Loading

Improved X ends for Prusa with clamped rods

by jonaskuehling, published

Improved X ends for Prusa with clamped rods by jonaskuehling Mar 1, 2012

Description

I liked the idea of older Prusa x ends, that use two screws for each rod to fix it, but the implementation was rather weak. With iteration 2's solution (press fit, closed ends) it's a mess to disassemble the x axis, so I thought up a new robust version and since modifying one of the existing files would have taken even more time, I designed it from sratch and added even more upgrades like my clamped LM8UU holders!

Prusa's snap in LM8UU holders have one huge problem: motor- and idler-mount are supported via those holders, so the whole holder can twist unevenly and warp the bearings agains each other. This is avoided with my design.

The rods are fixed using one screw for each, tightening the built in clamps. Their screw heads are easily accessible from below and there are nut traps at the top. I made open end rod holes, to be able to push the rods all through. The system achieves a rock solid fit AND easy adjustability - what else you should need?

The leadscrew nut trap uses two bridges inside instead of a circular hole, so there's no "post-print-processing" (i.e. drilling) necessary in this point.

The design is entirely parametric, so you can even use professional leadscrews like trapezoidal ones with bigger nuts. I additionally attached STLs for use with metric trapezoidal leadscrews TR8x1.5 and nuts of 17mm wrench size (15mm height). For standard Prusa requirements use the "threaded rod" files.

Recent Comments

view all
Hi, what infill should I use so they are strong?
Hi.
I have the same problem.
How can be this problema fixed?
Thankyou in advance.
Hi i'm trying to slice this things and I have problems slicing the x-end-motor. Error message is:
"Warning,. The input file contains a hole near edge.
22.000000, 64.200000, 7.200000-0.000000, 64.200000, 7.200000 (not
manifold). You might want to repair it and retry, orto check the
resulting G-Code before printing anyway"
How can be this problema fixed?
Thankyou in advance.

More from 3D Printer Parts

view more

Liked By

view all

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag. Print Thing Tag

Instructions

There are thin support layers inside the M3 rod clamp holes that need to be drilled through after printing!

The STLs are generated for 0.25mm layer height - change the parameter in OpenSCAD for your needs.

Hi, what infill should I use so they are strong?
Hi i'm trying to slice this things and I have problems slicing the x-end-motor. Error message is:
"Warning,. The input file contains a hole near edge.
22.000000, 64.200000, 7.200000-0.000000, 64.200000, 7.200000 (not
manifold). You might want to repair it and retry, orto check the
resulting G-Code before printing anyway"
How can be this problema fixed?
Thankyou in advance.
Hi.
I have the same problem.
How can be this problema fixed?
Thankyou in advance.
How to use a trapezoidal TR8x1.5 threaded Z screw with 10mm Z-Rod with LM10UU. Is it difficult to change in Open SACD and how ?
Jonas, your design is (are) brillant.
I got someone to print one of these for me on a replicator 2 out of ABS, and one of the clamp bolt holes gave way as I was tightening it up. I just added a washer, and it's fine, but that part is perhaps slightly to weak.
Ohyeah, the motor part printed like a champ. Tomorrow the opposite side, let's hope it goes just as well :)
Hi, i cant slice using slic3r.... it just give me a very bad slice...

Any advice?
try running it through cloud.netfabb.com I had to do this to get it to slice, but I haven't printed it yet.
Hi, great job, but how i can print that, with or wihout support material? abs, pla? fan? regards!
Can be printed without support. The choice for ABS or PLA is up to you, both will work. The use of a fan will help with PLA while beeing counterproductive with ABS (warping).
Fantastic! Both X ends and Jonas's X carriage printed. Everything fits like a glove. Many thanks Jonas!
Can you explain how the leadscrew nut is retained?  Is it simply friction or is there a mechanical component.  Does the design require multiple leadscrew nuts or a single nut on each side.  Thanks for a nice design.
Jonas, Thanks!just printed those, any idea as to the distance between x-ends on a prusa i2? Thanks again!
Hi! This x ends works with mendelmax? Thanks in advance!
I'm still printing this, but these very thin bits in one part of the idler... I'd say they're a minor bug. They don't print cleanly and aren't structurally useful.
Would it be a huge job to allow for 12mm x and z smoot bars?

