josefprusa / prusamendel Goto Github PK
View Code? Open in Web Editor NEWMy own RepRap Mendel redesign
Home Page: prusa-mendel
My own RepRap Mendel redesign
Home Page: prusa-mendel
Hello, and thanks for all the great work so far.
This is possibly inconsequential or done for extrusion reasons, but one side of the bar clamp is 0.5mm thicker than the other.
The problem seems to be in this line:
translate([9, outer_diameter/2+1, 0])cube([18,05,20]);
It should probably be something like:
translate([9, outer_diameter-slot_width/2, 0])cube([18,05,20]);
I have carefully measured this in Netfabb, and noticed it when I printed out the felt branch. Distance of motor centre (or Z-leadscrew centre) to Z-rod centre on Z-motor-mount = 30mm, same distance on X-end-motor and X-end-idler = 29mm. Which means the Z-axis motor coupler will ALWAYS run eccentrically.
And this is made even worse as NEMA17 motors are 42.3mm square on the end http://www.zappautomation.co.uk/pdf/steppermotors/SY42STH47-1684A.pdf , so 21.15mm from centre to edge, while the distance from the centre of the motor shaft to the 'wall' between the motor and the Z-rod is 20.8mm, so motor is pushed EVEN FURTHER away from the Z-rod. I guess this is why triffid in Issue 20 https://github.com/prusajr/PrusaMendel/pull/20 moved the motor centre, but in doing so causes the X-leadscrew to run EVEN MORE eccentrically as the motor centre moves to 31mm from Z-rod centre.
I suggest removing 1mm of the wall from between the motor and Z-rod, to allow NEMA17 motors to fit and maintain the 30mm spacing. The X-end-motor and X-end-idler need to be amended to have the 30mm spacing between Z-leadscrew centre and Z-rod centre, which is currently 29mm. I think this may have been an issue for a long time!
From what I understand we are documenting this primarily on the reprap wiki (http://reprap.org/wiki/Prusa_Mendel).. even if this was not intially the plan the documentation there is now much further along then the README file on github. To avoid confusion could you please simply provide a link to the reprap wiki documentation in that file so that people are not trying to build Prusas with out of date or less complete documentation then they have to.
PS. Prusa Mendel really is an excellent improvement over Mendel, cheers for making things better!
http://www.thingiverse.com/thing:11838
Does anyone have objections to this being merged with the project? After seeing Theodleif's design - it feels like the current design has some unnecessary parts.
I think it's time to merge the Greg's Wade Reloaded into the Prusa to fix a few issues.
http://www.thingiverse.com/thing:18379
Prusa mentioned a few issues with the current STL's for the greg's extruder parts 2 months ago: https://github.com/prusajr/PrusaMendel/blob/master/CologneCourseNotes.txt
Merging the reloaded version should fix parts of both of these issues:
https://github.com/prusajr/PrusaMendel/issues/34
https://github.com/prusajr/PrusaMendel/issues/39
I find the small gear in wadesbits.stl prints more successfully when it's round-side down. Is there any reason why it's not like this in wadesbits.stl?: http://www.thingiverse.com/thing:2521
The X and Z motor brackets have slots like the original Mendel. Presumably this is to allow the use of NEMA14s, although why it is not just two sets of holes I don't know.
The Y bracket only has holes for NEMA17 which is inconsistent.
The hole for the filament no longer seems to line up with the centre of the idler. I.e. I think it has changed position from the original. This will mean bought in hobbed bolts will not be compatible.
When compiling you get "WARNING: Ignoring unknown variable 'layer_height'."
This is used in module motor_mount_holes(), without it the holes don't protrude threw the object
The Bed drilling template for lm8uu has the wrong hole distances. It should have 28mm like the template for lme8uu has.
The original ones go all the way through both sides. Is there a reason why they don't on this one?
Some people are having print quality issues due to the belts' teeth rubbing on the bearing guide's flange. I thought you might be interested in trying a crowned pulley like the one I just tested. I think it would simplify the build somewhat as well, as it's only one printed part per pulley.
I have a more complete description on Thingiverse:
http://www.thingiverse.com/thing:14117
http://www.youtube.com/watch?feature=player_profilepage&v=foQmpN2twf8
DaleDunn@f104c98
http://www.thingiverse.com/thing:13482
After every print, the position of my Z-endstop seemed to change. The thing in the link above solved the issue. It should definitely be merged.
I see a lot of people (including myself) having trouble with the Z Coupling in this repository cracking due to stress.
Maybe we should consider a design like Nopheads?: http://www.thingiverse.com/thing:9622
It looks like Nophead's design is the one PrusaJr prefers: http://www.flickr.com/photos/prusajr/6045851714/in/photostream/lightbox/
The M4 hole size is 4.5, which is reasonable but the M3 is 4.4, i.e. almost the same. Looks like a typo as 3.4 would be more reasonable.
I don't think the STL files belong in the repository, it's sort of like putting a binary file like a .exe in a repository because they are compiled from code.
I would suggest releasing them every so often in packages, along with the compiled documentation (pdf, html, wiki. These should not be kept in the repo either, IMHO), just as releases are done with software.
What do you think?
While I can understand the desire to minimize the use of vitamins, it seems to me that the current friction-fit between the idler/motor brackets and the X axis' smooth rods is a poor trade-off. The brackets I were sold in my initial parts buy had nut traps and bolt holes to retain the X axis smooth rods. When working with poorly aligned machines (e.g. poorly sized Z axis couplers), I have found it invaluable to be able to lengthen the X axis smooth rods a tad, allowing me to apply force on the Z axis bushings and minimize X axis wobble. This is not possible with a mere friction fit. Moreover, I've noticed that even without this pre-loading, the friction fit brackets permit noticeably greater wobble than the old parts due to flexing of the pressure-applying finger.
z-bar-top-clamp_4off.stl
pulley.stl
y-motor-bracket_1off.stl
http://www.roymech.co.uk/Useful_Tables/Screws/Hex_Screws.htm and everything else I've checked says that M8 nuts are 13mm face-to-face, but metric.scad has them listed as 14mm. this causes the nut traps in the X-ends to be too large, giving backlash on Z.
why do you subtract 0.4 from diameter?
