How to solve loud thump at power off?

[Craig Sawyers]

I suppose that a sound reinforcement application has really robust loudspeakers that can tolerate a full-power transient on power up or down.

But what they should include is DC output protection, because no loudspeaker can survive an output transistor going into second breakdown and connecting a loudspeaker to the DC rail.

But like Kevin says - this is rarely the case.

Craig

[dougietamson]

I've tried tackling the issue from the mains end.

Back to basics this morning:
1st thing I tried was to try a simple pull the plug out of the wall socket while powered up, usual crack/thump.

Next try, switch off at the wall socket while powered up - no thump - we're getting somewhere.

I tried an 6 way mains extension lead with a built-in neon power on/off switch and switched off using the built-in switch while powered up - no thump.

Doug

[dougietamson]

I also tried a different mains switch, I pulled this one from from a Carlsbro 150W PA, dual pole with indicator light, same crack/thump.

[Craig Sawyers]

Well that is interesting. Switching off at the wall socket or extension gives no thump, but using an internal switch gives a thump.

There are two types of switched socket. Some switch both live and neutral, and some switch just live, and neutral (and ground) are just looped through.

So maybe interrupting just live might be giving different switch-off behaviour as compared to interrupting live and neutral with a double pole switch.

Just speculation at this stage.

Craig

[Boulevardier]

Another possibility - if Doug did those switching tests in fairly rapid succession, capacitors may have retained charge in the brief intervening time, and not had to recharge/discharge at the next switch-on/switch-off....?
Mike

[dougietamson]

Originally Posted by Boulevardier

Another possibility - if Doug did those switching tests in fairly rapid succession, capacitors may have retained charge in the brief intervening time, and not had to recharge/discharge at the next switch-on/switch-off....?
Mike

I repeated the tests a few times and waited a while between tests as I did notice that the reverb spring did still make some sound after the power off.

The 'good' switches did make a small thump at power on, same as the 'bad' ones.

I'll have a look inside the 6 way socket lead and try another multi socket that has a switch on each socket.

Doug

[John_BS]

If you could post a schematic of the amp we may be able to modify the DC time-constants to reduce the bump.

John

[dougietamson]

Originally Posted by John_BS

If you could post a schematic of the amp we may be able to modify the DC time-constants to reduce the bump.

John

Sorry, I don't have a schematic, I've tried googling but nothing.

[Silicon]

Could the mechanical 'snap' of the mains switch be affecting the reverb spring and giving a thump before the DC power decays?

Is the reverb spring assembly mounted on rubber to isolate it from the chassis?

[dougietamson]

Originally Posted by Silicon

Could the mechanical 'snap' of the mains switch be affecting the reverb spring and giving a thump before the DC power decays?

Is the reverb spring assembly mounted on rubber to isolate it from the chassis?

The reverb tank has rubber grommets and is attached to the 'wood' cabinet.

The on/off/on mains switch is at the rear of the chassis, I'm going to fit an IEC 3 pin socket and fit a switch that doesn't crack/thump.

[peter_sol]

yes. switching the live and neutral does give a crack as there is a double spark, not always in sync. take record decks, they always have single switches and rarely cause thumps even without a suppressor cap.
I suspect it does no harm to the speaker.

[Craig Sawyers]

It depends so much on the design. The Quad 33/303 has absolutely no start up and switch off thumps, even though the preamp and power amp are single rail and the power amp has an output capacitor.

Craig

[Radio Wrangler]

The 303 has a regulated power supply, could they have controlled its turn-on dv/dt so that the current in the amplifier output blocking capacitor is controlled?
I've not got the circuit in front of me, but as there has to be a certain number of coulombs put into/out of that capacitor, the only way to do it silently is slowly.

David

[Craig Sawyers]

According to Baxandall, who designed much of the 303, the only reason they used a regulated supply was that they were all paranoid about blowing up power transistors. This was their first foray into transistor power amps, so they decided to be ultra cautious.

Of course the audio fraternity ascribe the use of a regulated supply to some mystical sonic benefit. Nothing could be further from the truth.

I have no idea why the 33 turns on quietly by itself, or the 303. Although Quad never listened to their designs, contenting themselves with competent design and measurements, they got so much right.

Craig

[Radio Wrangler]

But on having a regulated PSU from concerns about transistor voltage ratings, then it's almost cost free to pick the turn on speed as a bonus. Baxandall would have spotted the opportunity and the result was a well house-trained amplifier.

David

[Aub]

Originally Posted by Craig Sawyers

Of course the audio fraternity ascribe the use of a regulated supply to some mystical sonic benefit. Nothing could be further from the truth.


Craig

Isn't a regulated supply supposed to stop the supply sagging on peaks Craig? Or is that rubbish?

Cheers

Aub

[Boulevardier]

Originally Posted by Craig Sawyers

According to Baxandall, who designed much of the 303, the only reason they used a regulated supply was that they were all paranoid about blowing up power transistors. This was their first foray into transistor power amps, so they decided to be ultra cautious.
Craig

D Self seems to confirm this in one of his books/articles (can't remember where). Self seems quite incensed about the use of a supply regulator in the 303 (which he claimed was completely unnecessary), and reports that when he challenged Peter Walker on it, Walker gave the same explanation as Baxandall's above.

Mike

[Radio Wrangler]

They were taking a bit of a step into the unknown, so they played it careful. What size surges you can get on the mains is unspecified. How much voltage rating you need in hand to be reliable in the field is unknown, so they played it careful. It worked and was probably the first 'good' trnsistor amplifier. They (and others) learned from it.

David

[Cruisin Marine]

Why not add a simple push button switch to a simple delay cct. that disconnects the audio/ and or speakers - that would be a simple and effective answer.

[dougietamson]

Update: I've fitted a new power switch and the load bang/crack sound at power off has gone

Doug