In these and yours x Carraige?

How would I go about it?

Thanks.
Just a short update of my notes concerning those spindle nuts:

They do still sell hexagonal TR8x1.5 nuts with 17mm wrench size as I wrote in the thing description. Only their square nuts have changed to 18mm, but you an still get the nuts I made the STLs for!
wht change layer height
Open the .scad file with OpenSCAD and modify the value for the variable called layer height.
Great parts Jonas. Where did you get the spindles? Was it in Europe? All I have found is quite expensive.
SHAMELESS PLUG ALERT!

techpaladin.com/store/#ecwid:category=1941327
&
amp;mode=product
&
amp;product=8445245

I can ship to Europe for under $10 in most cases
In fact, I bought them from you before I read this. Unfortunately nuts were out of stock... now they have come back but it is not worthy anymore to pay the shipment for one nut:( :(
absolutely incredible. THANKS!
Well, thank YOU ;-)
Is anyone using this with 32mm OD GT2.5 Pulleys? I don't think they work with this motor end...
thingiverse.com/thing:25047
GT2 36 Groove friendly x-end
by delusr
I found you have to sand down the edges of the pulleys a wee bit, or else do the same to the interior of the motor holder arm.
I tried to, but found the same thing.

I haven't had much luck in general with any x-ends that have the motor hanging off the back.

Using the idler end though and I like it a lot so far.
Do you only use one screw at the bottom with these or do you use a second one with springs on the top? Im concerned about that the springs will have problems with only the bridges inside the nut trap.
I use two spindle nuts for each x-end with springs and put an additional washer on top of the bridges before applying the springs. The necessity of a washer depends on the spring diameter - if the spring is only little wider than the spindles, a washer might be beneficial.
Well sounds logical. What washer size fits there? I think M8 is too big isnt it?
You're right, didn't think about that. I'm using the trapezoidal-spindle-version where my spindle nuts are 17mm in wrench size (while still 8mm spindle diameter), so common 8.4mm washers do fit in easily.

Maybe you should give it a try without any washer and choose springs that are up to 13mm in diameter (wrench size of M8 nuts) rather than like just 9mm..
Yeah, was what i thought. Hard to find some appropriate springs for that. Ill have a look.
Hi Jonas, Like your improvements but would it be dificult to adding an end stop like greg's then they would be perfect :-) Thanks.
Hi Typhoon, thanks for your compliment! So sorry, I'm quite buisy at the moment working on several projects that eat all my development resources :-) Btw. I heavily recommend switching to hall-o endstop for z homing and really comfortable endstop finetuning via potentiometer! No more complicated mechanical/optical endstop adjustment, all you need to do is glueing the little magnet underneath one of your xends and you're ready to go... See ikmaak.nl/eenblog/2011/11/09/reprap-improvements-i/
Hi Jonas. I had no idea about them. Very interesting indead. Thanks for getting back to me. Cheers. E
Looks great! I am getting the following errors, which is preventing a successful render in OpenSCAD.

WARNING: Ignoring unknown module 'roundcorner_tear'.
WARNING: Ignoring unknown module 'teardropcentering_half'.
WARNING: Ignoring unknown module 'roundcorner_tear'.
WARNING: Ignoring unknown module 'teardropcentering_half'.
WARNING: Ignoring unknown module 'roundcorner_tear'.

Am I missing some de
pendencies? I have all the .scad files you provided here in the same folder.
The mentioned modules are provided by my "jonaskuehling-default.scad" (last file in list on the left). Should work, if it's in the same folder.

You might double check that and please let me know if it's not working though.
Well, using an older version of OpenSCAD it compiles just fine. Not sure what the issue is. I'm looking forward to using these on my MendelMax!
So, after re-downloading the files, everything is good with the newest version of OpenSCAD. Must have had a corrupted file.
Jonas, could you please elaborate (or point me in the proper direction) for what you mean by "a properly calibrated machine". I'm very interested in these parts as well as the
&
lt;50mm bridge prospects. Thanks for these awesome upgrades.
I totally switched to slic3r since it was released as version 0.5.7 (it's at 0.7.0 at the moment) and achieved brilliant results. Calibration and adjustment is REALLY simple, compared with skeinforge etc.

