These are some very high-level instructions for building the Opto-Isolated Stepper Motor Controller. Not really a lot to say. Most of the construction is self-explanatory using the PCB artwork, the overlay, and the schematic. Note that the pad for pin 1 on ICs and such are square to help in proper placement. All ICs on the board are NOT oriented the same way, so pay attention.

 

Download in ps

info - General info about the controller.

parts.txt - This is an ASCII file with the parts list and some general info. This includes supplier info.

schmatic.ps - Schematic in Postscript format. This is a tad rough because I used an HPGL to Postscript converter to make it since Protel's Schplot doesn't do Postscript output.

overlay.ps - Overlay in Postscript format

pcb.ps - PCB artwork in Postscript form. This in a "normal" view. That is, it isn't mirrored or flipped. This is the view you want to use for iron-on methods.

test[123].exe - Test programs for parallel ports 1, 2, and 3. These programs step motor 1 back and forth.

step.[ch] - Source code for stepping motors. Really very simple. It is written for Borland's Turbo C. Porting to another compiler shouldn't be too much trouble.

If you're wire-wrapping, go by the schematic (you're on your own).

If you're using a PCB:

Using the overlay and schematic for direction, solder the parts in place on the PCB. Resistors, diodes, and axial capacitors are mounted "stand-up" fashion. MAKE SURE YOU CLIP U2's HEAT SINK ON BEFORE SOLDERING IT IN PLACE. It is sometimes impossible to get on afterwards.

Note that you have two choices for the 5V supply. It can be taken from motor supply via the provided 78L05. I have used this with no problems at motor supply voltages from 8 to 16V. The UCN5804B draws a very small amount of current. You can also provide your own 5V to the spare pad on the control input side. Note that even though this is next to the ground input from the PC, it is isolated from that ground (i.e. you can't take 5V from the PC). Connecting the two grounds would compromise the isolation.

I've used two different methods for cabling. The cheapest and simplest is to hack the Centronics connector off of a printer cable and use a continuity tester to determine which wires go to which pins. If you're using the kit, the cable used comes from MECI and has the following connections.

They are listed as color1-color2 where color1 is the wire color and color2 is the stripe color:

WHT-BLU (2) (direction)
ORG-WHT (3) (step)
WHT-ORG (4) (half/full)
GRN-WHT (5) (motor 1 on)
WHT-GRN (6) (motor 2 on)
BRN-WHT (7) (motor 3 on)
WHT-BRN (8) (motor 4 on)
GRY-WHT (9) (spare)
GRN-BLK (25) (ground)

The ground conection to pin 25 usually works fine. On some PCs you may have to connect pins 18 through 25 together.

The second method for cabling is to use a short pigtail connected to the board with a DB-25 connector on the end. This allows you to use a straight-through cable to connect the Controller to your PC parallel port.

Once you have it completed, connect a motor. The 6 outputs on the motor side are as follows (viewed from component side with motor connectors to the left):

B
V+
D
C
V+
A

B-D go to one motor winding and A-C to the other. Refer to the schematic and any docs on your motor for details. Apply power and check for 5VDC  on the output lead of the 78L05 (lead farthest from power connector).

Three simple test programs (test1.exe, ...) are included to help check the operation of the controller. These are all the same except for the parallel port they use for output. Test1.exe uses lpt1, etc. If you  know which port you're connected to (you may be surprised) then run the appropriate test program. The motor should simply rotate back and forth in each direction. How far it goes depends on the step angle of the motor.

If it doesn't work check component placement, cabling, etc. I'm assuming  some basic electronics building experience here (debug mode on ;).

If you are stuck, send me e-mail at wrb@cbnews.att.com and I will help as much as I can. Good luck and have fun!