McWire Successor

McWire Successor After coming across a number of people and a number of posts feeling frustrated with the McWire, I felt like there should be a single thread to discuss the problem, and how to deal with it. In addition to this thread, I've created a wiki page: [url=http://objects.reprap.org/wiki/Development:McWire_Successor]Development: McWire Successor[/url] Any and all discussion about a replacement robot are welcome, even if it restates stuff from other forums and threads. I have a couple of things to get us started: -The robot should be made from materials that can be purchased locally. -The robot should be able to be made with a minimum number of tools. What are all the problems that the McWire suffers from? What are the advantages it has? (Edit: I renamed this thread, and then reverted the name change when I realized it was silly. -Sebastien) https://reprap.org/forum/read.php?1,37355,37355#msg-37355 Tue, 16 Jun 2026 20:09:52 -0400 Phorum 5.2.23 https://reprap.org/forum/read.php?1,37355,50010#msg-50010 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,50010#msg-50010 Tiberius General Fri, 25 Jun 2010 07:39:12 -0400 https://reprap.org/forum/read.php?1,37355,49980#msg-49980 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,49980#msg-49980
Thanks,
Markus]]> Traumflug General Fri, 25 Jun 2010 05:09:09 -0400 https://reprap.org/forum/read.php?1,37355,49456#msg-49456 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,49456#msg-49456 Tiberius General Sun, 20 Jun 2010 05:52:39 -0400 https://reprap.org/forum/read.php?1,37355,49435#msg-49435 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,49435#msg-49435
BTW., I'll use threaded bars intentionally, as I expect to keep the machine with a Dremel in the toolstock. Some 600 mm/minute isn't that bad for milling.]]> Traumflug General Sat, 19 Jun 2010 20:30:01 -0400 https://reprap.org/forum/read.php?1,37355,49429#msg-49429 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,49429#msg-49429
I used Wolfgang's instructions for building the frame and the axis. This was incredibly easy. Initially I wanted to use M8 threaded bar for the drives because I wanted to use as much re-usable stuff as possible but Wolfgang convinced me during a long phone call that threaded bar is painfully slow. So I got the Mendel belt kit, a handful of 608 bearings (both will be need in the Mendel anyways) and found some T5x16 pulleys at the local electronics store (Conrad).

After splitting the Mendel Z-Belt I used the 5mm portion as the Y-drive and the rest as X-drive because they're a good fit length-wise for the WolfStrap's dimensions. The Z-drive remains a M8 threaded bar.
I used large metal corner braces, M8 bolts, 608 bearings and large washers to build idlers for the belts. The same corner braces were a perfect fit as stepper motor mounts (after 3 minutes with a Dremel cutting disc).

Overall the build was not very complicated and dead-cheap. Here is a rough breakdown of the costs:

20€ - all extruder hot end parts at grrf.de. lots of NiCr and Kapton left
30€ - RP parts for Wade's geared extruder bought from Wolfgang
20€ - Mendel belt kit from grrf.de
60€ - 3 NEMA17 steppers for the axis from grrf.de (they're expensive!!)
10€ - 1 NEMA17 stepper for the extruder from ebay
16€ - 2 T5x16 pulleys at Conrad electronics (incredibly expensive!!)
6€ - 10 608 bearings from ebay (used only 7 of them)
about 70€ for all the wood, slides, bolts, metal corner braces, threaded bar and screws with lot's of leftovers
------
round about 250€ total for the whole build of the bot and extruder. Much of the stuff I bought could've been cheaper but I was too lazy to search.
Add your choice of electronics and you're good to go. I went with techzonecom.com (which are still not here after 3 weeks :X ).

I believe this is the most cost-effective method to get your Mendel. Everyone who can mount a bookshelf and build an Ikea bed should be able to do this. For the construction you only need normal household tools except for the wood pieces and extruder drive bolt.
If you don't have your own circular saw and drill-press for the wood I think you can assume that there's a carpenter somewhere near you who can do it for you.
The filament drive bolt for Wade's extruder is easy to make if you have a drill-press with XY-table, otherwhise you can switch to Adrian's geared extruder.

