Friday, June 26, 2009
reliably repeatable positioning
I've been fascinated for some time by the idea of using cheap optical mouse inards to augment or replace repraps' existing methods of determining tool location. The main problem with existing designs doesn't seem to be accuracy (a 1.8 degree stepper, half-stepping, with a 10 turn-per-inch lead screw achieves 1/4000th inch / step granulaity), but repeatability. After hundreds of movements in various directions, how can we ensure we are in _exactly_ the same place as before?
Steppers and servos go a long way toward solving the repeatability poblem, but both are expensive.
A few days ago, I read a charming article about a simple binary pattern used by some pediatricians (via stadiometers) to measure height. (1) The pattern is called the Gray Code. It's essentially a highly accurate, fault-tollerant way of encoding position into a black-and-white pattern, easily printed on an axis (or a sheet of paper later glued to an axis) and it only requires an optical sensor with a few pixels resolution to read - simple enough for even our lowly Atmega microcontollers to do the required image processing in real time.
In short, it's a printable pattern perfectly suited for granting an optical mouse repeatability. Numerous people have extracted the graphical data from an optical mouse's ccd... see: http://www.contrib.andrew.cmu.edu/~ttrutna/16-264/Vision_Project/
A 600dpi laser printer could print a Gray pattern with 0.01 inch granularity, with each binary digit of the pattern encompasing a 6x6 dot square - 36 dots. Just an educated guess, but that seems like it should be granular enough for an opical mouse image sensor (which can do about 32x32 pixels at a time) to resolve.
Monday, June 22, 2009
There's a new kid on the block supplying parts for RepRap builders. They're based in Germany, so any orders from within the EU don't have to pay import duty. Checkout their site here.
Sunday, June 21, 2009
Roosevelt High School RepRap
Due to my business travel the group photo is a little bit abbreviated since it was taken the last day of the school year and three hours after everyone was let out for summer break! There are another five boys who have been contributing time to the project who will remain nameless for now LOL!
|From Roosevelt High School RepRap|
Pictured are their instructor Mr Ruff, my self, Ben and Ryan standing next to a Pro Light 1000 mill that we are using as a repstrap motion platform.
Progress is not over as Mr Ruff has kindly offered to come in Fridays (on his own time) to allow the boys to continue working on this project over the summer break.
We had a number of false starts and blind alleys that we went down. But the boys have learned how to design and cut acrylic parts with a laser cutter in the lab. They have cut a full set of Darwin Ponco acrylic parts that they hope to ultimately make into a standalone Darwin. Funds and time in the normal school year ran out to complete this effort.
The students have access to a Pro Light 1000 mill which is used in a lot schools here in the states. It has an attached interface box that has I/O ports that are controllable with M codes that we will use to interface to the extruder controller.
|From Roosevelt High School RepRap|
They also have an Epilog laser cutter which is an amazing precision machine; which after seeing it used I can understand why Zach was able to form Makerbot and go into production of the CupCake with it.
The lab also has a Z-Corp’s powder printer that has not been reliable in a school setting and was the inspiration for building a RepRap for the group. It also takes up significant space in the lab which could be put to better use.
The design goal is to be able convert the mill to a repstap by simply placing a drop on platform that holds multiple extruders for potentially doing multiple colors or building duplicate objects at the same time. The base design holds three heads and has provision for wingman extruders out to each side for objects that are less then 140mm in width.
|From Roosevelt High School RepRap|
Ultimately the front plate will be obscured by the PCB boards for the electronics. More to come!
Thursday, June 18, 2009
McWire Build: Cut your own x stage.
First I would like to say hello. My name is Charles Gantt (scphotog on the forums) I am in the process of building a McWire Cartesian bot. I am building it with the intent of printing Darwin or Mendel parts and then converting it into a small mill. In the 2 weeks I have been planning and building my McWire I have ran into only 1 major problem. The Stages would not print to the right size. I would like to make a note that only the X,Y,Z stages were not printing right. All the other files printed to the correct size.
