Next Steps

I now have 1 power comm board, 3 stepper boards 6 opto boards and 3 stepper testers assembled, programmed, tested and ready to deploy.

I am ready to order all of the remaining Cartesian robot parts from McMaster, Amazon, Farnell et al.
This leaves the extruder assembly, which Zach has a line on, and by rough count ~ 60 RP parts to complete my Darwin clone. Ordering the correct quantity of RP parts from AlphaPrototypes (expensive) or hoping that a set of RP parts falls into my lap are currently my only options. Ordinarily at this point I would pull out the drawings for the parts and start figuring out how to fabricate each one, but there are no drawings only STLs. I know there is a plan to have AOI drawings for each part, but if I understand the issue having the parts in AOI only provides a pre STL rendering, not a manufacturable dimensioned drawing. I am perfectly capable of building a "Darwin like" apparatus from the information currently available, but this does not advance the state of Darwin or the REPRAP knowledge base. At a quick glance there are a number of off the shelf parts that can be used without compromising the design. here are a couple of examples: Use Panduit cable clamp #CCS31-S8-M (mouser #644-CCS31-S8-M) and a 3mm (1/8") spacer of your favorite plastic (to keep the back of the boards from shorting out on the M8 rods) a M4 screw, washer, and elastic lock nut instead of the board mounts for the X, Y, and Power comm board. A similar approach using the same cable clamp and some 3mm (1/8") plastic angle stock ( glue it up if you can't find it) gets you mounting brackets for the opto boards. 1 millimeter (3/64") aluminum sheet will make fine opto interupter brackets and interupters. DOing this drops the number of RP parts required by 20%. Just to be clear I get it that we want to build as much of REPRAP through RP as possible, that is the point, but we have a chicken an egg problem. We can't build RP parts for one another if there are no REPRAPs. So my question for the collected brain trust is how do we move forward together? If the answer is wait for RP parts to appear, I am ok with that. On the other hand if there is general support for the idea of alternative fabrication I would propose dividing the RP parts into two categories, those that we can document for alternative fabrication like the ones I've mentioned, and those that we need to source through other means RP, CNC, resin casting... The goal here is to get onto the exponential reproduction curve sooner.
What are your thoughts?
Best,
Dan

You need to talk to Joost for alternatives for those parts you are missing. He did his out of wood!

Forrest,

Is there a place where there are dimensional drawings available or did he make the parts by guess and by golly?

I will follow up with Joost.

Best,
Dan

Joost talks about how he did it in the Sunday, June 17, 2007 entry in the builder's blog here at reprap.

This is the quintessential RepRap question, and one that I've spent a good long time both physically attempting to solve, as well as mentally dissecting. There are a few options. It basically comes down to how much you want to spend, how much time you have, and how good you are at manual construction.

If time is no obstacle, then wait until we are printing parts, heh.

If money is no obstacle, or a smallish obstacle, get a commercial 3D positioning device. attach to this our extruder (part kits coming soon...) and you'll be ready to get down to the work on actually extruding.

If skills are no obstacle, build a system from scratch. This will require good skills in construction, but is cheap and can have good results if you have the time and the skills. Emphasis on skills (something i've learned the hard way)

The middle solution, and one I think holds good merit would be to hack together a system from cheapish off the shelf parts. The idea James tossed to me, and I like is to combine a $50 cross slide vice (X/Y positioning) with a $100 drill press (cheap Z axis) Making mounts for these then becomes your hardest challenge. Get a bit of angle iron that bolts onto the bottom of the cross slide and use that to mount the motors on. You can easily mark and drill the holes for this. Remove the handles, then use silicon tubing + hose clamps to attach the motor to the stepper. You could probably even find gears on McMaster if you wanted more torque / better resolution. You would then clamp onto the wooden build platform (vice included...)

The drillpress could be similarly modified, possibly by taking the return spring out, or replacing it with a smaller spring. You wouldn't need to use the motor, just the Z drive it has. You might even be able to use this as a *light duty* milling machine if you really need to.

