the grub screw on one side of the front only, then tighten the belts and clamps on both sides. on the side with the loosened grub screws, the pulley will turn over the axle. On the other one, it will turn the axle a little (I do this one first.) once done, tighten the grub screws.


MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts
Re: Design questions.....
October 02, 2018 11:21AM
Thanks, it occurred that it might be a manufacture issue but it looked more expensive than just using idlers.

That printer has very much put me off closed box designs, a job that would have take me 45mins on an I3 - lift it out of an enclosure fit the bits put it back, took me about 4 hr fighting it out with all those M3 nuts and bolts that hold the case together sad smiley .


ATB. Aamcle

Edited 2 time(s). Last edit at 10/02/2018 11:23AM by aamcle.
Re: Design questions.....
October 02, 2018 02:21PM
Pulleys without bearings are cheaper than pulleys with bearings. Mounting two pulleys on a single shaft means you can use two bearings on the shaft instead of 4 on the pulleys.

When wondering why anything is done in a commercial printer, your first thought should be cost reduction. Everything done is intended to make it cheaper, not better.


Son of MegaMax 3D printer: [www.instructables.com]
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Design questions.....
October 02, 2018 03:34PM
If it can be "cheaper and better" that's the best case, but "cheaper and good enough" will generally win. This is why you see things like zip ties being used to hold stuff together. (I use some of those too on my lead screw printer.) They'\re good enough that they either wont' be a limiting factor, or else the limits that they impose will fall within acceptable for the design goals. in the case of 3D printers, nylon zip ties (As an example) may allow the bearings to move, which is no good if you're looking for sub-micron levels of accuracy, but they can be tightened enough to keep the tolerance down well below normal 3D printing tolerances. (~81 micro steps/mm is a tolerance +/- about .01mm inherent in a typical belt-driven axis anyway, plus the tolerance of the nozzle itself, so where's the sense in keeping things super rigid anyway? Add in changes like thermal expansion of materials, and just vibration in materials, and most 3D printers just don't need super accurate machined bearing holders.

I hear where you're coming from on the closed box design, but when I compare the closed box makerbot clone design to my I3 (Of which, as of now the only piece left in it's "redesign" is the extruder motor) the closed box is so far superior that it's not even funny. It's far better at keeping the orthagonal axes at right angles to each other, needs far less maintenance, though I'll admit that when it does, it is more time consuming to do. The dual belts for the Y axis were a real pain to adjust to ensure an orthagonal relationship between X and Y for me. I ended up building a jig to set the X carriage perpendicular to the Y rails, so it's within 0.25mm over the span of the 250mm of printable X travel. The margin of error is not zero, but it's good enough for anything that I'm going to print. once set, with the design the way that it is, it's less likely to drift over time without very noticable problems like loose belts, whereas my I3 would drive all the time (I had to re-tighten the frame nuts about twice per week until I finally gave up and used loc-tite on them, and then every time I moved the printer, I still had to adjust things.)

Anyway, with the pulleys the parts are:

1x 5mm rod, 2x 625F bearings, 2x pulleys, 1x spacer. A high quality kit may include 2X thrust washers

With idlers:

1x 5mm rod, 2x 625F bearings, 2x bearing idler pulleys, 2x thrust washers, 2x locking collars, 1x spacer. Higher quality might include 4x thrust washers.

This represents a cost increase for zero quality increase. it does increase the difficulty of assembly/tuning, but actually reduces the finished product's abilities, and increases points of potential failure.


MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts
Re: Design questions.....
October 03, 2018 04:30AM
Thanks All
Sorry, only registered users may post in this forum.

Click here to login