nippe.teknopapu.net

While digging through my pile of hardware I found a LCD display made by Epson. I’ve salvaged it from somewhere ages ago and decided to see if it works. After a little bit of google fun I found enough info about it in order to get it running with my Arduino. The display is Epson EA-D20025AR and it is Hitachi HD44780 compatible, but there’s a catch. The connector is only 14 pins and it’s wired differently than you would first think. The small marker on that 14-pin connector that usually marks pin 1 is used to mark pin 2 in this case. I found out that there has been lots of questions about this particular display, so look no further – here’s the pinout for it:

Pinout for EA-D20025AR

(continue reading…)

I did a quick trip to visit some relatives and at the same time picked up two old printers and one scanner. After disassembling them I found out that there’s some nice steppers for my plotter project, though some redesigning of the driver circuitry is needed since the stepper motors from the printers run on 24 volts and my original circuitry is built for 5 volts. There’s also another little thing – two of the motors are bipolar and my circuit is for unipolar motors, but luckily I now have 3 unipolar and 3 bipolar motors so for three axis I should have enough unipolars to do the drive mechanics. There’s also a possibility that I get another scanner next week and get another unipolar stepper motor from it – fingers crossed that it’s exactly the same as the first scanner I disassembled – at least the model number is same :)

Here’s some pictures of the loot.

Disassembled printers

On this picture, there’s two disassembled printers. I was a bit amazed since the printers, a Canon BJ-200EX and Canon BJC-2000 contained almost exactly same circuitry and drive mechanics even though the BJC-2000 is significantly newer product. The main differences were in the driver ICs used to power the stepper motors, BJC-2000  uses only a single chip to drive the bipolar stepper when BJ-200EX uses a combination of two chips. Loot: 2 Mitsumi bipolar 24V stepper motors, 2 Mitsumi unipolar 24V stepper motors, lots of gears and pulleys and some nice motor driver ICs.

Planning..

Checking if the drive mechanics from Canon BJC-2000 would fit on the old Mustek 1200CP flatbed scanner’s mechanics. The idea seems to work but I need more precise drive for the belt in order to get enough precision for the plotter. The goal is around 600 DPI for extra precision, though less will also do at this point. Actually I think that even 100 DPI would be enough for relatively simple shapes.

I was having a slow sunday with a hint of hangover while I remembered that I have an old flatbed scanner stashed in my closet. After some digging I found it and immediately took it apart. Pretty simple design, with a Mitsumi stepper motor. That gave me an idea to make a x-y plotter with it. I’m still hunting for some more parts but should get some old printers this week that will donate their internals to my next creation and then it’s just fabricating and coding.

The motor, Mitsumi M35SP-7N, has 48 steps in full 360 degrees rotation and runs nicely on 5V, though it can’t run very fast. But that doesn’t matter at this point :)

Testing...

After I extracted a driver chip (ULN2003A) from the scanner’s PCB I was ready to hook it up to my Arduino and run some tests. It appears that with the scanner mechanics in place, the scanner bar achieves about 600dpi precision, which should be good enough for my needs. There is a small issue though. Since the bar is driven only from one end, there is slack in the movements of the bar – especially on the end that doesn’t have the drive belt attached. This slack is visible to eye and it will cause problems in the future when there’s weight from another driving mechanism on top of it, so I’m trying to locate a similar scanner (Mustek 1200CP) and take another driving mechanism from it for the other end of the bar.

The testing software I’m using is just a simple implementation of stepper class found in Arduino examples, MotorKnob example to be precise. For the final software, I’m planning on using slightly modified GCode interpreter from RepRap project. The modifications are due to my motor driving circuitry that differs a lot from the RepRap, which has more advanced and better driving capabilities than the setup that I have here at this time.

Stay tuned for future updates on this :)