If for some reason the darwin i'm building doesnt work out, I'll be trying this approach. The most important thing we can do now is to get more people actually to the point of printing. As soon as we're reliably producing parts for a small enough price, we'll be able to speed up the iterative design process. We cant iterate and improve unless we have the capability to cheaply make new designs a reality.

We're close, and all it really takes is one or two people to break through, and things will get interesting. I'm thinking about offering a bounty to kick start this process, but am still working out the details in my head. =)

Also, if you'd like to discuss the design I've suggested, we should start a new thread.

A potential starting point from the ever formidable Vince Gingery...

[www.lindsaybks.com]

Ok,
I am stumped.
The builder blog I am looking at only goes back to 7/2/07.
What am I missing?

Thanks,
Dan

There's a list of links to old archives by month. It's on the left hand side, safely hidden below the ads and other stuff your brain is programmed to ignore.

here, i found some info on joost's efforts.

forum post: [forums.reprap.org]
blog #2: [builders.reprap.org]
blog #1: (i cant find it)

and of course: [www.reprap.nl]

[builders.reprap.org]

About the bounty Zach mentioned: why don't we pool resources and each put in a little for the winner to take ... I'm willing to put up $US100 (probably easiest transferred via Paypal) for the first person to reproduce a full set of Darwin parts.

Not much but its a start. Anybody want to add to that?

Thanks for the links

Got it.

Thanks

Edited 1 time(s). Last edit at 07/19/2007 01:11AM by Dan P..

Both Vik and Zack are building proper Darwins and Joost is bootstrapping one very nicely. I suspect that they will be cranking out Darwin parts as soon as they get used to printing in PLA and HDPE. Given my own experience with HDPE I'd have to massively redesign the parts to have a decent chance at printing them successfully.

Edited 1 time(s). Last edit at 07/19/2007 01:15AM by Forrest Higgs.

Dan,

I have documented some of my attempts here [forums.reprap.org] as well. I used a set of PDF drawings for the extruder (thanks to Bruce W). These are *not* dimensioned, but can be measured of the drawing (no time yet to finish this - they are attached).
Also, I have more material as in the forum post but did not yet get to finalizing it. I can post it here, but then it is at our own risk if it works or not ...
If you have any questions, let me know.

Joost

those PDF files look very useful. i'm going to add them to the extruder release files on sourceforge. probably later tonight.

Ok, Patience isn't my strong point.
I have been evaluating my (our) near term options for building Darwin. In an effort to reduce the number of low value high cost RP parts that I need to get made I have started blueprinting (2d dimensioned drawings) the current Darwin RP parts. My first attempt is the corner bracket. Darwin uses 8 of these. I have used AOI (though painfully) to extract positions and dimensions from the current corner bracket stl. I have redrawn these in a CAD program and created the attached PDFs.

Some of the features of the corner bracket cannot be made for all practical purposes without RP(the internal captive nut for the grub screws for example), some don't make sense to detail until the basic dimensions are set, and some are cometic (radiused edges on all the exterior edges) so I have left them out.
When I am done I hope to have a functionally equivalent part that is freely interchangable with the final RP part.
What I need from this clever group is some proofing of the dimensions.
If you have access to the real dimensioned drawing for the part or an RP part and caliper please take a few minutes to cross check my dimensions.

The drawings are the view labeled front in AOI, the view labeled left an the view labeled top. All dimensions are in millimeters.

Once I have a consensus drawing I will build or have built a proof set to verify that the part is a functional replacement for the original RP part.
The next part in my sights is the diagonal tie bracket of which darwin uses 20.

Thanks for the help.

Best,

Dan

Edited 1 time(s). Last edit at 08/05/2007 04:11PM by Dan Putman.

Hi Dan:
I have found it most easy to use Rhion 3D to open the STL files and simply save the file as a Rhino file converted into inch. Rhino 3D will do this automaticlly "asking what system of measurment you want before saving" once saved re-open the new file and use the tools to measure directly off the drawing.

Rhino 3D is a free trial software, but will only allow you to save 25 times. You can get around this by cutting and pasting many parts on a single work surface before saving.

