Author Options:

I need help correcting impediance on speakers. Answered

Here's some background on what I'm dealing with: I run a haunted house in Nashville TN called Devil's Dungeon. We have a sound system that is controlled from a rack with 6 2 channel power amps. 4 of them are 260W by 2 channel and 2 are 500W by 2 channel. I have 10 cd players running to a 24 channel mixing console and the way I control each room on a separate channel is by using the post-fader sends (yeah, it's a really old peavey, but the new ones don't have post -fader sends per channel) as a line out per channel. IE: i run a line directly from the post-fader send per channel to the power amp's input. Here's my problem: There are more speakers than there are outputs, and the speakers are placed where they are needed in a particular scene. It is a large building and it would be highly laborious to try and correct the impedance by putting extra speakers and with series/parallel combinations and as we are already open this season and I don't want to re-engineer the entire system. Is there any way to correct the impedance with some kind of variable resistor? Everything I've seen so far that could handle the power is pretty large, and if possible, I'd like to put it in a relatively small project box. Can this be done with a potentiometer? I only need to correct it by a few ohms and the only POT's I've seen have been in K increments .



9 years ago

The main problem is that the lower impedance on the parallel speakers are making the Amps run really hot. I have a 24"fan blowing air through the rack, and one whole side panel is removed to help air circulation. What I was hoping to do is make something that could be adjustable so i can make 3 or 4 to correct multiple lines, but have one design for all. As for running new lines, the scenes are painted, and decorated, and the cables and speakers are behind, or inside props. Plus some of the runs are up to 250' from the sound room. Nacho: When I turn the system on, the amps begin to clip at a very low volume. Sound quality isn't that important, to a degree, but they get real bad distortion at levels needed for the scene, and amps will trip the built in circuit breaker if it gets too loud for too long. half of the speakers have a secondary volume control on them that is a riostat like gmoon was mentioning, but it has the impedance correction on them to balance it out to 8ohms regardless of the setting. I'll take pictures of the sound booth when i get there tonight and post them by the morning so you can see what I'm working with, maybe it'll help explain what I'm looking to do better. thanks to all for your suggestions so far!

the secondary volume control on the speakers are like this:


Yeah, that's an L-pad... Can you tell us the existing impedances? (output, speakers, etc.) Wiring the L-pads differently can probably get you there...

everything is considered "pro audio" and are peavey preformance amps. everything is running on 8ohms

I'm a little fuzzy on this-- if the amp's output impedance is correct for the speakers, it might not be source of the problem. Is the mixer driving the amplifiers too hard? Are there low / high impedance inputs for the mixer, and can you switch them? I.E., could you be clipping at the inputs of the power amps? (have you tried connecting another device--iPod, the CD players..directly to the power amps?)

No, because if you switch the outputs around and run it to a line with correct speaker impedance the clipping stops. The signal doesn't clip at the board either.

OK, I interpreted "everything is running on 8ohms" as meaning that the power amps and the speaker were indeed correctly matched...

Have you tested the impedances of the speaker networks? Just to see if the L-pads are correctly matched to the speakers. If an incorrect-resistance L-pad is connected, you'll have a problem. Also check if the speaker wiring is consistent and sane.

You can do use an ohmmeter to test speakers (unlike most inductors)--you should get a lower value than the actual impedance, like 3 ohms for a 4 ohm speaker, 6.7 ohms for an 8 ohms speaker, etc.

If we think the speaker's impedance is too low (too large a load on the amps), then try connecting one side of the L-pad in series with the output. If the L-pads are 8ohms, then one half should be an 8ohm rheostat. Adding extra resistance in series should decrease the load....

I put an ohmmeter on the lines going to the speakers at the rack, about half of them were within an acceptable 1.5 ohm variance, but the other half was reading in the 4's and one even was at 2.6. I forgot to take the pictures last night, but i'll try and have all that stuff together and better explain what i need in terms of packaging.

. The 250' runs may be causing problems. Are they in the bad bank? A run that long is likely to introduce impedance, capacitance, &c; problems.

250' shouldn't be too bad for a low-impedance signal (it would hammer a high-impedance one, like a unbalanced mic or a guitar, though...) Scurge, it almost sounds like paralleled extra speakers were added to some rooms--with no thought to the total impedance. Might be only the "bad half" causing the amps to overheat.

. With most modern amplifiers, impedance matching is not as critical as it used to be. Try just hooking up the speakers in parallel and gradually raise the volume. If you don't get any distortion, you're good to go.
. If you get distortion, try an impedance matching transformer.

I'm all for trying the existing wiring, or adjustments to the existing wiring.

The only thing wrong with a transformer solution is we're dealing with the amplifier output, and a high-wattage transformer is expensive. And you shouldn't skimp on the rating, since haunted houses run those amps maxed for hours at a time.

For instance, this 100 watt Weber Impedance Converter Transformer is $48... (the 2x100 watt amp in my car only cost about $22.)

Weber transformers and chokes page.


9 years ago

Well, more speakers will give you more options than too few... have you explored all combinations of series / parallel with the existing speakers? That's the simplest approach....unless they are hardwired together in the walls, with only two leads per room to your control room.

Otherwise--yes, a resistive load can be used with the speakers (inductive load.) It's not going to improve your sound quality, however. But they can be used to match the speaker impedance to the output impedance of the power amps.

There are several ways to do it. One is to use fixed resistors--power resistors in the range / wattage are you need. Without more detailed info, it's hard to be specific.

For example, an 8 ohm, 100 watt resistor can be used in parallel with an 8 ohm, 100 watt speaker to give you a 4 ohm, 200 watt load. Don't expect it to sound as good as two parallel speakers, however.

And obviously, it won't be as loud, the resistor is soaking up 1/2 the power. So you'll be attenuating the output, and tossing away a good bit of output. Luckily, 1/2 the power isn't one half the loudness... (loudness is a logarithmic function, and 1/2 as loud is 1/10 the power. ) It's very likely you'll need to heat-sink your power resistors, probably by bolting or epoxying them to the metal project box.

If you need a variable, low-ohm, high-wattage resistance then a rheostat, or an L-pad is the way to go... Searching for "L-pad" will give you more information (although they are often used for attenuation only, not impedance matching.) Fixed power resistors will likely be cheaper...


9 years ago

The short answer is no. The wattage of your resistors will need to approximately match the wattage of your speakers and/or amps, and 200+ W resistors ARE big. Is it an actual problem? I thought modern amps were pretty robust, and it doesn't sound like your setup needs "maximum fidelity." You could put "enough" large power resistors in series with the speaker circuits AT THE AMP side to prevent blowing fuses or whatever, and let the speaker wiring float relatively uncontrolled...

If you knew exactly what impedance you needed to correct by then you could add resistors or those other things, like resistors but they work by impedance, the name has slipped me...