Its just not right!

Well, Im sat here marking GCSE coursework with Copperrap 2 sat at the end of the desk.(image shown to the right)
The more I look at it, the more I realise that it just isnt right! If I do finish building it as per the current plan (see other posts), then I dont think the final machine will function in anything like the way I would wish. The print area is too small in the X and Y direction, and too large in the Z direction.
I am going to scrap this version and have a serious re think! Mendal is so obviously well though out, and I think I need to create something similarly well designed. I may go along the Bodgeit route and build a bodged Mendal design, but I think that the Mendal design doesnt lend itself to that approach for exactly the same reason as I have just stated - it is carefully thought out and bits that are included are there for a reason.
oiaohm made a very good point when he said that the Z axis is easiest to configure by mounting it above the bed, and I may take that into account. As I see it now, CopperRap 3 is likely to be a larger version of the top half of CopperRap 2 with a Z axis connected directly to the head. I could use two NEMA 23s to control the X axis, another NEMA23 to control the Y axis and a NEMA 17 to control the Z (its nice to have the motors available now!)
Anyway, just thought I would get that off my chest, back to marking now!

The full monty

Ive really got into this sketchup lark, once youve mastered the basics it becomes just like playing with Lego! You build components, copy and place them, snapping them as appropriate to existing locations. This model took me about an hour to produce:-

It shows the full geometry of Copperrap 2. The red pipework is already built and the carriages at the top are as I described earlier. The grey posts are 8mm screwed rod seated in skate bearings. In this version, they will all be connected with a belt so that the bed can be lowered by a stepper motor. The extruder will be fastened to the large green carriage.

Ive been asked by a couple of people for a copy of the sketchup file, here it is...

Click here to download the CopperRap 2 Sketchup model.

Proof of concept

Its time to see if the method of construction may actually work!

I printed the drawing out from Sketchup, as a sectional plan view. This was extremely frustrating! Sketchup seems to deliberately offer a limited amount of print options and getting it to accurately print is another matter! The printout did reveal that the position of the cutouts wasn't quite central, I hadn't noticed that when I was working on the model.

I taped the printout to a piece of polypropylene cutting board in true "Bodgeit" style - I didn't want to stick it down because I needed to lift the drawing occasionally and see what was underneath.

I drilled through the printout with a 2mm drill to mark the position of the KD fittings.

The cutouts were produced by drilling the large central hole with a spade bit and the cutout ends with a 10mm bit. The spade bit doesn't give a very neat edge and is tricky to clean up. The straight cuts were done with a stanley knife, which is quite difficult to do, the first cut is easy and seems to go to a depth of 2mm but subsequent slices don't make much impression. A hacksaw leaves a very rough edge, so I will have to look at a different technique, I'm thinking that a hot blade may work well, or possibly mill the plastic with a dremmel.

The bogey assembly is made using a short section of 8mm aluminium tube which I cut using a pipe cutter. The KD block is drilled to accept a 6mm bolt (I bought a box of mushroom head bolts from Wickes, but they only have up to 6mm) Sticking the lot together gives:-

Now for the proof of the pudding! I fastened everything together using ¾ inch wood screws, which are a little too long as they project through the other side of the board slightly ( I will dremmel them flush)

Well, that seems to have worked, the bearings run very smoothly on the 15mm pipe (that's a copper section showing, on Copperrap 2 I'm using chromed pipe which may even be smoother.) I'm very pleased with the accuracy of this technique, I need to see if the 2mm gap I have left between the sides and the top (you can just see the pen lines on this picture) is enough. The board will be folded over to make the top part of the central carriage shown here:-

The end section will be bolted on using the 6mm bolts that hold the two bearing bogeys.

New year approaches and I have a shed load of work to do before we return to school so its unlikely that I will make much more progress before next week. There are two more cutouts to make on this carriage, 5 more bogeys and an end board. To make the other two carriages needs four more cutouts and 10 more bogeys. It isn't obvious from the picture, but the two Y bogeys are connected together with a length of 8mm threaded rod through the KD fittings of the bogeys. If you use a nyloc nut on the rod, it is a very tight fit on the end of the chromed tube.

Once the carriages are made and fitted, I will have to decide on how I connect the Y stepper. The X stepper is shown on the picture, it is fastened to the larger Y carriage. I could use two steppers, one on each side, or try to make a more complex setting, maybe run a shaft across from one side to the other with a belt connected to each side. An even more complex method would involve a longer belt (or cable) running right round the top of the carriages.

I had originally planned to have a threaded rod at each corner of the bed to drop the whole bed down after each pass of the extruding head. Ive seen reports that this is liable to binding so I may use an alternative method. Bodgeits comments about the bed only having to move in one direction got me thinking about using a long wedge to control the Z bed.

The stepper would pull the two purple wedges forward with a threaded rod, which would lift the red wedges and hence the main bed. If friction is a problem, it would be easy to make the wedges run on a couple of skate bearings. My original design has a good 250mm of Z travel, but as this is only a rep-strap (ie it will make a "proper" Mendal), then I don't need anything like that amount of travel.

