Sunday, June 12, 2011
Getting Rid of Mendel's Nuts
I’ve been working on something for a bit, and the results are starting to surface. I made a rash promise in the #reprap IRC on freenode.net to put up a teaser on Sunday, so here it is. It’s a RepRap variant with no fasteners in the frame, using sockets to accept the (in my case) 9.4mm aluminium tubing struts. You can assemble it with my favourite tool - the mallet! So far the only screws are those needed to hold the motors, belts and extruder guts in place (though as the extruder isn’t built yet so that ain’t saying much). Holes are provided for adding glue to the structure, and for the particularly fastidious they can be drilled and tapped as well. I’ve even made the glue cavity the shape of an M3 nut so you can in theory make strategic joints releasable to aid portability if you really do like your nuts..
Um, ignore the basket of fresh clothes. Suz thinks the workshop is some kind of laundry. Even keeps the washing machine there.
The shedding of M8 rod has other advantages besides removing the need for many fasteners and making the whole thing lighter. A good one is that the frame struts can now be smooth enough to run the axis skids on - running a skid on M8 thread is a teensy bit like sliding down a grater. This removes the need for some lengths, lightening the structure still more. Make it from hunting arrow tubing and you’d be going featherweight. Use bamboo and go eco. Many options.
The design of the new Z axis hinges on the use of a Sarrus Linkage, ahha, ahha. I’m hoping something cunning can be achieved with a DC servo and string, but until then a NEMA17 and a piece of M3 threaded rod will do. No there is not a lot of Z travel in my quick bodge (unless you count the slack when it wobbles up and down at present), but there is enough to get on with. The Sarrus is actually a great fit mechanically for the Mendel-style frame, because the Sarrus gets narrower as it gets taller - just like the Mendel’s triangular profile.
The X axis is right at the top, and a one-piece extruder/carriage slides back and forth across the top two runners. I’m still coding the brackets and idler for that. Yes, coding. The whole thing is parametric. The Y axis is more firmly in reality. This is a cunning sled that pulls itself together with a couple of rubber bands. This causes the sloping faces of the runners to move up the side of the rod as they travel, until they hit the top. This kills wobble automagically and is remarkably tolerant of mis-aligned Y bars. It’ll even work upside-down! The proper Y motor bracket is still being designed, so for mechanical testing I used fancy stuff called string...
The vertexes are going to change a bit from the picture. They’ll have to be a bit wider to accommodate the X carriage, higher so the Z axis does not scrape it’s butt on the ground, and braced for extra stability. None of it needs support for printing.
That’s enough for now. I’ll get back to it. More news & photos as I sort releasable stuff out. Just remember that I’m going to be releasing this as soon as it is vaguely usable, and as yet it isn’t!
Um, ignore the basket of fresh clothes. Suz thinks the workshop is some kind of laundry. Even keeps the washing machine there.
The shedding of M8 rod has other advantages besides removing the need for many fasteners and making the whole thing lighter. A good one is that the frame struts can now be smooth enough to run the axis skids on - running a skid on M8 thread is a teensy bit like sliding down a grater. This removes the need for some lengths, lightening the structure still more. Make it from hunting arrow tubing and you’d be going featherweight. Use bamboo and go eco. Many options.
The design of the new Z axis hinges on the use of a Sarrus Linkage, ahha, ahha. I’m hoping something cunning can be achieved with a DC servo and string, but until then a NEMA17 and a piece of M3 threaded rod will do. No there is not a lot of Z travel in my quick bodge (unless you count the slack when it wobbles up and down at present), but there is enough to get on with. The Sarrus is actually a great fit mechanically for the Mendel-style frame, because the Sarrus gets narrower as it gets taller - just like the Mendel’s triangular profile.
The X axis is right at the top, and a one-piece extruder/carriage slides back and forth across the top two runners. I’m still coding the brackets and idler for that. Yes, coding. The whole thing is parametric. The Y axis is more firmly in reality. This is a cunning sled that pulls itself together with a couple of rubber bands. This causes the sloping faces of the runners to move up the side of the rod as they travel, until they hit the top. This kills wobble automagically and is remarkably tolerant of mis-aligned Y bars. It’ll even work upside-down! The proper Y motor bracket is still being designed, so for mechanical testing I used fancy stuff called string...