I have taken lot's of pics during all stages of the build. I started to edit the German WolfStrap wiki page to include my experiences. I'll update the english one as well if I find the time.]]> Tiberius General Sat, 19 Jun 2010 17:40:07 -0400 https://reprap.org/forum/read.php?1,37355,41428#msg-41428 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41428#msg-41428 -------------------------------------------------------
> I've found that the opto-sensors in use for the
> Mendel can see through blank paper, but can't see
> through paper I have printed black with an inkjet
> printed. This makes it easy to make custom
> encoders.

For linear encoding you could theoretically use of the online barcode generators.]]> annodomini2 General Fri, 02 Apr 2010 16:41:55 -0400 https://reprap.org/forum/read.php?1,37355,41329#msg-41329 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41329#msg-41329 mccoyn General Thu, 01 Apr 2010 16:35:57 -0400 https://reprap.org/forum/read.php?1,37355,41300#msg-41300 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41300#msg-41300 -------------------------------------------------------
> I see what you're saying, although in the specific
> robot I've linked to's case: the hobby servos I've
> played with seem to have decent holding torque
> simply as a result of the high gear ratio.
>
> I can understand why there is a potential for
> cumulative error, however since the servo's
> themselves have potentiometer closed loop
> feedback, I'm hoping that the PID controller's
> acceptable error will be accurate enough for the
> intended purpose, and will manifest itself as
> noise rather than the cumulative error as you
> would see with missed steps on a stepper.
>
> Because of the incredible potential of this
> avenue,($70 3 axis platform with high print
> speeds) I'd like to measure the inaccuracies and
> see if I can pull them out of the robot six sigma
> style.
>
[objects.reprap.org]

[forums.reprap.org]

> Does anyone know of someone who has successfully
> written inverse kinematics code to control a delta
> bot/tripod in EMC2? (Other than FESTO I mean.)

[forums.trossenrobotics.com]]]> nite69 General Thu, 01 Apr 2010 13:31:04 -0400 https://reprap.org/forum/read.php?1,37355,41290#msg-41290 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41290#msg-41290
5 ft 7959K24
4 x 57105K11

Total was $38.70...pulleys were expeeensive. I have to find a way to make them.

This discussion over a cable drive system seems very promising cost-wise.]]> Shayfir General Thu, 01 Apr 2010 12:31:51 -0400 https://reprap.org/forum/read.php?1,37355,41289#msg-41289 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41289#msg-41289
-------------------------------
I've been slow to use the cable drive I've been working on because the 1/32" cable I ordered runs up over the other windings or runs over to the edge and then winds over the previous windings. The cable is stiff but does work. When you're winding it don't let go! Rats nest! Looks good and works well while I'm twisting it on and off by hand.

My analysis of the problem/s is the capstan is not perpendicular to the platforms travel and not level.

I've designed a rather complicated method to be able to adjust EVERYTHING!

Using some 1 1/2" pipe with holes drilled and tapped to use bolts at three equal spaces around some bearings I think I should be able to tweak it till it's just right. There are several diy cnc builds that do this.

From reprap

Now gotta build one to see.
[picasaweb.google.com]
The sketchup file is here:

[dl.dropbox.com]
]]> Arvin General Thu, 01 Apr 2010 12:19:17 -0400 https://reprap.org/forum/read.php?1,37355,41224#msg-41224 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41224#msg-41224 Shayfir General Wed, 31 Mar 2010 22:35:45 -0400 https://reprap.org/forum/read.php?1,37355,41108#msg-41108 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41108#msg-41108
The encoder outputs are usually preconditioned in the encoder, so they come out as a digital signal ready to go into an arduino.

There is usually a timing belt and DC motor in the printer, too. Get those for future use. Also, watch out for stray ink! People often get rid of their printers because they have become messy.

fdavies]]> fdavies General Wed, 31 Mar 2010 06:33:51 -0400 https://reprap.org/forum/read.php?1,37355,41075#msg-41075 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41075#msg-41075 rocket_scientist General Tue, 30 Mar 2010 20:21:08 -0400 https://reprap.org/forum/read.php?1,37355,41073#msg-41073 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41073#msg-41073
fdavies]]> fdavies General Tue, 30 Mar 2010 19:44:21 -0400 https://reprap.org/forum/read.php?1,37355,41020#msg-41020 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,41020#msg-41020
A second way to do this have a guide bar with two well spaced holes. The cables go through the holes, and the guide bar is driven left and right by a thread drive running of the drive rod. This way, the spacing is more assured, since it is still possible to jump threads at the extreme ends without the moving guide holes. I will see if I can draft a drawing.