If you'd like to give it a try, I explicitly recommend RichRap's trilogy "Slic3r is nicer" which you can find here: richrap.blogspot.com/2012/01/slic3r-is-nicer-part-1-settings-and.html

Follow his guide for your first steps with slic3r and I'm sure you will earn awesome results within days
..
If your printer works as intended and extrudes as intended, even long bridges come out easily. At least you will have a good starting point for further tweaking to even get the best quality possible out of your machine :-)
Awesome, thank you. I've read the slic3r is nicer article and now I'm bridging some nice gaps. Next is your print, thanks a bunch.
That's good news!
you've really done them with Slic3r? How did you get the clamps printed - wouldn't you need Support material for it?

I currently got a 0,5mm nozzle only and PLA but i'd like to give it a try. Just wondering if could really do that with slic3r cuz i didn't see any support structure for the top of the motor mount and the clamps. Would be a bit too far distance for bridging wouldn't it (still rather a newbie :))
With a properly calibrated machine bridging up to 50mm and more is possible. The upper motor mount support as well as the rod clamps only use bridging, so no support is needed at all.
Good -and much needed - job!

The only part that might be missing is moving the X axis motor center of mass back where it belongs, in line with the X axis instead of cantilevered.

The other background question: with a properly squared machine and well calibrated prints, one shouldn't be in need of so many springy parts. Any particular reasons?
I sticked to the x-motor position since overall stiffiness of the x ends was more important to me than balance of weight. The x motor itself doesn't perform fast movements despite of little z elevation and possibly retract-lifting via slic3r..

The previous position inline with the x axis gives less opportunities to support the mount beefy enough, I found. There ain't too many possibilities for the motor mount at all at least if one would like to stick to the prusa belt lineup, tried out various on my way :)

(Might be time for vertical x axis....)

I didn't get the point you asked with springy parts, could you please specify your question? Thanks in advance ;-)
About the motor position: having heavy parts with the center of mass needlessly suspended away from the point where they are connected to the structure is calling for new oscillation modes, just like adding another pendulum. And the motor itself is the main source of excitatory vibration for the system.

There is room to properly secure that motor in line with the x axis inside a (ventilated) box-like structure, it only adds some mms to the X belt length.

About "springy" design: the section connecting the z bearing holder to the rest of the structure appears to be quite narrow, so it will flex and twist rather easily. This may be wanted, or not.

Again: since we are currently limited in speed by the extruder, this may well not matter.
Sure you're right. I actually notice visible effects of oscillation above printspeeds like 85mm/s for perimeters and 130mm/s for infill with Marlin firmware. But at this speed, the whole Prusa frame wobbles so seriously, that I think it doesn't matter anyway, if the x-motor joines in :-) In my eyes the frame design needs a huge redesign first - and with this said I would'nt stick to a horizontal x-axis anyway.. So I didn't want to spend too much work on this..

I finally got your flexing concerns. The reason for the pretty thin (3mm) support wall stabilizing the lm8uu clamp is, that I liked to let the clamp as much freedom as possible to evenly fix the bearings without twisting any part of the rest of the x end. At the bottom, the clamp is connected on a
wider range to the main x end body, the connection width is about 10mm for the height of the rod clamp body (ca. 15mm). The "thin" support wall up to the upper motor mount support bridge delivers additional stability, but doesn't take the whole force.

I'm sure there might be even better ways to bu
ild this, but the performance of the parts is really satisfiing for me so far..
I really, really like these.
Thank you i really hate zippy ties

you must be really patient ,,ie wot kind of print time are you getting.

and how much fill ?
Print time should be comparable with standard Prusa v2 x ends.. I use 25% rectangular infill (slic3r) for ABS and I have pretty rigid results. The rod clamp design doesn't depend much on a special infill level and all parts that should come out strong but flexing are rather thin, so most infill values lead to nearly solid infill in these parts anyway (rod clamp, lm8uu clamp). 25% is good and strong, you shouldn't need much more than this.

Oh and I use a 0.35mm nozzle, 0.25mm layer height, 3 solid layers and 3 perimeters.
These look really good. I think I may have to make a set of these for my Prusa :) Nice design
Top