It never fits a bar
bar-clamp.scad: "threaded_rod_diameter/2-0.4"
When printed for LM8UU bearings, the center hole in the X-Carriage removes structure from one of the LM8UU holders. The center hole should be subtracted from the carriage plate, but not from the LM8UU holders.
The nut size in wade-big.scad is still presumes edge to edge measurements for m8_nut_diameter
ie line 53 "cylinder($fn=6,r=m8_nut_diameter/2-0.5,h=nut_trap_depth+1);"
Should read something like "nut(m8_nut_diameter,nut_trap_depth+1,true);"
I use Slic3r and have the latest version. I delivers the following error when adding metric-prusa-lm8uu/x-end-idler.stl by it's self:
Warning: the input file contains manifoldness errors. Slic3r repaired it successfully by guessing what the correct shape should be, but you might still want to inspect the G-code before printing.
It then delivers this error after clicking Export G-code:
Warning: The input file contains a hole near edge 30.000000,2.000000,0.000000-30.000000,2.000000,0.000000 (not manifold). You might want to repair it and retry, or to check the resulting G-code before printing anyway.
The resulting gcode ends before bridging the top of the rod holes.
If I update the config for LM8UU and build the x-end-idler in openscad, it slices fine.
The BOM in README does not list the y-bushing mounts.
As discussed in Github #4285 the MMU2S suffered idler homing problems.
The SW people were looking for solutions while changing sensitivity values.
I found that the motorshaft and Idlerdrum cannot be perfectly aligned. Also that the motor mount was not perfectly square to the Idler housing centerline.
My route as a mechanical engineer was to eliminate the over-determination .
Based on tolerances on the Idlerdrum connection to the motorshaft and the mounting plenum of the motor self, I changed the motor mount into one providing 2 extra rotation freedoms.
My machine is now working fine without any changes in sensitivity values with the latest firmware.
Read my doc please and try my .stl's on printables under MMU2S+ ๐คญ
I reach out to Prusa engineers to understand what I did and why.
For discussion available,
Henk van Ooijen
Hi, just been looking through your files and noticed that the LM8UU & LME8UU pre-generated stl's are the same! Problem is solved by changing false-true in OpenScad, rendering and exporting, though thought I'd point this out for others so they don't print straight from stl's!
@prusajr: Could you please tag the version you printed for the reprap class in Aachen? Thanks!
At the point where the ziptie cutout is made, the thickness of the walls drop to a point where it becomes very easy to break the part while removing it from the build plate.
I would suggest increasing the inner diameter of the ziptie cutout by 1mm.
This could be basically development note (to myself) but I could not find any better place where where to document the ideas behind my changes.
My idea is to make the Z rod channels on the X ends a little bit smaller. The goal is to add stability and make it easier to print since I noticed if printing one by one, the ends seem to get too hot easily depending on the material you use to print them.
With linear bearings, we can make the Z rod channel about the same height as the support beam on the motor X-end.
To achieve similar on the idler X end, we can make the triangular shape ranging from the start of the support beam to the Z rod channel.
I'm very sure this will work without problems using linear bearings. I'll try printed lm8uus and igus bushings with that before integrating into master.
This causes an issue with the extruder body, which uses M3 bolts. The nuts at the top of the extruder body are impossible to fit into the nut traps without considerable reworking of the nut traps.
The value given in metric.scad is 5.3mm.
By my measure, it should be 6.28mm.
The drive pulley M3 grub screw hole consists of a cylinder and cube. Where these transition there is a step. The following code gets rid of the step:
//entrance
translate([0,-3,15]) cube([5.8,7,3],center=true);
//nut
translate([0,0,13.6]) rotate([0,0,30])cylinder(r=5.8 / sqrt(3),h=2.8,$fn=6);
//grub hole
translate([0,0,9]) cylinder(r=3.3/2,h=10);
I've now built most of my Prusa Mendel but I've now run into 2 problems:
I have yet to find the y carriage in github that is shown in this vid: http://www.youtube.com/watch?v=0MvUD-tuOX0
The Z shaft couplers don't work (as Prusajr mentioned earlier)
The first problem is the important one. Could you please post the y carriage or let me know where to find it?
Thanks,
Rob
Not able to download the Zip File the status is shown as Network Error
vars=[
//[m8_dia, m8_nut, m4_dia, m4_nut, m3_dia, m3_nut, bush_dia, mot_shaft, bush_rod, bush_outerdia, bush_length]
[8.4,2,5,9,4.4,7,11,5.3,8,16,11],//metric
[9,16.4,5.5,11,5.5,10,11.5,5.3,7.9375,16,11]//SAE
];
Alterations to the 2nd number in the SAE space (16.4 here) does not change the size of the nut hole. Also if you look if both of these setting work the same wouldn't the nut diameter of the metric 8mm be 2mm?
Structural rods (with dimensions) , motors, gears etc. should be listed in BOM with adequate detail.
The two arches meet a knife edge.
Error when trying to pull it in:
$ git submodule update --recursive --init
Submodule 'thingdoc' ([email protected]:prusajr/ThingDoc.git) registered for path 'thingdoc'
Cloning into 'thingdoc'...
ERROR: Repository not found.
fatal: The remote end hung up unexpectedly
Clone of '[email protected]:prusajr/ThingDoc.git' into submodule path 'thingdoc' failed
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.