The corner bracket is 1.750 X 1.750 X 1.250, the top center hole is .550 or 14MM if you like that metric stuff, the surounding holes on top are .312 or 8MM there placment is .375 from the outside edge of each corner, you get the idea.

Edited 1 time(s). Last edit at 08/05/2007 07:24PM by Richard.

Thanks this helps.

I know Rhino 3D, I will give it a try.
I posted the measurements in metric units to be consistent with the rest of the project.
I have attached an updated corner bracket front please double check that I got it right.

What are the diameters of the holes on corner bracket top and their respective locations?

Off to load up Rhino.

Best,
Dan

Dan,
I have some more drawings I have been using solid edge which is what the original Darwin drawings were made in from there I have been saving as a pdf file the size when printed is correct and then I have been cutting them out and pasting to wood the drawing. I will post some more here for the x and y axis parts this week.

Bruce Wattendorf

Thats great.
This is just what we need. The more parts that we can make by conventional means the better.
When I have the dimensions right on the corner bracket and Diagonal tie bracket I will post to scale PDFs for the same purpose.

Best,
Dan

HMMMMM
Well the side of the corner bracket is a little harder to explane but here goes.

It is 1.250 in height X 1.750 in width. devide it in half both ways with a line so you have four divisions (+) each hole is set in .375 from the edge along the center lines, the top hole and side holes are all .197 or 5MM and the bottom hole is .312 or 8MM.

the other side if you turn the bracket 90 degrees is the same except the large hole is on the top center and all other holes are .197 or 5MM.

Your attached PDF drawing is accurate.

I should point out that I have rounded down on the decimal equivalent of 8MM the acutual decimal value is .317

Edited 2 time(s). Last edit at 08/05/2007 10:27PM by Richard.

Great,

Here are drawings that ~ correspond to the dimensions you measured.
Plus a bonus Diagonal Tie Bracket to proof.

My plan is to replace the captive nuts with EZLOK threaded inserts with the appropriate holes over sized to give about the same size grub screw. The choices are 10 32 slighty undersize the 5mm grub or 1/4 20 slightly larger.
I am currently thinking of making the proof parts out of nylon or delrin.

All suggestions are welcome

Best,
Dan

You could make the corner pieces from two parts, with additional pegholes, and still use captured nuts.

Your way is probably better.

I thought of preserving the captured nuts, and if the torque needed on the grubs is too high, it is a better way to go (compression over shear).

The reason I am leaning towards the inserts is that getting batches of parts CNCed is easier. If there is demand for say 10 sets thats eighty corner brackets and 200 diagonal tie brackets. The price could be quite reasonable. As I think I said up post I got a quote of $563 for eight RP corner brackets, it is hard to believe that the total cost even with the inserts wouldn't be much cheaper.

As I look at the RP parts list the bed corner looks like the next thing to deconstruct, but I am open to suggestion.

Thanks for the thoughts.

Best,
Dan

The drawings look good to me Dan, I'd say your well on the way, however I'm not so sure nylon or delrin is the best choice to hand fabricate the parts. Both materials are tough that's for sure and that same quality makes them a pain to home fabricate. In some respects you would do better to have/make the parts from aluminum and this would also help with the set (grub) screw issue.

I thought I replied to your note, but I don't see it in the the thread.
In any case. I have 6/6 nylon 1.25"X2"X24", 7/8"X7/8"X36" and 16 3 foot 8mm rods coming from MCmaster. Price is $138 plus shipping. The nylon is about half the price of Aluminum and Mcmasters' price was better than any of the speciality plastic vendors I could find. I am sure at the right volume 1.25 X 1.75 nylon extrusion is available, but this should do nicely for proofing the parts.
It looks like EZLOK Finsert #270-332 for the Diagonal tie and #370-332 for the corner bracket, but we will see what we see.

Real life is intruding on my efforts so it is probably a couple of weeks before I get to have a go at fabbing up some parts.

Pictures, drawings, and progress report to come.

Progress of a sort....

