Dead amp - ?speakers miswired

[danglin]

Asking for your thoughts on an issue I have encountered with a background listening setup for an exercise space/workshop we have just built. New construction.
Instead of buying all new more expensive amps for this setup, I had a scarcely used Audiosource amp rated at 60 watts per channel and able to drive a 4 ohm load to drive four in-ceiling B&W 8 inch 2 way speakers rated at 8 ohms with 90 watts per channel power handling and 90 dB sensitivity. Not long after getting everything installed and wired, I began to notice sound cutouts that occurred simultaneously with the amp clicking. Cutouts lasted seconds to a minute or more. Within a couple weeks, one channel and then the other died.
I had checked wiring and noticed that the speakers were not wired with two wires per speaker like I have installed myself at a different property. Come to find out, the four speakers were wired in parallel. The amp death still seemed odd, so I asked the installers to come back and look. Turns out the amp was seeing a two ohm load, and the speakers measured a 3.8 ohm impedance at the voice coil.
According to B&W, this is “OK” and “typical,” but not what I was expecting for an 8 ohm speaker, though I know impedance varies per frequency.

Neither the impedance at the amp, nor the speakers themselves, had impedance measured at installation.
I have a new amp now, but waiting for the installer come up with a solution. One thought is to run the two wires per speaker like I have done before - the other would be to wire each channel in series, which would double the impedance. Both solutions would require fishing, perhaps opening ceiling/wall.

Is what I am describing represent typical installation practice or expected speaker impedance, actual vs rated?
Sorry for the epic tome.

David


Quote Reply

 

[Pogre]

Others can answer this better than I, but in general parallel wiring speakers like that halves your impedance. If all 4 were parallel wired together then yes, that could definitely present an ugly load to any amp.

 

[danglin]

I do understand that parallel wiring reduces the impedance seen by the amp, but the surprise was the lower actual impedance presented by the B&W's at the voice coils. 4 ohms at the amp would have been an OK load, but not two ohms.

 

[Pogre]

As far as what B&W rates the speakers at, sometimes those impedance numbers are a little bit fudged. The impedance of a speaker is all over the place and the ratings are based more on an average than 1 solid number. They may very well dip down to 4 ohms at a particular frequency. Hang tight for better answers tho. I'm still very much a learner myself with electrical loads and speakers.

 

[lovinthehd]

The impedance measured directly on the driver (dc resistance) isn't quite the same thing as the operating impedance range. Just assuming how speakers are wired before hooking up an amp....

 

[danglin]

I have a followup question, per Alcoholic Jedi above. I have a nice Fluke multimeter, but no frequency analyzer or sine wave generator. Is the measurement of resistance measured at the speaker +/- terminals a true measurement of speaker impedance? From what I am reading, it seems not to be. I measure 3.9 ohms at the speaker terminals of my 8 inch 2 way B&W ceiling speaker, which are rated at 8 ohms.

 

[lovinthehd]

I have a followup question, per Alcoholic Jedi above. I have a nice Fluke multimeter, but no frequency analyzer or sine wave generator. Is the measurement of resistance measured at the speaker +/- terminals a true measurement of speaker impedance? From what I am reading, it seems not to be. I measure 3.9 ohms at the speaker terminals of my 8 inch 2 way B&W ceiling speaker, which are rated at 8 ohms.

Try this https://www.audioholics.com/loudspeaker-design/loudspeaker-sensitivity

 

[Verdinut]

I have a followup question, per Alcoholic Jedi above. I have a nice Fluke multimeter, but no frequency analyzer or sine wave generator. Is the measurement of resistance measured at the speaker +/- terminals a true measurement of speaker impedance? From what I am reading, it seems not to be. I measure 3.9 ohms at the speaker terminals of my 8 inch 2 way B&W ceiling speaker, which are rated at 8 ohms.

With a reading of 3.9 ohms for DCR, your speakers would have a nominal impedance of about 6 ohms, not 8 ohms. As a rule, DC resistance is about 2/3 to 3/4 of the nominal impedance. B & W speakers produce transducers with impedances that dip low at low frequencies, and I wouldn't be surprised that yours do dip to below 4 ohms at certain frequencies.
Your Audiosource amp is most likely unable to drive your ceiling speakers. I suggest that you try another more powerful amplifier and see if it resolves the problem. It's a question of amplifier power supply capability.

 

[TLS Guy]

I have a followup question, per Alcoholic Jedi above. I have a nice Fluke multimeter, but no frequency analyzer or sine wave generator. Is the measurement of resistance measured at the speaker +/- terminals a true measurement of speaker impedance? From what I am reading, it seems not to be. I measure 3.9 ohms at the speaker terminals of my 8 inch 2 way B&W ceiling speaker, which are rated at 8 ohms.

We get these problems here with these systems continuously. A speakers manufacturers spec means little. Since it is B & W speaker it almost certainly dips to the four ohm range.

When doing these background systems in a commercial space I strongly recommend using a 70 Volt PA type system That would be your best option. These systems do not need to be stereo.

I would recommend you get a 70 volt PA amp. Then put a transformer at each speaker. Then you can connect them all in parallel on the side of the primary transformer. You then make sure that the addition of the watt taps on the secondary side of the amp adds up to 80% or so of the total power of the amp. Make sure you do not go over 100%. That way you have a very stable system and not blow an amp again. You do not have to put the same power on each speaker with this method.

Your other alternative is two use two, two channel amps, or one four channel distribution amp. For what you are doing you need either to go the pro 70 volt rote, or use a professional four channel distribution amp. Your installer sounds very inexperienced to me. That was no way to engineer a commercial sound system.