Parallel Port Opto Isolation ???

[GrahamIT]

Any electronic designer types here that can point me in the right direction for information to do with using opto-isolators with a parallel port, specifically for Mach1 and 2 software.

I'm trying to design my own breakout board and so far I think I'm doing a reasonable job, but really need this feature so as not to fry my works laptop, only I've not used optos before.

[sendkeys]

i remember seeing a schematic at embeddedtronics.com dont know how great it is.

[GrahamIT]

Didn't look too promising, I'm in need of a little more detail really, thanks for trying though.

[fyffe555]

Graham,

I've been playing around the same subject for a while (longer than I'll admit here ok so i'm hoping to learn something too) and here's a few things I think I've figured out. Hopefully people more in the know will be along shortly.

Most of the schematics available on the web, while they might look pretty, have nice colours etc, do have a few problems and I've not seen one I like. Given that an opto basically lights an led one side and so switches a gate on the other, the isolation is provided by the lack of electrical connectivity across the gate. Supposedly. The problems with some schematics I've seen is that even though they're using opto's they sometimes share a common gnd or 5v suppy to power both sides of the opto, or take the PC's PP 18-25 gnd pins right through to the drivers gnd.

If you're trusting the opto to break the electrical path then on each side you need a seperate gnd/5v supply on both sides as well as the connection to your input or output (lpt etc) to either take the state high or low depending on your need ( technically lousy english but hey).

You also need to consider the opto's characteristics for switching speed, but also V+ and ma necessary to light the led. Some might work directly from the parallel port, others might need a seperate 5v supply and you use the ports state to take the optos led high or low.

You also need to consider direction, some pins on the PP are bidirectional, some are in, some are out. Some sw uses pin 1 for input to an axis for example. Once you know which is which then what are you going to isolate? Everything? One way or bidirectional? Just inputs to the PP? Or outputs for Step and direction too? Outputs to relays?

If your drivers have opto's there's arguably no point in isolating 'upstream' of that. I wonder if the added opto might added switching time to the whole thing and so *might* cause some problems if you're at higher switchrates. Don't know if this is true or not, just a thought.

On the PC side you'd have to use power from the PC side to power the opto gate/led. On the driver side you need to power the gate/led from the driver side. Watch out when you're laying out the pullup/pulldown resistors to work out the switch state, one layout might take a PP high state and opto switch output result in a low. Watch out when laying out opto's with gates working in different directions.

My efforts so far with eagle haven't resulted in a board I would like to make! too complicated just opto isolating PP inputs and outputs to relay switches. The drivers I want to move to have optos so axis output opto in a breakout is unneccessary. I think.........

I'm still working mine out, have just sent a PM to someone here on the same subject in fact, so I'll be interested to see where it goes.

[HillBilly]

I've had good luck with this circuit.

Darek

[pminmo]

Thats an excellent circuit. The only issue might be speed, don't run step rates any higher than 10khz.
Unless you goto a faster opto isolator.
Phil

[GrahamIT]

That's more like it, though I thought opto's were rated in MHz not Khz range.

Thanks Hillbilly, gives me a good starting point to work from.

[pminmo]

The typical propagation delay of the mct opto isolator is 50us on it's worst transition. Depending on how well they match up with each other, you maybe able to run faster. You will only know when you get your setup running. A couple of things you might consider is going to the 6N135 opto isolator for increased speed, you might also consider using a uln200x series driver to replace the 7407.

Phil

[GrahamIT]

Do I need to worry about propagation delay in this instance, as it would be a constant and not have any effect on the motors performance unless the delay became longer than the step rate. I dont think this is going to happen as I'm only using full and half step drivers, maybe it would become a problem if I were to use microsteps though.

[pminmo]

If each one was matched, but in reality they will vary and be within spec.