Embroidery machine retrofit.

Unfortunately, after looking closer at your pictures of the board neither the 7200M nor the 7200E pinouts match that of the 7200J. The GND connections are wrong.

Another thing to note is that both the 7200M and the 7200E are low level drivers - they don't have step/dir inputs but rather phase excitation inputs. Assuming 7200J is the same I can see two scenarios: 1) There is additional logic on the board to drive these from step/dir inputs on the Centronics input. This will complicate reverse engineering. Or 2) The phase inputs are driven directly from the Centronics connector. If so I'm not sure how feasible it is to drive the thing from LinuxCNC.

Well I am still thinking my best route is to scrap list on ebay the current board. Maybe somebody will want it. Then just get a couple cheap stepper drivers and an lpt breakout board. It will save me a fair bit of headache over trying to reverse engineer this. 

I have another larger one I still need to figure out as well. It is self contained but not sure what I will do there as software seems hard to find and hardware key locked. But thats a project for later.

So dug back into it a bit more. The white box on top looks to be an scr motor controller. The motor has 4 wires going to it. 2 about halfway down (blue and red 16 ohm) and then  at the back (white and black around 8k ohm)

It has a 3 wire control coming from the bottom. Tried to measure any power on the pins but did not see anything.

The 120v also goes into a transformer and then goes directly to the bottom board. So any dc conversion is happening there or possibly in the missing controller.

I think i do wanna keep the scr controller if possible 

Birdman3131 wrote:

Well I am still thinking my best route is to scrap list on ebay the current board. Maybe somebody will want it. Then just get a couple cheap stepper drivers and an lpt breakout board. It will save me a fair bit of headache over trying to reverse engineer this.

This is probably a good plan IMO. But first make sure the steppers are two-phase (unipolar or bipolar). If they are e.g. 5-phase drivers are more difficult to find I think. How many wires are coming from them?

About the SCR controller: it look like the 3 wire control interface goes to an isolation transformer on the SCR card. Nice, but it means the control signals are not simple analog DC. They must be some kind of AC or PWM? It also looks like there might be a feedback signal going back to the control board. You probably need an oscilloscope to figure this one out.

Not sure if I need to try and get this thread moved to one of the build forums. As at this point I have started ordering parts.

The steppers have 4 wires each so I assume they are the more common type.

I messed up slightly in ordering the parallel port breakout board. The ones I have used in the past had jumpers for if a set of pins were inputs or outputs. It looks like the one I bought does not have that and I am thus stuck with only 5 inputs when I need 7.

Current options I have for getting extra inputs. Not actually sure some are viable.
1. Use an ESP8266. Connected either via serial or usb (I assume wifi would be less than ideal)
2. Spend the extra money on a lpt breakout board that is toggleable.
3. I have 4 serial ports on the computer. Not sure if they could be made to work but I figured I would ask.
4. Combine X and Y homing into one pin as well as upper thread and lower thread into another. Likely my best bet after #2 but not sure.

The sewing motor's extra set of wires does have me confused. I do have an oscilloscope and will see if I get any output when I spin it. They are 8k Ohm and I am assuming at this point they are either a position reading or possibly a brake of some sort.

Only wires going from the control board is the small 3 wires going to the isolation transformers. The other connections are in/out 120v.

arvidb wrote:

Or 2) The phase inputs are driven directly from the Centronics connector. If so I'm not sure how feasible it is to drive the thing from LinuxCNC.

LinuxCNC supports a variety of direct-excitation schemes, both with the parallel port stepgen and the Hostmot2 version.