Fully Rack & Pinion Drive Wallace Video.

Some of you may have seen my Rack & Pinion Y drive before.

It took me some time, and lots of thought, but I've finally put together an X Rack & Pinion drive.

You can see it in action here:
Wallace R&P Video

The files for the X drive are on Thinkiverse here:
Wallace X R&P parts

Here are some print samples:




I'm planning on trying the PinkPanther girl next followed by the tornadow.

Any sugestions for other prints to test? I can go up 200mm.

I will be building more of these to sell over the next couple weeks.

---

Komb'
Glankonian Protoforms

Cool, isnt weight on the carriage the-lower-the-better, though? Dont think there is a way to do that neatly with a pinion though. Besides, this cant oscillate.

this is pure innovation! what speeds are you able to get? can you do a circle 4 inches in size single walled 1 inch up at 60-100mm/s? if so I'm sold on the idea.

What are the advantages of a eavl and pinion over the standard belts and pulleys?

---

greghoge.com

HUGE 3D PRINTER PARTS SALE!!!

This is beautiful. Here's my request:

1. Print a new set of racks and pinions.
2. Install them in your machine.
3. Goto step 1.

See how many iterations you can get.

Great job...

Whats the advantage with this?
I thought people didn't like this idea cuz it's 'too slow' ie: leadscrew
Also, with this design, you have the extra mass on the x axis?

Please do't beat me up over this, its just a curiosity question.

---

--------------------------------------------------------
Custom all metal CoreXY
- Duet 2 Wifi w/ PanelDue 7i
- 330mm x 360mm x 500mm
- 750w Silicon heater

Custom Mendel90
(Backup printer - Old reliable!) - Sold

Komb, that is awesome, you've got a big win right there.

The prints looks good.

How is the backlash and wear resistance?

Jasper1984 Wrote:
-------------------------------------------------------
> Cool, isnt weight on the carriage
> the-lower-the-better, though? Dont think there is
> a way to do that neatly with a pinion though.
> Besides, this cant oscillate.


I was told that the weight of the drive motor on the carriage would be a problem. But I had to try anyway.
I can put a smaller motor on the extruder, as the EckerTech extruder was designed for a slightly smaller motor.

Not sure what you mean vs oscillate. The X-carriage seems to have less up/down (twisting on the bars) motion than before.

jamesdanielv Wrote:
-------------------------------------------------------
> this is pure innovation! what speeds are you able
> to get? can you do a circle 4 inches in size
> single walled 1 inch up at 60-100mm/s? if so I'm
> sold on the idea.


I tried both the Slic3r default speeds 30mm/s etc and the same speeds I've been running my Prusa and Ecksbot at. 48 shells, 60 infill, 160 moves.

The video had the outer shell at 30mm/s as I had to do that to reduce ringing in the print when using the belt. I tried the octopus at 48mm/s outer shells and I can't see a difference from the previous one.

I'll try that circle alter and report back.

---

Komb'
Glankonian Protoforms

gmh39 Wrote:
-------------------------------------------------------
> What are the advantages of a eavl and pinion over
> the standard belts and pulleys?


I'm presuming you mean rack and pinion.

I originally replaced the belt under the Wallace bed as:

1. I didn't like the serpentine layout of the original design.

2. When it had gone slack it was a pain to re-tighten

So, I replaced it with the frist R&P design.

The top belt had a zip tie spacer, so the occasional tightening was easy.

After running the Y R&P for so long with no problems and people asking when I'll do the X, I did it.

Main advantage, no re-tightening.

Mogal Wrote:
-------------------------------------------------------
> Great job...
>
> Whats the advantage with this?
> I thought people didn't like this idea cuz it's
> 'too slow' ie: leadscrew
> Also, with this design, you have the extra mass on
> the x axis?
>
> Please do't beat me up over this, its just a
> curiosity question.

Less belt tightening.

Runs at same speeds as my belted machines.

True there is extra mass. Dosen't seem to be a problem though...

I'm nice, I wouldn't do that.

jzatopa Wrote:
-------------------------------------------------------
> Komb, that is awesome, you've got a big win right
> there.
>
> The prints looks good.
>
> How is the backlash and wear resistance?


Ah, Good question!

I put some lithum grease on the Y R&P when I noticed dust coming from my extruder gears.
Seen no wear since then.

If the gears are meshed properly there is no visable backlash. Heringbone gears help that.

However, if the gears are too tight, you get a ripple in the print. If the gear isn't fully meshed, slight gap, you get backlash.

Edited 1 time(s). Last edit at 11/01/2013 11:53PM by Komb'.

---

Komb'
Glankonian Protoforms

Wow, very nice! I suppose a bowden extruder would even out the weight gain from the X axis motor on the X carriage if you really wanted to counter the extra weight arguement.

Also, I wonder if it would be possible to keep the X motor static, and have the rack move with the X carriage? Just a reversal of the present arrangement.

---

Yvan

Singularity Machine

@ Yvan

Holding the motor out there and placing the Rack on the Carriage is doable. Would take a bit more to keep the motor steady though.

I'll give it some thought.

How ever, this set up is working just fine.

@ Everybody

The Pictures I promised:





I think I was running a little too hot on her.

