What to do with unsaleable speakers?

[bluesman]

I don’t think you know exactly what’s wrong yet - you’re trying to treat an unknown disease on the basis of guesswork, but you need facts to make the diagnosis. The fact that tweeters “often fail” in that model does not mean that yours did.  You describe the imbalance as “severe” but say you can’t hear anything over 10k.  Your tweeters start putting out at close to 3k, as I recall, and that’s where timbre, brilliance, “air” etc start to be defined but none of the fundamentals and major harmonics are that high.

 

It’s a bit hard to believe that you can easily hear a gross imbalance in SPL from lack of output above 3k, especially if you truly have no functional hearing sensitivity above 10k. It’s a truly rare audiogram that breaks sharply at a specific frequency and even rarer if it’s bilaterally symmetric - so you almost certainly have a progressive loss of threshold sensitivity that starts at or below 1k.

All you know for sure (assuming that it’s constantly on the same side despite swapping input or output channels/cables) is that there’s a problem in one of your speakers.  There are several things you can do to get a more accurate diagnosis.

 

1. A simple, cheap SPL meter is very useful, but if the imbalance is truly big, just download a free SPL “meter” for your smartphone 
2. You can use recorded music as a test source, but you might want to use an online sine wave generator to ensure consistency of input levels and signal between channels. You can use the audio out jack on your phone or computer - SQ is irrelevant here.  Obviously an external DAC is also fine to use.
3. I’m pretty sure your speakers are designed for biamping.  Remove the jumpers and connect your amplifier’s outputs to the bass input terminals ONLY.  Listen to and measure the SPL from each channel from a close, fixed distance on the axis of each driven speaker.
4. Move your speaker leads to the high frequency terminals and do it all again.  If using a tone generator, start at 1k and go up from there.

I don’t know where those terminals are in the crossover chain, i.e. whether there’s any filtration between them and the drivers or if they’re direct (which I doubt). You’ll need to find out from Revel, because it will affect the information this test will give you.  If those terminals are inputs to the crossover, any imbalance could still be caused by a driver or a component in the crossover.

If the imbalance is audible only through the bass terminals, the problem is almost certainly not your tweeter.  If it becomes audible at some point while sweeping the frequency upward using a tone generator into the full frequency inputs (with jumpers in place), you’ll know where it starts and that should narrow the search.  Just for completeness, you might also want to check the jumpers themselves and the terminals to which they connect.

 

And if it really is a tweeter, you can simply remove the jumpers and wire up to a pair of external add-on tweeters. There used to be many on the market.  You could even use a good, small, inexpensive passive “monitor” as your new tweeter - there are many excellent ones for less than $100/pair.  If your HF hearing is that compromised, this should be fine - just sit them on top of the Revels.  To the inevitable critics about to post how terrible this solution is, I don’t think the phase and sensitivity differences will matter much to someone with poor high frequency hearing - especially when the stated alternative is to use boundary effect as a fix and listen to them as they are now.

 

Good luck!

[bluesman]

17 minutes ago, Mike Rubin said:

Thank you very much for this detailed precis!  
 

I will be home a lot for the next few months - nearly all of us will- and this gives me a chance to undertake more detailed testing.

 

Just a few things:  

 

These are jumpered now, so I can play with the two sets of terminals that you suggest.  
 

I also have a system test record that plays tones at about 40 closely-spaced frequencies.  I haven’t had the patience to start at the bottom, but, beginning at about 2500 hz does reflect an audible imbalance, although the degree depends on the end of the couch that I occupy.  I will use my phone’s SPL app to verify more precisely; that hadn’t occurred to me.

 

Finally, if I end up having to mate outboard tweeters, I will just move on from these speakers.  Their market value isn’t really that high snd I’ve gotten my money’s worth ovr the more than 15 years I have had these.

 

Thanks again!

You’re welcome - but to be honest, I suspect it’s not a tweeter and new speakers may be the best choice. Having said that, it’s always worth a thorough evaluation first.  Measure very close to each driver - you won’t learn much from the ends of the couch. Hold the microphone right in front of the cone. A fixed measured distance is best, and a tripod or other stable support to hold it at the same angle makes the measurement even more repeatable.
 

I still listen to the Rogers LS3/5a monitors I bought new in 1976, and my son uses my KLH 19s (circa 1969) in his home theater system. Age is just one factor for both speakers and for their owners  

[bluesman]

1 hour ago, Lone Mountain Audio said:

Another simple test is to swap tweeters from right to left/left to right.  Does the loss follow tweeters or stay the same with the crossovers?  That narrows it down fast. 

 

Brad 

It’s a lot easier to move wires between terminals than to swap tweeters, especially if you do it at the amplifier.

[bluesman]

2 hours ago, Lone Mountain Audio said:

 

If you are speaking of a biamp or triamp speaker input panels, many of these are not direct inputs to the tweeter.  They are usually inputs to the passive network and a HF section of a network still outputs something to the midrange; a passive biamp network is not a brick wall protecting the tweeter.  Plus there are many parts prior to the tweeter so if the network is bad (has bad parts) then you might think its the tweeter when its really the network.  

 

The only failsafe way to check is to physically disconnect and switch the tweeter between right and left.

 

Brad 

 

You might want to read my post above discussing this.  Here’s the relevant quote so you won’t have to search for it:

 

”I don’t know where those terminals are in the crossover chain, i.e. whether there’s any filtration between them and the drivers or if they’re direct (which I doubt). You’ll need to find out from Revel, because it will affect the information this test will give you.  If those terminals are inputs to the crossover, any imbalance could still be caused by a driver or a component in the crossover.”

 

So as I said above, start with the wires and go to the driver itself if the problem is tweeter-related. I think it’s probably not, based on the description.  A truly gross loudness discrepancy is unlikely if it’s the tweeter, which starts getting its drive at 2700 Hz in this system. This is especially true for a listener with what he describes as a severe high frequency hearing loss in both ears, unless he has a bizarre cochlear disorder with severe recruitment (which is also very unlikely).

 

I think he’s much more likely to find the problem driving only the midrange down and will never have to test the HF circuitry or driver. I could be wrong, but the only data we have suggest that I’m not.

[bluesman]

36 minutes ago, Lone Mountain Audio said:

Sorry Bluesman, I missed that post and said the same thing you did.  Forgive me!  I agree tweeters are so small and delicate they usually either work or are dead- I've not seen them reduce level and stay consistent response wise.  I am with you, it is likely midrange crossover or some unusual failure.

 

Brad 

No worries, Brad.  But this isn’t a matter of opinion - the physics of sound tell the story. The combined SPL of two sound sources of equal levels is an additional 3 dB, which (although clearly audible to listeners whose hearing sensitivity to the frequency range in question is adequate) is not a major difference. And I assure you that a tweeter crossing over at 2700 Hz is not putting out an SPL equal to that of the other drivers in that system - so its contribution to the speaker system’s output is less than 3 dB above the output of the other drivers. Even if totally nonfunctional, it would simply not cause a gross level discrepancy in comparison with the functional side.