Reducing Volume

[russv]

I had read plenty about using attenuators, light bulbs, iso booths, etc.

I am not a amp tech, hardly, but why wouldn't this work. Build a box with one or more speakers (old pieces of crap) tear out the cones, pack them in foam, and wire up so the impedence is what you need. Run this along with your cab. So instead of running an extension cab for more volume, you run this box for less. You could add switches to add/subtract number of dead speakers you are running, or dummy loads, etc.

I am really curious, what folks think of this idea.

[Old Tele man]

...each "dummy" speaker has its own DC-resistance and AC-impedance, and adding them into the path "changes" the load being 'reflected' back through the OT to the tubes.

...yes, it COULD be done, but it'd take LOT's of carefull planning and design to ensure that each added "dummy" speaker didn't change the TOTAL effective load impedance (an "L-pad" would work).

...also, Ted Weber already sells something exactly like what you're describing.

[russv]

I guess I understand that part. ie. when I wire my interal combo speaker (8 ohm) on y cable to my extension cab (8 ohm), I plug into the 4 ohm tap.

If I could come up with a wiring scheme, that always had 8 ohm, then more dummy would = less noise.

Would this sound like crap? It seems to me this would be better, as you driving really speakers.

[John Phillips]

If you tear out the cone of a speaker, the voice coil will fry when you put power into it. It won't be suspended or move properly and won't cool itself. I don't know what the percentage of power handling you'll lose will be, but it could be as high as 90%, and certainly well over 50%. It also won't have the same electrical characteristics as when it's suspended and allowed to move in a magnetic field, so it won't have the same effect on the tone of the other speakers. This is why the Weber Mass attenuators that use voice coils as part of the load also use high-power resistors and other parts as well. It's not that easy...

There is one very simple thing to remember: if it sounds like a cheap, simple and too-good-to-be-true means of reducing volume, it is... too good to be true. There is a good reason why properly-designed commercial attenuators are fairly substantial, quite complicated and expensive bits of kit.

About the only exception I can think of is using simple high-power resistors as parallel dummy loads, and even then they change the tone because they don't have the same impedance characteristics as speakers. You can find such things very cheaply if you know what to look for though... try an old electric heater or kettle ;). You're looking for about double the resistance of the speaker impedance you're trying to duplicate, which you can find either by calculation from the power rating and the supply voltage, or by measuring it with a multimeter. Unlike light bulbs, high-power resistors and heating elements don't change resistance much with temperature.

[russv]

Fair enough. I like I said, no amp tech here. Getting a dummy load isn't hard, or expensive online. I probably will just get one to use to tame things when I need it.

[John Phillips]

If it helps, this should illustrate the problem! :)

http://www.theangelconversations.com/pics/VoiceCoilResistor.jpg

That's a voice coil from a 300W speaker, and a 100W power resistor.

You can probably see that it's unlikely that the voice coil would be able to handle even close to the same power as the resistor, like that - the wire is simply wound onto a glassfiber and paper former (which is a good thermal insulator), whereas the resistor is encapsulated in an aluminum heatsink which itself can be bolted to a larger heatsink or the enclosure - let alone three times as much.

The only way the speaker coil can handle that sort of power is because it's suspended in a very narrow gap through which air is forced as the cone moves, so it cools itself very effectively.

... although on that one, not quite effectively enough :(.

[russv]

Great illustration. I am need nothing as high as 100 w. I may play around with an old speaker and epoxy the cone to deaden it. I love this kind of info, and thinking about homemade solutions.

[John Phillips]

This may be useful...

I had another one of those 300W drivers with an intermittent fault in the voice coil, so in the interests of science I decided to sacrifice it and find out just how much power it would take :).

As expected, once I was able to get at the coil I was able to bypass the fault, which was between the braid and the coil itself, and get some clips on it. I rigged it up with a Variac, current and voltmeters, and put it on an absestos tile...

It was a nominal 4-ohm speaker BTW. In all cases except the first, the current draw fell and then stabilized as the coil heated up.

