Speaker Protection Circuitry

[admin_exported]

Hello all,

I'm new to the forum and to the world as hifi as well. I've had several audio systems in the past that were built around SPL/high volume levels, but never anything that qualified as hifi.

So lately I've been doing some research and am working on putting together a system. So far what I plan on putting together is:

Receiver: Yamaha RXA-810 7.1 networking reciever (100W x 7)

Fronts: B&W 683 or 684

Amp (fronts): Adcomm GFA-555 (200W x 2)

Sub: Image Dynamics IDQ-10

Amp (sub): Butler Audio 2150 (500W bridged)

The center channel and surrounds I have not decided on yet, and may have to wait until a later date due to budget limitations. For now, I'm just focused on getting a good stereo system set up. The Amps are already purchased - the Butler Audio amp is a leftover from an old mobile set up, and the Adcom amp I recently purchased off Ebay.

So one thing that has me a bit concerned is running a pair of $1000 - $1500 (USD) B&W speakers off of an older Adcom amp. Although the amp is highly praised it has no speaker protection circuitry, and I've seen a couple stories online where people had their speakers blown due to DC output from the amp. So would anyone be able to recommend a reliable method for protecting speakers from DC output / amp failure? Is there perhaps an in-line speaker protection circuit I could purchase online? I've also heard that polyswitches are often used for speaker protection, would this be a reliable method? Of course, I would also be looking for something that would not greatly affect the sound quality.

Thank you for any replies. Also any comments on the gear are welcome because, as stated, the speakers and reciever have not been purchased yet. I'm still debating between getting the 684's or 683's, and will be taking a trip down to the local shop to have a listen sometime this weekend hopefully.

 

[Mooly]

This is a tricky subject and something I have looked into and worked on myself.

DC offset protection... which is what you are refering to... detects unwanted DC at the speaker terminals and uses a relay to disconnect the speakers.

The problems are that the usual methods of detection employs an "integrator" which has a finite time constant. This means that if your amp fails and puts say 60 volts DC across the speaker, there is a finite time taken for the circuit to react. It's of the order of milliseconds but that isn't quick enough to stop a speaker cone hitting the end stops with probable damage. It might stop the speaker voice coil catching fire (really that is what can happen) but not guarantee the speaker is undamaged.

The relay is another problem. Most are hopelessly underated for the task and also being mechanical have a finite speed of operation. To break a high DC current flowing in a large inductive load (the bass driver and crossover) is yet another problem as the inductive nature of the load "draws an arc" as the relay contacts open. In some cases this just welds the contacts together.

That said there are Ebay offerings and so on that would be better than nothing.

Polyswitches are no good as they are a non linear resistance and would cause severe audio distortion when driven hard as the resistance fluctuates.

If you are really worried about this problem you could, and this is DIY doable, add two large electrolytic capacitors in series to each speaker. They would be connected "plus to plus" or "minus to minus" (it doesn't matter) and inserted into either the positive or negative speaker lead (again it doesn't matter). That would protect from DC. They would have to be large in value (say 6800uF 63 volt types) but they could easily be put into a small plastic box.

A fuse may be worthwhile even on its own. Value... as low as you can get away with without nuisance blowing.

 

[Anonymous]

Ah yes, I'm not sure why adding a capacitor hadn't occurred to me.

This would, however, affect the frequency response - most notably at lower frequencies, correct? And the purpose of the (relatively) high capacitor value would be to try to mitigate this effect, no? (Since capacitive reactance decreases with increased capacitance).

This is definitely something I will be looking into.

Cheers.

 

[Mooly]

That's correct

You need two caps (four in total for two channels) because you MUST make the cap bi-polar so that it is OK whatever the polarity of any DC applied. That means two identical caps and the total capacitance is then half the value of one cap. So two 10,000uF caps in series make a 5000uF. That would give a "cut off" frequency of around 4Hz for an 8 ohm speaker, give or take depending on the impedance curve of the speaker in question.