I tried printing them from 3 different computers, 4 different programs including autocad 2006, and 5 different printers, one of which was a professional Roland plotter. The same problem every time. The drawing was scaled too small. So I came to the conclusion that I would be better off reverse engineering the stages and creating new technical drawings. The first one I have finished is the X stage. I would like to post a .jpg of that drawing so you at home can reference it if you decide to cut your own plexi for your McWire. I will also edit this post and post the source file in .dxf format for those of you who want to play with the design.
Please take note that this design will most likely not match up with the laser cut acrylic stages from rrrf. I did not design it to be compatible because I will not be using the laser cut parts. This might not be as pretty but it sure is cheaper. Each stage is costing me less than $4 in plexi as I am cutting them from scrap I get from a local scrap shop.
For those of you who would like to follow my build which has a complete uptodate list of parts you need to build a modern date McWire you can find it on my web site The Makers Workbench. Please feel free to sign up and post comments or ask questions.
Here is a .jpg of my revision. A much larger version can be found here.
I would like to thank Dr. Bowyer for letting me join in on this discussion.
Thursday, June 11, 2009
While catching up on the blog, I found a really useful post by Tony (copied below):
I was able to get some very nice results from the original GM3 powered BFB extruder, all the same principles apply, keep the melt zone short and support the PTFE close to the heater. Stability of the motor speed and the quality of the motors were the only slight issues. I ran both single and twin motor versions for many months without a problem. I still believe this original extruder design is very good and it certainly proved itself reliable over that time.
For those who had problems getting them to work I strongly suggest you blueprint the design exactly as it should be, by all means sharpen the thread and undercut it with a hack saw if you must but this should be the limit of the mods. The most common error I have heard about is the omission of the small reaction washer behind the main feed screw. Miss this out and you will never grip the filament tight enough. Build the extruder exactly as the instructions and it works, mess with the PTFE add custom heaters and coach bolts for the drive screw(what!), and you are in unknown territory. Run it how it should be first, get some experience in printing, you will then start to see where modifications would pay dividends, the original one works very well and starts reliably from cold.
Coach bolts? Who would do such a thing! :-)
Time to re-check. If it's only me and one or two others having problems with the BfB drive, then it's probably either my setup or incorrect assembly.
Re-state the problem.
Initial problem: heater/motor control worked fine, but extruding at 250C and any speed only resulted in 2-3mm/minute, almost too slow to see. Tried various temps and speeds, same result.
Removing heater, I was able to manually pull the filament through (screw thread not biting).
Tightened all the bolts - still no change to extrusion. Tightened to the point the acrylic started to crack (on the 2mm plate holding the bolt/filament roller). No change. Springs fully compressed. (reaction washer in place)
Disassembled, sharpened bolt thread with die. Better grip, but no better extrusion.
Examining the heater and drive separately, found that the heater required a reasonable force, but extruded OK, but the drive system still wasn't strong enough - although it seemed more than strong enough during testing, it still wouldn't extrude.
I then experimented with some different nozzles and drive systems. I found some nozzle designs needed less force, but needed careful machining to be reliable.
Replacing the m8 thread with a coach bolt , which had a much larger, sharper thread really gripped the filament very well - but still didn't extrude. WTF?
I think the problem is the thread is very grippy - but the pitch is about 4 times larger than the m8 thread. The coach bolt drive is very strong, but drives faster than the m8 thread - a few mm per revolution. I think pushing too fast chokes the extruder, stalling the filament, and stripping the filament (confirmed by examination). It might work very well, but needs to be geared slower (or driven by a stepper).
OK, so after reading the BfB forum and posts, I stripped down my extruder to have a closer look.
A careful inspection of the extruder parts revealed this:
I'll file the channels out, reassemble to the original BfB extruder, and try again. I think it *might* just work this time (famous last words...)
As we say in computing, PICNIC (Problem in chair, not in computer)....or should that be PIANID (Problem in Assembler, Not In Design)...