PCB Heated Bed

Overview Thermistor

Here's my homegrown heated bed. I decided to make my own because prusa's needed a bigger board than I had available.

Resistance measures at 0.9 ohms although my multimeter isn't exactly accurate with stuff like that. Eagle's length-freq-ri ULP also pegs resistance at about 0.9 ohms.

It reaches about 116°C with ambient of 10°C, a felt underlay, and my power supply's 12v rail drooping to about 11.3v despite two 12v 50w lights loading the 5v line.

Here's the eagle BRD file for those interested: Heated Print Bed.brd.

Straightening my wade's idler

Just need to add a proper surface now, and see how it goes!

First Extrusion! Working Axes!

First Extrusion! Very Exciting!

I haven't posted in a while, been waiting for parts, money, ideas to come. Over the past couple of weeks I've made significant progress though. As you can see below, my printer is no longer all wood, and so this blog is probably now rather poorly named, however I'll keep it here until I can think of a better name.

Here's the current state of my frankenprinter:

Output Side
Top View
Electronics
Z Belt Tensioner

All the axes are working, if a shade stiff. Now I need to construct a bed somehow, and I can get printing!

I have bought a piece of blank PCB which I'll etch a heater pattern onto. Here's my current design:

Heater Pattern
Please comment if you think this won't work, or will work well!

Big thanks to Mark Lagana and Greg Frost for printed parts, MrAlvin for the mega (and a ramps board not pictured- I'll use it when I get around to removing my pololus from my current board, or have to replace them), and MakerGear for a lovely hot-end

FiveD on Arduino Firmware: First confirmed print with photo!

First confirmed print

FiveD on Arduino firmware reached an important milestone today- first confirmed print with photo, reported by Markus Amsler here in the forums.

Big thanks to Jake Poznanski, Markus Amsler, Markus Hitter and Stephen Walter who have all provided heaps of patches, to those who have provided just a couple of patches to scratch their personal itch, and also to all those who have simply tried this out and reported how they went :)

For those of you just viewing this now, forum, source, wiki and I can occasionally be caught on the reprap IRC channel too.

Re-orienting towards Wooden Prusa Mendel

After a bit of a sabbatical from constructing to concentrate on firmware, I'm now at the point where I need my printer finished to make any more meaningful progress. In that time I have discovered the brilliantly simplified Prusa Mendel remix, which I think will be significantly faster and easier to make from wood. To that end, here's my first print for a long while, the prusa x-end-motor part:


Prusa x-end-motor

FiveD on Arduino - a call for testers

A while back I started work on an alternative firmware, FiveD_on_Arduino [forum thread][source][wiki]. It is designed to fit on an atmega168 (run your whole reprap from the extruder controller!) yet be upwards compatible with other chips (328p, 644p, 1280 so far) and extensible.

This firmware has reached the point where it needs wider testing. You'll need a system that has gnu make, avr-libc, avr-gcc and friends to compile as I have no idea how to wrangle the various IDEs into doing what I want. You'll also need avrdude, or some other way of getting the compiled firmware onto your atmega.

git clone git://github.com/triffid/FiveD_on_Arduino.git
git branch -r
git checkout release-candidate-triffid
cp config.h.dist config.h
### edit config.h and Makefile in your favourite editor ###
make
make program

If you have ideas on how to make this firmware easier to access, please post in the forum thread ("compile and load in my favourite IDE" a given- but what needs to change?). Other ideas and contributions are welcome too :)

Thanks to traumflug for his numerous contributions, jakepoz for gen3 electronics support, and numerous others who helped with suggestions and patches.

RepRap Motherboard, Homebrew Style

Homebrew Motherboard

Here's my homebrew motherboard freshly kitted out.

The arduino sits in a reverse shield of my own design, with pin sockets either side and bridges going from one line of sockets to the other. Hard to see in the photos are some wire clips that hold onto the mounting holes to help keep it steady. It's surprisingly snug with this arrangement, barely moving even when I connect the usb plug.

The ATX connector is made simply from some tinned copper wire. I'll probably put in a proper connector when one becomes available. The 12v connector in the middle of the board is vestigial, and will be removed when I add some pins for the ATX 12v extension. Multiple 12v and ground wires are absolutely essential, considering how much current will be drawn from them, as well as the likelihood of vibration.

The jumpers near the motor controllers are connected to MS1-3 so I can change the microstep amount simply and quickly. Z needs none, extruder needs it all and I'm still undecided on how much X and Y need.

At the bottom of the board next to my MAX6675 are a pair of power mosfets for controlling the extruder heater and one other item- bed heater or fan or something. The mosfets are IRL3803s which according to my calculations can handle 16A with no heatsink and 4.5v on the gate.

The headers between the arduino and the atx connector are for the end-stops which I'm yet to make.

Bottom View

The bottom view shows a rat's nest of point to point wiring which I'll clean up if it proves to be problematic. Signal lines next to high frequency high current PWM lines are usually a recipe for disaster, but the motor modules are so small that I'll just have to see how I go. Twisting and decoupling helps immensely, but is by no means a magic pill.

Wooden X 360 Z Assembly

End View
Back
Side Plate
Front

Here's my X carriage 360 end assembled. Despite the significant errors made in these pieces, they fit together surprisingly well. I haven't fitted the adjustment bolts as the wings hold on quite tight enough without them. I think this is the first bearing runner I've made to date that doesn't try to rotate around the bar while moving along it- all my errors must cancel each other out, or this particular section can deal with gross imperfections better than any of the others I've made.

In other news, I found some linear acceleration stuff to put in my firmware and I seem to have most of the bugs out of it.

The acceleration code ganked from the official FiveD firmware acclerates by a set amount each step but steps get closer together, so acceleration is exponential and deceleration is logarithmic, making really high speeds impossible to attain. The above mathematics allow a constant acceleration profile with only a little integer math each step, which is ideal for us.

It needed some adaption because it was created to move a fixed distance in a certain period of time rather than change from one speed to another while moving a fixed distance, but everything needed to adapt it is there.