McMaster suprised me and the day after I placed my order there was a box of goodies on my doorstep. I thought work on the brackets et al would have to wait, but suddenly I was ahead of schedule (this seldom happens).

I trotted off to the shop drawings of the corner bracket in hand and proceeded to make a smaller piece of nylon out of the big piece I ordered. I ordered 1.25"X2"X24" 6/6 Nylon, what I received was 1.265"X2"X24" 6/6 Nylon. No problem, but the first of several little discoveries.

My shop consists of a combo cut off/ upright bandsaw, and a tabletop drill press and most of the usual sort of handtools powered and not.
The first task to cut a 1.75"X1.75"X 1.265" block from the raw stock. I set up my saw to do the cutoff, measuring carefully (twice), turned on the saw and lowered the blade. a little cutting, but the nylon was melting as much as it was cutting. So I grab my handy can of cutting oil and me the saw and the nylon block are much happier. Even so the first cut took ~ 20 minutes. A little re jigging of the cutoff clamp as the material is now to small to just put it in the clamp and make the second cut and off we go. Lessons learned this cut only took ~ 15 minutes. There was a significant amount of deburring required to atone for the sin of letting the nylon get to hot.

Dinner and my bride were calling so that was it for last night.

This morning I headed back out to the shop and started laying out the corner bracket. I grabbed my block of nylon from last night picked the 2 best edges, my cuts were good, but McMaster's were better. These became my layout datums (all measurements referenced to these edges or the top of the piece.

Using my Starett machinist square I had the part laid out and ready for drilling in about an hour.

I bumbled around for the longest time squaring up the drill press table attaching a drill vise and positioning nylon block vise and table for drilling all the holes.

I started with a pilot hole and it too wanted oil. I progessed the hole to its final size and was pleasantly surprised that the larger diameter drills cut cleaning and quickly with far less oil.

3 hours pass, I have the finished part in my hand and it is very handsome.

I grabbed 3 8mm rods inserted them into the corner bracket all at 90 degrees to one another. The angles look good, but something isn't right. I take a careful look and I can see 8mm rod protuding into the 5mm* holes. I oversized the 5mm holes to accept EZLOK inserts, my first thought was that this was the problem and I had just screwed up big time. After I calmed down I went back to my desk and did a quick tolerance study of the corner bracket as drawn. Sure enough the 5mm holes for the 8mm Z rod holes interfered with the 8mm X and Y rod holes and this was without oversizing them for inserts. Richard was kind enough to help me with the original part measurements and they seemed pretty solid. I measured this and moved that and finally came up with a drawing that doesn't have any interferences, but the clearance is only .004" see attached drawing. .004' is almost exactly the goal for the accuracy of REPRAP (.1mm), but the minimum layer thickness is close to 10 times that thickness. So either my dimensions are way off, or we have a part of REPRAP that REPRAP can't build.

Notes on drawing The red dotted lines are the projections of the holes from Corner bracket top. The remainder of the drawing is from corner bracket left.
I removed the interference by increasing the clearance between the X and Y rods.
To make a buildable part the clearance needs to be larger yet.

Finally there is nothing that makes you appreciated the talent and skill of others than trying to duplicate it your self. Hats off to the REPRAP team.

Best to all,
Dan

would anyone here be willing to maintain the collection of PDF files we've generated for making these parts? right now i've put as many as i could find in the cartesian bot release up on sourceforge, but its hard to keep track of all of them.

just shoot me a PM or an email and we can work out the details.

Guys send me your .stl's and I will quote you the cheapest possible cost for your prototypes for the reprap in ABS. You can upload at absprototypes.ca

Here are a couple of photos of my nylon corner bracket proof part.
I have ordered received and installed EZLOK Finserts. The inserts have excellent holding torque. I can make The "270-332" inserts work everywhere except the top corner bracket that holds the Y motor mount. This has sent me back to the drawing board one last time to clean up some sloppy dimensions and switch to a shorter EZLOK Finsert "170-332" which will solve the Y motor mount problem and also make the part mold-able without need of a sacrificial core. Of course I will need to test the shorter inserts pullout resistance, but I don't think that will be a problem.