The vertexes are going to change a bit from the picture. They’ll have to be a bit wider to accommodate the X carriage, higher so the Z axis does not scrape it’s butt on the ground, and braced for extra stability. None of it needs support for printing.
That’s enough for now. I’ll get back to it. More news & photos as I sort releasable stuff out. Just remember that I’m going to be releasing this as soon as it is vaguely usable, and as yet it isn’t!
Comments:
<< Home
Another vote for the Sudo mallet as tool of the year. I like the repurposing the top bars as x sliders. Very efficient. One gripe, when a project has as it's main goal removing the nuts, the blog post should be far more creative. How about "Emasculating Mendel" or Announcing the new Mendel Eunuch" or "Castrating Mendel"
That is just so cool.. It looks like its come out of Millet's and is the going camping version of Mendel.. 3D printing while at Glastonbury?
Or the Hitchhikers 3D printer... backpack the world with your reprapable 3D printer spread the word.. far and wide.!
You just need a solar powerpack to power it.
Or the Hitchhikers 3D printer... backpack the world with your reprapable 3D printer spread the word.. far and wide.!
You just need a solar powerpack to power it.
I've been playing with a similar design, but using FIMO polymer clay to make the corner "brackets". I used 10mm Al tube and a bit of balsa wood which should make it light enough to be driven by small servos. I'll try making some of these nutless parts when they are released.
It seems to me that you'll have to redesign the z-axis to have a horizontal threaded control rod. Instead of the rod passing straight up and potentially sticking through your printing surface it should pass from one center joint across to the other horizontally where it won't harm anything by sticking out the end. I'm still not sure how that would work if you end up with a motor hanging out one side though...
The threaded rod on the Z axis is fixed to the deposition bed (or will be shortly). It won't go anywhere :) The tricky part is detecting Z=0.
I'm really looking forward to seeing this! I'm guessing that it should bring the price down by a bit and make the vitamins much easier to source. I'll probably start printing one out as soon as you do a first release.
What is the width between the two top bars? I'm wondering if a MBI MK5 hot end will fit.
What is the width between the two top bars? I'm wondering if a MBI MK5 hot end will fit.
I can see what linkage you can attach to the z-axis motor that doesn't require a hole in the desktop holding the device. I do see that if you have the z-axis braced horizontally, you have very simple and effective mechanism, see http://en.wikipedia.org/wiki/File:Jackscrew.jpg
PS: Sorry, forgot to mention, I really like the simplicity in the overall design. To me, it looks very elegant.
If you only use half the sarrus linkage's range of motion you can easily use a vertical Z leadscrew. Other than that, I'm not certain.
Also, I am chomping at the bit to build this. I love the way it looks, the simplicity and that it is parametric.
So, Vik, I'm assuming you are developing this in OpenSCAD. If so, I would love to help. Any way you would be open to collaboration to help get this out faster?
Thanks!
So, Vik, I'm assuming you are developing this in OpenSCAD. If so, I would love to help. Any way you would be open to collaboration to help get this out faster?
Thanks!
Actually, I've now sketched up with two variants on a solution that would allow full range of motion on the Z axis with a stepper motor mounted just below the base of the Z axis and one with it mounted on the bottom of the print bed. A Z=0 switch should be pretty easy to incorporate as well.
As far as detecting z=0, if z is when it's flat, I know there are small detect switches that have a tiny trigger, normally open, that might be able to fit within the sarrus linkages so when it closes, it gets the "signal", to pardon the pun.
http://www.radioshack.com/product/index.jsp?productId=3190382&filterName=Type&filterValue=SPST is the perfect example, tiny and simple, albeit overpriced (even after the employee discount (RS corporate, don't read this!)).
Post a Comment
http://www.radioshack.com/product/index.jsp?productId=3190382&filterName=Type&filterValue=SPST is the perfect example, tiny and simple, albeit overpriced (even after the employee discount (RS corporate, don't read this!)).
<< Home