In the side view you can just notice an artifact from the Y R&P. I have adjusted that and it should not be visable in future prints.




Coulden't figure out why it was deforming at the top until I noticed a cool draft blowing through the room.

Posted it anyway. Shows nobody is perfect.

Have moved printer back to its usual secluded corner of my work room.

Lighting is not so good in there, but the prints should behave better.

---

Komb'
Glankonian Protoforms

Quote
jamesdanielv
this is pure innovation! what speeds are you able to get? can you do a circle 4 inches in size single walled 1 inch up at 60-100mm/s? if so I'm sold on the idea.

Not being sure how fast the .4mm J-head could go, I decided on 75mm/s for this test.

After the Bottom fill, and the time it took to get my video capture software working... the print took 4 minutes. Probabily 6-7 including waiting for bottom fill and capture software start.

Pic first:


And Video:
YouTube

Enjoy,

---

Komb'
Glankonian Protoforms

Congratulations. Welcome to the wonderful world of reprap. [www.youtube.com]

---

Detlef

Excalibur Hotend

reprapclean32.png

reprapzone.blogspot.de

Quote
theodleif
Congratulations. Welcome to the wonderful world of reprap. [www.youtube.com]

Not sure where you are going with this...

I freely admit to borrowing form those projects. Says so on my first R&P design, referenced from my latest design.

However, I must say I really didn't like the bars coming out of the units like that. It seemed a dangerious design to me. Plus I don't have enough space for machines as they are.

That is why I went with a totally within the machine aproach. If I had just used that design for the X, I would have been done months ago. But I woulden't have been happy.

---

Komb'
Glankonian Protoforms

Komb i'm convinsed. give the printer a name and set it out into the wiki wild.

I've run into some problems printing 1 of the parts, now that I am making another whole printer.

Taking some time to make printing part easy.

Also replacing the Z end-stop holder with another design. Current one rocks too much.

I've been calling the entire Modified Wallace design the Wallce GP or Wallace G1

I'll be putting the entire Modified Wallace design on Thingiverse as soon as I have another frame assembled.

I'm also working on a compleate assembly guide. Frame assembly instructions partly done...

No Idea how to add it to the wiki.

---

Komb'
Glankonian Protoforms

here the wiki page is made. it just needs to be setup. as a reprap member you have permissions...



[reprap.org]


when i have setup wiki pages and have files to add, i have been posting in the experimental forum and labeling it as a file placeholder, then attaching files their,and linking to them in the wiki. wiki works fine for pictures.

@ jamesdanielv

Thanks, I'll get to that as soon as I can.

@ Anyone building an R&P Wallace.

As some had asked me to try printing a rack from my R&P printer, replace the existing one with the new, Rince & Repeat, etc.

I tried printing one on the Wallace G1 and the Ecksbot. Ecksbot made my originals.

The WG1's was about 2mm longer! So, it re-calibrated the E-steps for x-y and tried again.

This time the difference was fractional.

I had to place both against a hard stop, mark where they ended then measure the difference with my calipers.

.45

I have re-re-calibrated x-y E-steps and am printing again.

So, I realise if I did the reprint-replace rack test and didn't re-calibrate for any differences, I would quickly become way out of spec!
However, if careful measurments were made at every replace cycle, the differences should average out and the experiment would continue indefenetly.

So, You MUST make your own E-step measurements and calculations when tuning the machine!

---

Komb'
Glankonian Protoforms

Thanks for the update, Komb'. I'm really interested in characterizing the propagation of error through generations of replications, so I look forward to any more results you can post about this.

Do you think it is possible that, over many generations, even if you calibrate for absolute size, the surface roughness (waviness, etc) of the rack/pinion will degrade to the point where it will be unusable? People often mention "averaging out" when using printed pulleys. But in that case, the printed pulley is mating with a commercial off-the-shelf belt, which maintains the precision. Your machine offers a much better scenario to test this effect because, after the stepper motor, there are no more vitamin parts in the transmission chain.

When the Wallace G1 is printing out a duplicate rack, when it is printing the very end of the new rack, is its carriage at the limit of its own rack? Or is the new rack being printed offset at some diagonal angle so that it fits within the workspace of the printer?

The Rack is printed diagonaly across the bed.

So far the teeth of the Racks look identical.

---

Komb'
Glankonian Protoforms

would accumulative error be avoided by printing front and back halves of the linear bar rack, and then turning one print around to a designed herringbone style. The print coming from both sides reduces this error, so that 10 generations later the error has cancelled out?

then the only error apparent is the z height of the grooves, which should resolve itself as long as at least 3 teeth from the pinion are always in contact with the linear bar rack.

Edited 1 time(s). Last edit at 11/10/2013 04:45AM by jamesdanielv.

One way to quantify the error is to attach a vernier calliper to the axis of interest then compare set distance against measured distance. Some callipers even have the option to log the data to a PC automatically. Also even though you are re-calibrating the axes, there is still the possibility for error to drift between the start and end points of the axis.

Quote
Komb'
No Idea how to add it to the wiki.

You are not alone. I know jamesdanielv already started a page for you, but you (and many others I think) might find something useful in this small tutorial I just put up: Wiki Tutorial.