[Capnden]

I have used a Chinese TB6560 control board for about 3 years. I made all the fixes in one form or another and the board performed fairly well except when I was running very long programs for etching PC boards when it would miss a few steps and lose registration. I had a lot of confidence in Toshiba, I don't believe they would keep putting out a chip that didn't work so I looked first at the outputs from the parallel port. These looked good and clean. I then looked at the inputs to the 6560. Here is where the problem was. The signals were noisy and the low was right about 0.8 volts which is the max spec low for the chip. I looked at the board schematic and the whole buffer circuitry is backwards. Opto isolators should have come first and because they degrade the signal the 74HC14 should have been after the isolator to clean the signal up again. I started looking at why the signal needs to be buffered in the first place and decided it didn't. The output from the parallel port uses 5V logic and so does the 6560. There is no need for a buffer.
I removed the 74HC14s and the 9 opto isolators for the XYZ signals and removed the IC sockets they were in. I then wired jumpers from the input pads where the 74HC14s came out directly to the output pads where the isolators were effectively connecting the parallel port directly to the 6560 for the signal lines. Presto everything worked great. I have run several tests and made several PC boards since without any indication of missing steps.
If anyone else is having trouble with this board I suggest you give this a try.

[pudzian]

I have used a Chinese TB6560 control board for about 3 years. I made all the fixes in one form or another and the board performed fairly well except when I was running very long programs for etching PC boards when it would miss a few steps and lose registration. I had a lot of confidence in Toshiba, I don't believe they would keep putting out a chip that didn't work so I looked first at the outputs from the parallel port. These looked good and clean. I then looked at the inputs to the 6560. Here is where the problem was. The signals were noisy and the low was right about 0.8 volts which is the max spec low for the chip. I looked at the board schematic and the whole buffer circuitry is backwards. Opto isolators should have come first and because they degrade the signal the 74HC14 should have been after the isolator to clean the signal up again. I started looking at why the signal needs to be buffered in the first place and decided it didn't. The output from the parallel port uses 5V logic and so does the 6560. There is no need for a buffer.
I removed the 74HC14s and the 9 opto isolators for the XYZ signals and removed the IC sockets they were in. I then wired jumpers from the input pads where the 74HC14s came out directly to the output pads where the isolators were effectively connecting the parallel port directly to the 6560 for the signal lines. Presto everything worked great. I have run several tests and made several PC boards since without any indication of missing steps.
If anyone else is having trouble with this board I suggest you give this a try.

I was writing my post almost at the same time that you posted this so I didn't see it at that time.

I would like to try your approach first since it does not involve to (de)solder anything and is easy to rollback. Could you please have a look at my images and tell me exactly how to place the jumper wires?

Thanks!