drawing.pdf

EDIT:

After looking at my drawing (sorry it is stuck in pdf), I can see that a much simpler solution is to combine both methods. Use a threaded rod to hold the cable, but have a "c" shaped piece of metal that is threaded to slid up and down the rod as it rotates, and has two openings for the cables to go through so that it winds even on and off the threaded rod. No need for a gear drive. In fact, the next step is to put the motor on a slide, and mount the "c" shaped block to the platform, so that the openings for the cables to come in and out does not move, but the motor moves left and right as it winds first one way then the other. 4 roller bearings at 45 degrees in angled slots would allow it to slide easily without the motor moving or flopping around.

Mike]]> rocket_scientist General Tue, 30 Mar 2010 12:27:27 -0400 https://reprap.org/forum/read.php?1,37355,40910#msg-40910 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40910#msg-40910 -------------------------------------------------------

>
> I bought some 50lb test braded fishing line. No
> luck. It started out great but seemed to flatten
> and then rolled up on the other layers. Still
> worked but the steps would not be the same length
> for those versus the lower layer.
>
> My next thought is kite string. I'm worried it
> will do the same thing. Mayb I just need to find a
> source for really thin steel cable. Did some
> searching earlier with no luck. I'll try again.
> Arvin

I gave it some thought a while ago while designing a cable drive. If the cable has multiple wraps around the capstan, and the capstan is winding on/off a screw of suitable pitch, advancing/withdrawing, it should allow the cable to wind on/off in an involute manner.

[forums.reprap.org]]]> murrayd General Mon, 29 Mar 2010 20:16:38 -0400 https://reprap.org/forum/read.php?1,37355,40902#msg-40902 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40902#msg-40902 Shayfir General Mon, 29 Mar 2010 19:15:03 -0400 https://reprap.org/forum/read.php?1,37355,40718#msg-40718 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40718#msg-40718
I can understand why there is a potential for cumulative error, however since the servo's themselves have potentiometer closed loop feedback, I'm hoping that the PID controller's acceptable error will be accurate enough for the intended purpose, and will manifest itself as noise rather than the cumulative error as you would see with missed steps on a stepper.

Because of the incredible potential of this avenue,($70 3 axis platform with high print speeds) I'd like to measure the inaccuracies and see if I can pull them out of the robot six sigma style.

Does anyone know of someone who has successfully written inverse kinematics code to control a delta bot/tripod in EMC2? (Other than FESTO I mean.)]]> JohnnyCooper General Sat, 27 Mar 2010 16:00:50 -0400 https://reprap.org/forum/read.php?1,37355,40698#msg-40698 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40698#msg-40698
For Example, your typical Mendel peice has around 10-100 thousand lines of Gcode, each with a very specific number of 2 degree steps on all 3 axises. If in those thousands of lines of gcode you loose or overshoot even 1 step, the whole part will be visibly flawed. 3-4 and the part will likely be mechanically useless.

The Delta Robots suffer from the fact the motor sits at the fulcrum of each of the axises, requiring each axis to use it's holding torque to simply keep the head in place against gravity & the motion of the other axes. In a screw or belt drive the the motors only need hold against a much less aggressive force from the 2 other axes, and almost no holding against gravity.

Delta robots are a great research project, but I don't really see any advantages for this apliticaltion, besides the fact that they look freekin awesome.]]> spacexula General Sat, 27 Mar 2010 10:31:07 -0400 https://reprap.org/forum/read.php?1,37355,40668#msg-40668 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40668#msg-40668 [letsmakerobots.com]

or the Festo iFab if you are looking for a bigger work envolope and prefer steppers:
[www.youtube.com]]]> JohnnyCooper General Sat, 27 Mar 2010 03:48:18 -0400 https://reprap.org/forum/read.php?1,37355,40657#msg-40657 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40657#msg-40657
Blog Team Open Air

Mike]]> rocket_scientist General Fri, 26 Mar 2010 20:26:42 -0400 https://reprap.org/forum/read.php?1,37355,40653#msg-40653 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40653#msg-40653
http://shayfir.blogspot.com/

