Monday, September 29, 2008


Second try at cutting steel

In which your narrator actually manages to cut the thin steel plate... read more

Wednesday, September 24, 2008


First try at cutting steel

Some time ago I resolved that I would take a crack at making Tommelise cut the thin sheet steel to do the field offsets for the linear tin can stepper that I've designed. I bought a little diamond cutting head for about $9 that works with my Dremel.

Field offsets are how you turn an ordinary air core electromagnet into a multi pole magnet suitable for making what is known as a "tin can" stepper motor. You can see here what I am talking about.

I plan on using a 3/8-24 inch threaded rod for a lead screw for now till I figure out how to cut my own lead screws. You can buy that kind of threaded rod in most good hardware stores. One of the gripes that I have with the Haydon linear stepper motors is that the 3 mm lead screw is far to flimsy for the lengths I'm using. Making my own linear steppers lets me fix that problem as well as controlling the costs. Now I just have to write a little routine that will generate the profiles that I need to cut.

read more



STL and Forks in the Eye, and new Slice Format?

So, I spent a while ( well more than a while ) fiddling with some STL files, trying to get them to slice in skeinforge.

STL is really nice in some ways-- easy to parse, easy to understand.

But it utterly sucks when it comes to trying to figure out hwow to represent a solid.
But, the STL Im working with has holes in it.

See, commerical cad packages dont seem to try to create "good" stl, -- its just for display, so they are like swiss cheese. I tried for hours to fix them with meshlab, remesh, and other tools, but they are painful. I'm using SolidWorks, which is a fantasic solid modeler, but creates unsuable STL.

Then, i realized that it has internally a brep represenation of each face-- just what i need to create quick and accurate slices.

It took only about 4 hours to whip up a plugin for Solidworks that will slice a model and output the edges, loops, and faces for a solid-- with exacting precision and no holes. Now I just need a way to get that into skeinforge to do the rest of the steps.

Then i started wondering: what if we had a standard slice format that contains the boundary rep of each slice, so that if it can be created in AOL or other cad packages, its easy to avoid STL?

Something like this, which uses standard loop representations ( with ccw or cw direction to indicate "inside" )

solidName: "MySolid"
slice-z: 0
loop: "outer":
#x-y points
1.0 1.0
1.0 1.1
end loop
loop: "other"
#more xy-points
end loop
slice-z: 2
#more slices

The idea being, that this could be created directly very quickly and easily from the solid packages ( bsh script in AOL) and be an easy, portable way to then do more difficult slicing/dicing in another system...

Sunday, September 21, 2008


Milling around objects

While I have been able to mill gears with large tooth counts with milling data from Slice and Dice the number of times that Tommelise has to raise the head, move to a new part of the milling job lower it and start cutting plastic again has, not to put too fine a point on it, been excessive. A simple 18-tooth gear profile cutting job that ought to have taken 30-40 seconds was taking 8 minutes. My efforts for the past few days, as a result, have been aimed at reducing that.

Actually, it didn't take too long to get Slice and Dice to minimize the number of milling segments drastically.

I was able to reduce the 18-toothed gear job down to three loops; the outer square perimeter, the gear perimeter and the drive shaft hole. read more

Wednesday, September 17, 2008


Arduino Rebrain <> Sanguano

I have mentioned, on the forum, my attempt to make an Arduino behave sort of like a USB Sanguano by brain transplant.
I have blogged the details here : Rebrains.

Tuesday, September 16, 2008


507 Mechanical Movements

About a week ago, I stumbled upon two wonderful reprints of turn-of-the-century compendiums of simple (and not-so-simple) machines. Since both books were published in the US prior to 1923, they fall firmly in the public domain. Thus, I decided to scan and re-publish them in digital format for use by everyone.

Here is the first.

Note: the front and back cover art, though produced after 1923, appears to consist entirely of drawings from within the book itself, and a simple statement of the title and author. As such, I believe copyright claims on those works to be unlikely. If anyone objects to their presence in the digital copy, I can remove them, however I'd prefer to give the publishing house as much credit as possible for digging up these gems, and send some business their way.

More to come.

Sunday, September 14, 2008


18 teeth - living and learning

In which the narrator discovers that milling has only a passing resemblance to additive printing...

read more


Extruder assembly - Laser Cut parts kit

I recently purchased 2 rolls of filament, one roll of HDPE and one roll of ABS in anticipation of the completion of my RepStrap using the Laser Cut parts kit. I'm hoping to be making my celebratory cup within a week or two.

I went back to my partially assembled extruder mechanism and installed the proper bearing that I purchased this past week. I then tried to manually feed a piece of filament and found it to be a little too tight. I disassembled the extruder mechanism and found that the relief that was laser cut in item #5, the top plate, had a rough finish and was causing the binding. I took a fine file to it until the piece of filament ran smoothly through the slot.

