DC output protection circuitry

[priscared]

Gday
I have an old power amp, it doesnt have dc protection circuit on the output. Does anyone have a design for one of these circuits and suggestions on how about putting it in. I have a yamha p2200 power amp.
Thanks
Daniel

[traderbam]

Hi,
Not sure what sort of thing you are looking for. Here's a bullet-proof circuit that will protect your speakers against dc offsets of 0.6V or more. It has a detection circuit that triggers an opto- thyristor and then disconnects the speaker via a relay. It keeps the speaker disconnected until the dc offset is gone AND you have reset the circuit via a push switch. You need to access the speaker outputs and the + and - supply rails and ground.

If this is the sort of thing you want let me know and I'll explain it in detail.

[priscared]

that is exactly what i want. a passive device would probably be better as i could put it in my speakers as my amp is tight for space (yamaha p2200) but yeah if you could give me some details thatd be great
Daniel

[EchoWars]

No passive device is going to be able to detect DC and disconnect the speakers...at least that I know of.

Elliot Sound has a PC board (it's very small) that simply needs population and a relay (and the wiring to the board) that does the job in about as small a space as possible.

http://sound.westhost.com/project33.htm

Also..here is a design that I've used a few times...but if space is a concern I'd consider the ESP board.

[wintermute]

Hi,

you may want to consider some polyswitches in addition to the dc protection circuit. They are current tripping devices so protect against over power situations, which a DC protection circuit won't....

Link here: http://www1.jaycar.com.au/productVie...Max=&SUBCATID=

Note Jaycar have a much better explanation about choosing them in their paper catalog.... I prefer to put one on each driver rather than just on the speaker input.

Tony.

[richie00boy]

Not exactly a passive single device, but there is a Velleman kit (K4701) that uses a relay and a few diodes and resistors to disconnect a load in the event of DC. The thing requires no power supply and so can live in speaker cabs. Quite a small PCB, too.

[EchoWars]

Hmmmm...I thought that would be cool...until I saw that it needs a DC voltage of 10V before the relay will operate. Granted, smoked output devices will usually present a lot more than 10V, but for my own speaks I'd like something that'll kick in at a volt or two.

[traderbam]

Daniel,
Here's the circuit explanation. The circuit is designed for an amp with split rail supplies of about +/-36V. Component values can be adjusted for other rail voltages.

The left half of the diagram is the dc detector. The speaker signal is fed through a 10k resistor into a 50uF non-polarised cap. These set the time constant for the trigger. The transistors should have betas >100 and Vce >rail voltage. The two 2.7k resistors set the "on" current through the LED of an opto-SCR.

The right hand portion controls the speaker relay. When the amp is powered-up, the relay is initially open-circuit (speaker disconnected). The 2200uF cap charges up throught the 2.7k resistor and then the relay is activated, driven by the transistor. This provides a few seconds of turn-on delay thus mitigating any turn-on thump from the amp. The cap stops charging when the 24V zener holds the voltage at the transistor's base. This circuit uses a 24V relay.

When dc is detected the opto-SCR is triggered and the SCR, aka thyristor, switches on. It remains on until no current flows through it. It pulls the transistor base to ground and turns off the relay. The circuit will remain in this state even if the dc condition goes away. The only way to "reset" the thyristor is to short-circuit it using the push-switch, or by powering down the amp for a few seconds. If the dc condition still exists, pushing the switch will have no effect and the speaker will remain disconnected. (The cap also slugs the rate of voltage rise across the thyristor to prevent it from self-triggering due to junction capacitance).

The 10k resistor is a bleed for the cap. The 1N4001 diode shorts reverse currents from the relay coil. The 0.01uF cap prevents arcing at the relay contacts when the speaker is disconnected. That's about it - the 2.7k resistors should have about 1W power rating. For transistors I used ZTX653/753 but any similar transistors will do. Use a good quality relay to preserve sound quality - mercury contacts or whatever - and ensure it can handle the speaker currents, a 10A relay normally suffices.

I can assist with adjusting component values to suit your amps power supply rails, just ask. And if the diagram isn't clear enough I'll send you a new one.

[traderbam]

The trouble with a passive system is that relay power is needed. A really simple circuit could be made with a "normally closed" relay. Just put the relay coil in series with a resistor and connect across the speaker outputs. A cap across the coil to slug the response a little. The line to the speaker goes through the n.c. relay contacts. If a dc level appears that is high enough to drive the relay it will disconnect the speaker. This is belt and braces and may save your cones from the most common failure which is the output going short to one of the rails.

[EchoWars]

If a dc level appears that is high enough to drive the relay it will disconnect the speaker.

Which is exactly what the Velleman kit does.