anti-thump circuit

Status
Not open for further replies.
Bill Fitzpatrick said:
I agree but don't believe that a cap across the power switch is going to solve his problem.
I've run into this problem on a bunch of amps, and I have to agree with Bill here. A cap on the power switch won't do much except prevent arcing on the switch.

There are dozens of circuits around to prevent turn on and turn off thumps, and the better ones also will protect the speakers from a catastrophic amp fault. Paulb posted the most elegant and easy solution already.
 
First, the person who did your surrounds did a poor job if it is tearing after 3 months. The material does rot out after 10-20 years, but tears this early in life indicate a bad job or old surrouds were used. But a turn on thump is always annoying, and I try to design them out of the amps I build. For your situation, I suggest some options.

You could consider leaving your system on all the time with the volume turned to zero. Barring that, here are ideas.

Most new amps have a relay which disconnects the speakers on turn off and delays connecting them on turn on. You need to add this function to your system, and I have seen kits for this. A design which I know to work well is online, in the Hafler DH-500 manual, page 16:

http://www.hafler.com/techsupport/pdf/DH-500_amp_man.pdf

The relay board shown, dubbed board PC 9C , has a single chip that controls a relay for the speaker protection. If you don't care about DC offset protection, you can use just the relay and the diodes and passive components around it (5 components). It is set up to delay turning on the relay by having a very weak power supply that slowly charges up, but rapidly turn it off on power failure or switch off.

You can just ground the relay (connect the point shown as the collector of Q201 to ground) and have only the power on/off anti-thump protection. The values of the capacitors and resistors in the network powering the relay are chosen to have a slow build-up of power to close the relay, but a rapid decay on power loss. This is set up for a 48 volt relay, you might need to use a 12 volt or 24 volt relay instead based on your power supply voltages. In that case you will have to adjust part values to get the turn on delay the way you want it (1 second is usually fine). Possibly you will need a resistor in series with the power coming in to reduce the current, in addition to the components shown. If you just do the simple on/off anti-thump protection I just described, it's only 5-6 small parts and a hefty relay.

For a big amp, you will need a very high current relay, not the sort you find at Radio Shack. Bosch makes some good 12 volt ones for the auto market for battery switching, and there are good 24 v relays used as motor contactor relays you'll find in industrial catalogs, either should work. Good luck!
 
Perhaps a better way might be to use the relay right at the input to the output stage. Let me explain.

At switch-on, the relay contacts are closed, and the relay connects the upper driver base (assuming double emitter follower output stage) to a voltage source 1.3 volts above ground. This source can be as simple as a resistor from the positive rail and two diodes in series to ground. This forcibly holds the upper driver two junctions above ground, preventing current flow. The Vbe multiplier will hold the lower driver base four junctions below, so they too will prevent the outputs passing any sort of current.

After a short delay, say one second, a timer circuit pulls in the relay, and the voltage source is disconnected from the upper driver base. By this time, of course, all is stable, and the offset control has pulled the outputs to within a few millivolts of ground.

All is well and God is in his heaven.......

It works well, and it uses a tiny relay which is completely out of circuit once the amp has stabilized.

Cheers,

Hugh
 
Speaker saver circuit

If this solution or one similar has been proposed already, I am sorry.

In my home-built amplifier project, I too noticed this annoying thump at turn on and turn off. The cause of the turn on thump is a current in-rush that originates in the power transformer. When it is off, it has no magnetic field set up yet, and thus looks like a short circuit. This initial transient dies off quickly, but its effects can be quite damaging as the spike can be somewhere in the 50+amp range!

To eliminate this problem, I decided to place a series resistance (100-ohm, 10W does fine) in series with the transformer. The field has been well established after much less than a second, at which point, the series resistance is then removed via a relay (5V normally open). This relay is controlled by a 5V signal (I put in a small wall plug-in 5V regulated supply) that goes through a simple bit of "logic" circuitry. I used a small a simple RC circuit with the values of R and C chosen so that the time constant yields a sufficient charging time for the capacitor. The voltage from the cap is then compared to 2.5V via a comparator opamp. 2.5V is obtained by feeding 5V into a Voltage divider with equal valued resistors. As soon as the voltage on the cap goes above 2.5 volts, the opamp outputs its top rail and drives an output transistor (acting as the base current supplier) to supply power to the relay. I also tied this output to my speakers which are also relayed to protect them when not powered on. In this design, the turn on for the amp is completely controlled by the switch supplying power to the RC circuit.

On the plus side, you get rid of the in-rush current (well, at least minimize it), protect your speakers, and it moves the high voltage away from the control switch.

The drawback is that the 5V circuitry (transformer and regulation) are always on unless you want to flip two switches to turn on your amp which I don't 😉. But they don't draw much power anyway.

I realize this may be hard to follow without a schematic (I drew it up free hand and built it, but didn't bother with a formal schematic). But the hope is that I've possibly given you another option you can explore!

Please fire back any questions you may have about this.

~AAW
EE student
RHIT
 
Status
Not open for further replies.