Here's a picture of the part after filing:

I also filed the slot where the drive bolt runs while I was there. I then assembled the extruder according to the procedure but I left the 2 M5 bolts in step 1 and the 5 M3 bolts in step 6 loose. I placed a ;iece of filament stock in the slot and wiggled the extruder assembly until I got the best feel of the stock sliding in the slot. After tightening the bolts I still found a spot that bound a little when the filament is beyond item #4 in the assembly and the entering the hole in item #5. I took a small round file and filed this until there was absolutely no resistance when feeding the filament all the way down to the feed screw. The size of the slots and holes for the filament could be opened up a bit in the design. I don't think there is any requirement for the feed slot to be as tight as it is. This would also help the alignment between items #4 and #5 in the assembly.

I fired up the extruder electronics and software and voila, the drive mechanism is feeding filament! I'd like to figure out a way to measure the accuracy of the feedrate. If anyone has any ideas I'd like to hear them.

Regarding the laser cut parts, there seems to be a side effect of the laser cutting process when the cut is not through the entire part but rather to a depth less than the thickness. It appears as if the acrylic becomes brittle and rough in these depth cuts. I found that all it takes is a small file and a little careful filing to clean them up. I have done this with some of the other parts that have partial depth cuts. The nut cavities in the belt clamps are difficult to get with a file so I just scraped away any axcess with a small awl. This little bit of clean up is not a big deal and as we all know is really part of any assembly process using most types of materials. I don't consider this to be a problem at all.

The last of my parts should be here next week and I will be building the heater barrel assembly next. I'll also have the rest of my parts to finish the partially completed sub-assemblies. Almost there...

Saturday, September 13, 2008


Shifting gears

Wherein the narrator's long-held dream of building cheap gear trains becomes reality...

read more

Monday, September 08, 2008


My Roller & Rail progress.

I have been distracted by Jazz & Curries a lot this week, but seeing Timothy's nice chunky tube&flange design I thought I would also blog my similar stacked bearing arrangement :
Roll On!


PIC24 printing from SD card

A few pictures to show where we are, this output has been made without any stops and starts of the extruder.

The twin drive extrusion head that produced these parts is not quite the same as the standard Bitz from Bytes item.
The head features Opto speed control feeding back to a PWM output to the two motors.
Extrusion speed held +/-1rpm.
The temperature control is just an "on if under, off if over" control via a FET.
Temperature is held +/- 1 DegC
Round other side of the controller is an LCD readout of the setting and achieved values for temp and speed. Both temp and speed have adjustable POT's for local setting and quick testing, it saves so much time if you can make these changes on the fly.
The next step is to hook up the serial link and make changes via the G-Code.
I still have a fair way to go but first indications are looking good.

Build of the bracket, left and below, is without any fill

It aint perfect but when I get the cooling sorted out!
The rather nice rim on this "glass" is due to a failure in the Z-Drive part way through.
All parts are saved onto the SD-Card, ond only take a fraction of the space available.
Each of the parts above had well over 5000 lines of G-Code. I dont think we would have any problem putting a whole machine on a single card.

Sunday, September 07, 2008


Taking shape... of what, I don't know.

Managed to bodge together an experimental gantry and X axis assembly today. I'm not at all certain this is how it will end up looking, but it's something to play around with at the moment while I work on the Y and Z axis bearing assemblies.

The pipe fittings aren't the most precisely machined things in the world. It takes some fiddling - over-tightening some joints, leaving some loose - to get things lined up right. Even so, one of the bearing assemblies needs some further adjustment, which I'll have to do by re-positioning the skate bearings on the bolts. Oh well... that's why they're there.


RepStrap Build - Laser Cut Parts Kit - Z tensioner modification

I have made a modification to the Z axis tensioner that seems to solve problems that I was having when trying to get my Z axis working. I tested motion in the Z axis and was successful on the first try with the modified design without any adjustments. I don't have my Z sleeve bearings yet so I'm not running it too much but in my early tests the belt didn't slip and the motor didn't stall. The modification places the tension on the center of the belt and I think that the placement makes the Z axis load, from the perspective of the motor, appear the same in either direction. I know that the Z axis only moves in one direction during extrusion and it may not matter much but I think the load to the motor should be balanced. Some of the current and future efforts to mount different toolheads that will utilize the cartesian robot for applications other than 3D printing and may require the Z axis to behave the same in either direction of travel. In the original design, the tensioner is mounted on the diagonals and is placed offset to one side and very close to one of the Z pulleys.

Here's a diagram:

The modified design places the tensioner in the center.