Applied voltage / current drawn (initial) final / resistance (initial) final / power (initial) final

5V / (1.0A) 1.0A / (5.0 ohm) 5.0 ohm / (5W) 5W
10V / (1.8A) 1.6A / (5.5 ohm) 6.3 ohm / (18W) 16W
15V / (2.4A) 2.1A / (6.3 ohm) 7.1 ohm / (36W) 30W
20V / (2.6A) 2.3A / (7.7 ohm) 8.7 ohm / (52W) 46W
25V / (2.8A) 2.6A / (8.9 ohm) 9.6 ohm / (70W) 65W
30V / (3.0A) - / (10 ohm) - / (90W) -

At 5W, it got warm, but not enough to stop me holding it.
At 16W, hot enough that I could touch it but wouldn't want to hold it.
At 30W, too hot to touch even momentarily.
At 46W, the enamel coating slowly turned dark brown like the coil in the pic above.
At 65W, the coating started to smoke lightly and the coil was hot enough to burn a piece of wood held against it.
At 90W, the coil former started to smoke very badly and I didn't wait for the temperature to stabilise since a short would probably not have been too good for my Variac, which was already at its rated limit at 3A.

So - about as I expected. I would say that 30W was about as far as could be called a safe rating, since the coil coating did begin to degrade above that. And that's a 300W voice coil...

[Old Tele man]

...immerse the voice coil in light oil to act as a thermal coolant...it worked for "hams" trying to 'cool' 6L6's while pumping 80-90W from a pair.

[jjasilli]

OK if you're the experimentally inventive type. Otherwise Ted Weber sells the ready made dynamic attenuator you're looking for.

[russv]

That is a very thorough experiment, John.

I am not sure I am to mess trying to contain oil. Seem like a recipe for a bigger mess than 2yr lab can make.

Yeah, I do like to experiment. I probably should concentrate on play, but ya know, ones got to fiddle.

Would an attenutator, any type sound better in parallel? Most often you seem them in series, which make sense for impedence matching. Just curious if any has any experience.

[John Phillips]

^ ^ ^ ^ ^ ^ I love to experiment too :).

Would an attenutator, any type sound better in parallel? Most often you seem them in series, which make sense for impedence matching. Just curious if any has any experience.Yes. You don't get exactly the same tone, since the attenuator rarely matches the impedance curve and damping characteristics of a real speaker very accurately, so power distribution is different at different impedances and resonance is altered, but it seems a little more transparent than series attenuation.

You don't have to use equal impedances either, as long as the total is within the safe range for the amp. You can achieve 3dB reduction with equal impedances, 5dB with the attenuator half the impedance of the speaker, and 7dB with it a quarter. (Set the amp to 4 ohms for the last two, since the total impedance will be between 2.66 and 5.33 ohms.)

The attenuator which sounds best used like this may be very different from the ones which sound best used in series - you're probably looking for the closest match to a real speaker impedance, and the Marshall Powerbrake is a good candidate. (That's why I use it as my bench test load, FWIW.) I haven't actually done much experimenting with it in parallel though.

[russv]

Cool. Thanks for the great info.

[skylabfilmpop]

What you are suggesting is the equivalent to putting a 16 ohm speaker on a 4 ohm load. It doesn't reduce the volume that much,sucks some high end though. You could build a big box full of foam and put it over your head .....;) Seriously we have wrestled with non master volume amps being too loud since amps were invented I am sure. And attenuators screw up the tone although the MASS is the best I have heard. I have seen people screw a piece of plywood over 1/2 of their speaker cabs, use plexi baffles and all sorts of things. IN reality there are only a few fixes. 1) use an amp appropriate to the venue you play in. If you play small bars 10-20 watts is more than enough. Even a bassman or a plexi style 50 watter is too much power for small rooms unless you are a very dynamic player and use OD pedals for gain. 2) Get a Variac. These are about $75 on ebay. Granted that you haven't turned the voltage down, biased the amp there and then raised the voltage you will likely not hurt the amp, just wear the tubes a little more quickly. The majority of the touring and session player on the Divided by 13 website clients page are using Variacs, some of them night after night on major tours without failure. You can knock a 15-20 watt amp down to VERY polite levels with one of these and even do a lot to tame a 50 watter. Aside from the obvious choices of Variac, small amp, attenuator, iso cabs a good master volume mod is not a bad idea and some great amps like Matchless have them stock. A trainwreck style MV will effectively dial out at 10, is very transparent so it will let you go "non master" when you are in a big room.

[rockon1]

What if you cut the cone out (and perhaps the speaker frame if it worked) leaving just the spider and voice coil intact perhaps beefing up the spider making it stiffer and using like that? That way the coil would still move no? Bob