Evolution

I cant believe how time flies! Obviously I havent done very much with CopperRap in the last couple of months but the RepRap community has been hard at work. The big news is that the Darwin design has now evolved into Mendal which is a very different beast that seems to have cured some of the issues with the original RepRap design.

The RepRap blog also has some very good explanations as to why the bearing designs work as they do and Bodge It (http://repstrapbertha.blogspot.com/2009/12/mendel-stand-base-board-to-keep-it-neat.html) is doing some sterling work in making a "strap" version. The discussion between oiaohm and bodgeit raised some very good points.

The outcome of this is that I am still going to develop CopperRap 2 but I will be amending the design to incorporate some "evolution".

Bearing carriage design comes first. I havent got very long today but a quick play with Sketchup produced this:-

This design uses three skate bearings held in place by two Fixit blocks. Three sets of bearing bogeys are held at 120 degrees to each other and the whole lot is duplicated to prevent twist. Ive added a top plate with a stepper motor to show how this will look.

Ive made a discovery with the polypropylene board . You get a much cleaner cut if you use a sharp knife instead of sawing, and by scoring the board half way through, you can make a bracketed construction that doesnt need any additional fixing. In the above design, the poly board can be made as one unit and bent twice.

Here is another view showing the X and the Y carriages in place:-

Sketchup has allowed me to experiment with positioning etc without making mistakes, I can explore the geometry and work out how it all fits together which is something that I normally do in my head. It is a frustrating application to use though, but then again I suppose this is because it is so powerful and I an lacking in skill!

Ball Chain

After an interesting weekend (lets just say it involved being stuck on some cliffs until 3am and a helicopter rescue!), Ive made a lot of progress with CopperRap 2

Ive ditched the idea of using screwed rods for the X+Y axis, and Im going to use a toothed belt. The belts arent easy to get hold of and are quite expensive so Ive been looking for alternatives.

Many RepRappers have used Ball Chain, which is readily available in the DIY stores. I made some rough calculations and worked out that I can make a 14.98mm pulled with 16 "teeth" that will link in with the balled chain that I have bought.


And here is a pulley set up:-



It was made by spinning a piece of acrillic sheet in a drill and filing it until it was the correct diameter, then mounting it on a stepper and moving round in 22.5 degree steps. I held a dremmel with a suitably shaped bit and made a ball shape indent each time the motor stepped. It was very crude!

The gear was a good fit, and with a suitable jig, I can make high quality and repeatable gears.

I managed to test the gear, I tied six half litre bottles of water to the chain, and the NEMA 23 motor easily lifted them. I was about to video it, but cracked the gear when I was adjusting it - typical!

Home built Delta Robot

Home built Delta Robot: "

Theres an interesting thread going on at RoboterNetz.de (translated) about building delta robots. For those who haven’t seen delta robots before, be sure to check the video after the break for some really impressive agility displays. A delta robot usually has 3 arms connected to a single point at the end. This configuration allows the end point to maintain its “level” while the whole unit is moved, usually very quickly. There are a few people building them in this forum. The one pictured above is [Asurofilmchen]’s version, but you should also check out [Radbruch]’s.

[via Hacked Gadgets]

"

ReplicatorG

Its been an interesting couple of weeks!

I thought that once I had three stepper boards and a Sanguino it would be easy to start to draw shapes, attach a dremel to mill objects etc. but I was so wrong!

I managed to upload the reprap firmware to the Sanguino, once I had figured out that you had to press and hold the reset button, waiting until the Arduino software told you how large the sketch was, then release the reset button. I now works 100% of the time.

I then installed the RepRap host software on my PC and wrote some simple G Code files so I could test it out. The RepRap host wouldnt do anything with the G code files, it didnt even display them. I added the Ruby stl script to Google Sketchup and made a very simple file (I tried a circle with one quadrant removed so it looked like a pacman, and also a simple square). I was able to export the files as .stl files but when I imported them into the RepRap host software, they showed up as apparently random triangles all over the place.

I struggled with this for quite a while, but just couldnt get it to work.

Hunting around the MakerBot site, I came across "replicatorG". I downloaded version 0005 and had a go at defining a "machine" by editing the machine.xml file. One of the things you can specify is the port that you are connecting to, but ReplicatorG didnt seem to pick it up. It also wouldnt let me add another line to any G code programs. I emailed Zach Hoekan and had a reply before I had chance to boil my kettle! He passed the issue over to Adam Mayer who also replied very quickly. After swapping a few messages, Adam said if I could wait a few hours he would have version 0006 ready which would cure these problems.

The response from Adam and Zach was nothing short of amazing!

Sure enough, version 0006 arrived and cured the problem of being able to edit a g code file and also specify the serial port from the machines.xml file. It was at this point that I realised I had the wrong firmware, ReplicatorG needs the version 3 firmware!

Once I installed the new firmware, I was able to write a g code file and build it so that CopperRap printed it! Progress indeed!

At long last, I have a machine that will do something useful.

There are still issues with ReplicatorG, it doesnt do G2 and G3 (arcs) very well, in fact they are are completely wrong, but Im sure Adam will sort this soon.