DMX Optical Isolation

[mschell]

Great stuff guys. I appreciate all the feedback on this topic. I sent Mark's comment to the seller and here's his reply ..

"
Remember that the splicer's main function is to amplify the DMX signal in order to drive more DMX devices. "

And yet the seller calls it an 'isolator'. So which is it - a repeater or an isolation device?

[David_AVD]

The reason for isolation can be either A) to eliminate excessive ground loop currents (via the common reference wire) in systems spanning larger areas, B) to protect other segments from damage due to excessive voltages getting into one segment or C) both.

As for the diodes isolating the two sections, I don't see how that can be true. A quick bit of drawing shows that a large reverse polarity (or AC) spike onto the DC side of one bridge will first burn one or more of the diodes, then more than likely kill the second bridge. How much damage to the transformer secondary and DC side of the other bridge will depend on the surge. I'm happy to be proven wrong of course, but I don't think I am.

[smartalec]

this is one of the diagrams i was basing my original circuit from http://dl.dropbox.com/u/48217835/dmx...20isulated.pdf
ive got my circuit running better now, just finishing up on bits.

[P. Short]

The reason for isolation can be either A) to eliminate excessive ground loop currents (via the common reference wire) in systems spanning larger areas, B) to protect other segments from damage due to excessive voltages getting into one segment or C) both.

As for the diodes isolating the two sections, I don't see how that can be true. A quick bit of drawing shows that a large reverse polarity (or AC) spike onto the DC side of one bridge will first burn one or more of the diodes, then more than likely kill the second bridge. How much damage to the transformer secondary and DC side of the other bridge will depend on the surge. I'm happy to be proven wrong of course, but I don't think I am.

At this point I think that you're right, and that I was writing nonsense before. I think that the 'isolation' between the two sections is limited to two diode forward-drops plus the breakdown voltage of the regulator (or filter cap, whichever is smaller). Once that voltage is exceeded largely uncontrolled currents will start to flow, and various components will be damaged. If I'm right, the isolation between the two sides would be well under 100V, possibly under 50V, which is way too small.

[Jack Stevens]

Great stuff guys. I appreciate all the feedback on this topic. I sent Mark's comment to the seller and here's his reply ..

" The bridges do not provide a complete electric isolation but will be able to prevent a voltage feed back from any device equal to the 600 V rating of the diodes more then what is usually needed for protecting the controller.
Remember that the splicer's main function is to amplify the DMX signal in order to drive more DMX devices. "

And yet the seller calls it an 'isolator'. So which is it - a repeater or an isolation device?

Sales puffery. It's a repeater, not an isolator. Denny, it won't do what you're wanting to do. That 3-in-1 device will, though. It has isolated power supplies and opto-isolators for each section. I'd do it with RJ-45's, though, for your/my purposes. You could be by with doing just one isolator, rather than 3, if your universes don't require a lot of controllers.

[mschell]

Jack,
Yes, I was being somewhat sarcastic in my question - it's a repeater, not an isolator.

I'm thinking that something else is needed - both a RS485 output from the E681 and then an isolation device to protect the output stage.