Question about SVT4-Pro protection circuit & startup delay

[jamesmafyew]

I have an older blue-panel SVT4-Pro here that has an intermittent connection somewhere on the mainboard -- when I pull up gently on the heatsink, the relays disengage the outputs and the fan comes on full-speed. When this happens, the power amps are fine and have no DC offset and still pass signal (visible on the scope just before the relay). When I press down gently on the heatsinks, the fan shuts off and the output relays re-engage. Ok, so there's some cracked solder or a trace under there I have to find.

What I wanted to ask about, though, is the startup time before the relays click on. When it's working properly (i.e. when I'm pushing down on the heatsink), there is almost no delay at all between flipping the power switch and the relays clicking on. Is that normal? They come on fast enough that I can see a bit of DC bump on the speaker outputs on the scope. How long is it supposed to take? I would've figured a few seconds at least. It's been long enough since I had one of these come through that I don't remember how long the startup delay was.

[Sowhat]

What I wanted to ask about, though, is the startup time before the relays click on.

I happen to have one of those on the bench right now and it seems to be about a second for the relays to latch, this is not to say that this amp is working correctly, afterall, it is on the bench.

[jamesmafyew]

Interesting, thanks for checking. I fixed that amp (cracked solder at the ground pin for the LM35 thermal sensor), and it seems fine now, but the relays still click on almost instantaneously after power is applied. Checked all the caps in the protection circuit and they're fine; no idea why it would do that.

[Enzo]

Without me looking at the schematic, I THINK I recall that the main power relay comes on if it detects heater voltage. That should happen quickly. That allows the high voltage supply to turn on. The other relay comes on if the fault sensor is satisfied. No reason for that to take forever.

[nevetslab]

Both of the relays are set up, one per output channel, to sense for DC offset. There's a reference voltage to the (+) side of the comparator circuit from (-15V), thru a 22k resistor and a diode, placing -0.6V as the reference that's common to both relay channels (each fed thru 15k to the (+) input. The DC sensing off the amp output is thru 220k and a 22uF cap, yielding a 4.8 Sec time constant. Then, the two comparator inputs are diode-connected, with the compator output driving the relays' transistor switch. I can't recall the start-up time when all is normal...whether it's instantaneous, or has to time out from the turn-on transient coming out of the power amp circuit. I do know the relay will remain un-energized when there IS a DC offset (from failed output stage parts).

[Enzo]

Sorry, I was thinking of the SVT Classic.

There is more than nevet related, too.

Aside from the DC sensor, each relay also has a control RELAY A or RELAY B. Those are diode ORed to both the thermal sensor and to power up. In the absence of thermal or DC problems, the relays are timed by power up.

IC6 controls this. It sees a roughly 4.7v reference, and as soon as the power supply can charge up C16 to more than that, the relays click.

[jamesmafyew]

Ah ok, then I wonder if maybe C16 is open. I didn't check the caps on the breakout board mounted on the heatsink, and I don't have good experiences with the reliability of bipolar electrolytics in high-heat applications. Thanks.

[Enzo]

It is only one microfarad, shouldn't take all that long to charge up.