Here are a couple of pictures of the modification:

I took the 2 diagonals and mounted them horizontally, 1 above and 1 below the centerline of the belt. I think that the cartesian robot is rigid enough to deal with the lack of diagonals on the tensioner side. It will still have support in the horizontal direction on the tensioner side. If more support was required, diagonals could still be mounted on this side if the horizontal rods in the Y dimension were longer such that they are long enough for both sets of diagonals. The lack of diagonals becomes a feature though because when your part is complete you can just reach in and pull your part out without the diagonals in the way.

When mounting the horizontal supports for the tensioner take care to place them such that they do not hit the Z pulley flanges. With the motor off, move the belt by hand to make sure there are no clearance issues. When everything moves without interference you can fire up the electronics and test without breaking anything. Here's a picture of the clearance I'm talking about.

Mine are not perfectly aligned because this was merely a test to see if the design worked. I am going to replace the broken acrylic parts that mount the diagonal support rods with machined delrin or aluminum. I feel that these parts need to be stronger and made from a less brittle material. I have broken more than 8 of them. Haven't thrown any tools yet though! I must be getting mellower with age...

One last note: The parts that mount the tensioner to the rods, item number 89 on the Z Sub-Assemblies drawing, were originally designed for the rods to pass through them diagonally and could be re-designed specifically for their use with the rods mounted horizontally. They seem to be sturdy enough and are working fine but they could easily be re-designed to accommodate mounting on horizontal rods if a problem comes up.

Friday, September 05, 2008

I promised Forest some more pictures. Here they are, much improved courtesy of a Canon SD1100 IS. The original bearing assembly is on the left, a freshly assembled one with one of Andy's suggested improvements is on the right.

Moving the turnbuckle onto the inside of the floor mounting flange allowed for the use of one less nut per bolt, and moved the bearings closer to the flange, seemingly improving alignment somewhat. Excellent.

I haven't tried mounting additional flanges on the outside of the bolts yet. At nearly $4 each, they're already the most expensive part of the assembly by far. If I can get away without doubling the number of them needed, it's a big win.

Thursday, September 04, 2008


Thermal Managment

Just finished putting some of the extruder assembly together for the PlyRap. Being mindful of the issues other builders have reported when printing higher temperature plastics I wanted to do something useful to mitigate this.

The pictures show a short length of Aluminum Tube split with a hacksaw and placed around the PTFE Barrel mount. This should give the PTFE some degree of heat sinking and help it run a little cooler.

I would rather have used a finned heat sink but lacking a lathe I had to settle for a straight tube.

As is usual I used a hose clip as well. both to make a good contact and to hold it all together just in case the heat sink doesn't quite work as planned.


Repstrap and CNC progress update

First of all, please excuse the quality of the image, it was taken with an OLPC, not a traditional digital camera.

I haven't posted much, but not for lack of work. I've been busy working on a repstrap / CNC router design that's buildable for a low cost (< $200 including all tools and materials) from parts available in your local hardware store, with minimal or no tools.

It's been slow going, but a recent prototype linear bearing assembly has shown promise. It's my first to come close to meeting those criteria - inexpensive, no tools required, and all parts available at the local Home Depot. It's built using 1/2" internal diameter black pipe and fittings, 5/16th allthread rod and nuts, skate bearings, and double-eye-hook contraption I don't know the name of. :) The eye-hooks are used to tension the allthread rods, rotating the black pipe floor-mounting flanges and trapping the guide rail between the sets of skate bearings.

Total parts cost for this assembly at the local hardware store was about $15 US.

Now that I have a linear bearing that works, I hope to quickly move on to an entire repstrap capable of being assembled without so much as a single power tool.


Touch Probe and Pen Adapter Sleeve

Here are a couple of accessories for the Universal Pen Holder and Touch Probe Tool Head again these are documented here.

There is a Pen Adapter Sleeve for Staedtler Lumocolor Permanent Pen and a Touch Probe so that you can use it to digitize your own hand made objects.

Monday, September 01, 2008


Milling Stuff

In which Tommelise 2.0 goes operational in a mode not anticipated by the narrator...

Since the slave I2C chips haven't arrived, I decided to see if I could get Tommelise 2.0 runing as a 2.5D CNC milling machine since in that mode an extruder controller board isn't required. I began by milling blanks of poplar.

In this mode, I was able to debug the T2 positioning robot without using my 1/2-inch plastic sheet...
read more


Universal Pen Holder and Touch Probe Tool Head

I have designed, built and documented a Universal Pen Holder and Touch Probe Tool Head I wanted a way to put etch/photo resist directly onto PCB's, pen plot, photo plot, digitize items I had made ready for RP'ing and most importantly an easy way to setup and check mechanical alignments on my Darwin.

Hope this is of value to someone else as well.

Follow the link above to view the documentation.

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]