PS, I just want to say...great job everyone. The Reprap community is the greatest I've ever seen.]]> Shayfir General Fri, 26 Mar 2010 19:35:07 -0400 https://reprap.org/forum/read.php?1,37355,40255#msg-40255 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40255#msg-40255
Post a picture of what you're doing. I found some 1/32" steel cable at McMaster-Carr that might work. Minimum length to buy is 50 feet at $0.07/foot and 185lb strength. I will buy some when I can remember what the other thing was I needed!]]> Arvin General Mon, 22 Mar 2010 21:12:00 -0400 https://reprap.org/forum/read.php?1,37355,40250#msg-40250 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40250#msg-40250 I may be thinking of larger diameter pulleys than you are. For my drive pulleys, I am planning on using the 1 11/16" plastic sliding glass door wheels plastic pullies. I plan to put a plate on the other side of the pulley with a set screw for the stepper motor shaft, then bolt the plate to the pulley to drive it. I think that the 1/16" steel cable should work around that. If not, then I will have to re-think that. I should have this going in the next few days.

Mike]]> rocket_scientist General Mon, 22 Mar 2010 19:44:25 -0400 https://reprap.org/forum/read.php?1,37355,40248#msg-40248 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40248#msg-40248
I bought some 50lb test braded fishing line. No luck. It started out great but seemed to flatten and then rolled up on the other layers. Still worked but the steps would not be the same length for those versus the lower layer.

My next thought is kite string. I'm worried it will do the same thing. Mayb I just need to find a source for really thin steel cable. Did some searching earlier with no luck. I'll try again.
Arvin]]> Arvin General Mon, 22 Mar 2010 19:14:46 -0400 https://reprap.org/forum/read.php?1,37355,40235#msg-40235 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40235#msg-40235

Arvin, once I build a fairly standard McWIre but with cable drive, I will try plywood variants. I think for quick and easy the draw slides are a possibility. But the old fashioned direct drive to threaded rod will be cheaper and easier to make than the cable drive with pulleys and guides.

Mike]]> rocket_scientist General Mon, 22 Mar 2010 18:08:21 -0400 https://reprap.org/forum/read.php?1,37355,40184#msg-40184 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40184#msg-40184
I think the cable drive is going to be the one. Have spent the last week going through several versions of a machine that can be built with a jig saw and a drill.

Here's a look at the latest. Several iterations and I'm back to almost what I had to start. Lots of megs lighter on the hard drive though.

Thoughts?

From reprap

The sketchup file is: [dl.dropbox.com]]]> Arvin General Mon, 22 Mar 2010 09:51:19 -0400 https://reprap.org/forum/read.php?1,37355,40146#msg-40146 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40146#msg-40146
It's not really supported anymore because it doesn't do a very good job printing Mendel parts, that's the reason for the push for a successor.

All RepRap is really looking for is a cheap printer that is completely off the shelf, that can produce the RP parts for the RepRap flagship printer of the time.

I think the long term goal is to make it possible, with no access to RP parts for you to go from

Design->Hardware store->Mcwire->Mendel->Next generation

The next generation printer will not necessarily be printable by a Maker bot or Mcwire.]]> spacexula General Sun, 21 Mar 2010 23:49:47 -0400 https://reprap.org/forum/read.php?1,37355,40145#msg-40145 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,40145#msg-40145
Is the McWire bot the "official" bot because it's the best choice/most reasonable/made by Zach, or simply because it was the most well-documented at the time... It seems like the old page that listed RepStraps listed the McWire along with Hydraraptor, Tommelise one or two others, so I'm trying to determine the official-ness of the McWire.]]> sircastor General Sun, 21 Mar 2010 23:35:36 -0400 https://reprap.org/forum/read.php?1,37355,39876#msg-39876 Re: McWire Successor https://reprap.org/forum/read.php?1,37355,39876#msg-39876
Somethings you might consider trying:

Instead of wire or cable consider using Dacron fishing line. It is woven very strong, has minimal stretch and has excellent grip when wrapped around things. I bet you would only need 2 or 3 wraps around the dowel.

How about notching the dowel or inserting a smaller section of dowel in the middle to gear it down and prevent the travel up and back?

How about an aluminum rod with sandpaper or some texturing?

Auger now has some NEMA17s but still need to couple it to the drive screws. Life has been getting in the way but perhaps this weekend I will be driving the XY tables with RepG.]]> bjbsquared General Thu, 18 Mar 2010 20:07:51 -0400