Acoustic 370 turn on thump

[Randall]

Not sure where to go from here. I forced the issue with the customer, and when pressed he could not really 100% remember if there was ever a time when this amp did not do this. However, he said he recently sold another one of the same model, and that one didn't do it. He also said he would not be opposed to a standby switch, so if that ends up being the way this goes, could someone explain in a little more detail how that switch and resistor get wired?

[dmeek]

Another idea would be to insert a surge thermister in series with the power transformer primary. I'm not sure how to choose the best one. Maybe someone else here knows more about it

https://music-electronics-forum.com/...1&d=1587513261

[dmeek]

Here's how the standby switch would be wired.

https://music-electronics-forum.com/...1&d=1587517960

[Leo_Gnardo]

OK, this is interesting, I grounded the top of Q315, the mute JFET, and switched it on. Still has the thump. And this is with the preamp output cap still lifted.

That DC blocking cap on the output still has to get charged. Then discharged on power down. Thump, thump, anybody in there?

[olddawg]

Have you swapped out C313?

[Randall]

"That DC blocking cap on the output still has to get charged. Then discharged on power down. Thump, thump, anybody in there?"

Yes, but as I mentioned, I subbed in a 2200uF, and it made no difference.

"Have you swapped out C313? "

If you mean the .1Uf near the output, no I haven't. How could this be a factor?

[Leo_Gnardo]

"That DC blocking cap on the output still has to get charged. Then discharged on power down. Thump, thump, anybody in there?"

Yes, but as I mentioned, I subbed in a 2200uF, and it made no difference.

I recommended swapping in a new cap on the output as insurance against having the old cap failing as a short circuit and putting half the rail voltage on the speaker. Good, that's done.

Every time the amp starts up, the voltage on the amp side of that cap has to rise to its operating level, half of the rail voltage. At that instant, the amp will pass a pulse of current into the speaker. Thump.

Every time the amp is switched off the voltage on the amp side of that cap will fall from half the rail voltage to zero, passing a pulse of current into the speaker. Thump.

There's no escaping it. Unless you disconnect the speaker until the current pulse has passed - the standby switch as shown will let you do that.

[Randall]

I clipped in a 1K 6 watter as per the above diagram, thanks dmeek. It still thumps, only this time when the standby switch is closed. It does not do it if I allow for the output cap to fully charge to 45v, but this takes a full 20 seconds. Is this normal, or could the cap be at fault? I am leaning towards replacing it as Leo has recommended.

[dmeek]

The resistance will have to be less. It gets complicated. You could try 470 ohms which should reduce the time to 10 seconds.

I was going to suggest this circuit but the resistor would burn out with any signal while in standby

https://music-electronics-forum.com/...1&d=1587591492

[Randall]

What is preventing a 5w resistor from burning up with any signal in standby in the first drawing with the 1k resistor? Isn't it potentially taking all the output power?

[Leo_Gnardo]

The resistance will have to be less. It gets complicated. You could try 470 ohms which should reduce the time to 10 seconds.

I was going to suggest this circuit but the resistor would burn out with any signal while in standby

https://music-electronics-forum.com/...1&d=1587591492

I like the latest iteration, but use a 1K or 470 R instead of 10 ohms. SO WHAT 20 seconds, or 10, if the amp user is so impatient they can't wait less than half a minute let 'em suffer the thump. Learn, or let the amp abuse the speaker at every power up & power off. Sure it's not a great thing to do to the speaker but most of 'em have survived in spite of getting whacked by that current jolt. Hint: Don't try it on a JBL D, K or E 130. They're really sensitive to this kind of thing. Expensive to recone too.

[Helmholtz]

What is preventing a 5w resistor from burning up with any signal in standby in the first drawing with the 1k resistor? Isn't it potentially taking all the output power?

The signal voltage at full output is (a little less than) 45V x 0.7 = 31.5Vrms. So a 1k resistor would have to dissipate (31.5V^2)/1k = 0.99W. This is without clipping. With full clipping (square wave), power will double.

[Enzo]

power will double.

And even that won't burn out a 5W resistor.

As to the thump, as Helmz said, 30v output. After that you run out of power supply. So the thump doesn't exceed those power rails. SO the thump is no more stress to the speaker than the full output of the amp is. If you can slap and jam on your bass for four sets without killing your speaker, that thump won't hurt it.

[Randall]

So if I understand this, using a 470R 5w resistor on a SPDT switch as in post #24 would give me about 10 seconds to charge the cap, and 4.2 watts full power clipped signal into a 5 watt resistor, worse case. Of course if someone were to drive the amp at full clipped power, they are one their own anyways, as far as I am concerned.

Customer does want to spend an extra $25 to replace the output cap for safety's sake, so that is ordered. Does anyone see any problem using the ground switch for the standby? 50 year old cap in there after all.

[Leo_Gnardo]

The signal voltage at full output is (a little less than) 45V x 0.7 = 31.5Vrms. So a 1k resistor would have to dissipate (31.5V^2)/1k = 0.99W. This is without clipping. With full clipping (square wave), power will double.

So a 5W resistor leaves plenty of power headroom, not much chance it would ever burn out.

Note the most recent drawing shows a 10 ohm resistor. If that value is used, and somebody plays the amp in standby while the volume's turned up, sure that's going to generate some heat. But, as I said, it's not necessary to go there.