Outoput relay desing concept

[d3imlay]

I've been toying with some ideas for an output relay for DC protection.

Given a 200 watt amp with 80 VDC rails and assuming that the ouput goes to one of the rails, not many relays are capable of interrupting that kind of fault. AC no problem, DC big problem. The second function of the relay would be a delay on pickup when the amp is turned on.

Two ideas I'm considering are:

1. Crowbar the ouput with a triac like Peavey did on the CS800. A simple R/C network drives the gate. A relay is still used but it's main function is delay on pickup.

2. Current sensing rather then voltage sensing. Digikey sells small current transformers that could quickly sense fault currents, hopefully without the limitations of voltage sensing.

Has anybody ever looked at an output stage with a scope when an amp fails? Usually there is at least one failed output in each rail. Does one xistor fail, then cause the weakest xistor on the opposite rail to fail due to exceeding the SOA? If so, how long does it take for the second xistor to fail? I have a sacrificial amp that I can test some ideas. My plan would be to remove all of the ouputs except 1 on the positive rail and run it at some level until it fails. I've already installed a CT on the ouput to trigger a scope. I'd like to put it in the +DC rail but I don't have room

[Leolabs]

I did saw some with optocoupler.Try this:

http://www.schematicsforfree.com/archive/dir/Audio

[djk]

"1. Crowbar the ouput with a triac like Peavey did on the CS800. A simple R/C network drives the gate. A relay is still used but it's main function is delay on pickup."

Copy the Peavy schematic, it's also the same one that Quad uses in several of their products. Don't forget the SBS on the triac gate.

Also don't forget relay after the crowbar. BGW did this, as did Peavy did this on the newer versions of the CS800.

http://www.diyaudio.com/forums/attac...amp=1137409831

Protection Circuit - Crowbar - Zener vs SBS