One tweeter won't play and I think I know why... Can I get a second opinion?

[ratsark]

Hello everyone! I've been lurking here for a couple of weeks now, getting familiar with all this crazy audio stuff. I've finally made my first high-end purchases, an emotiva bpa-1 (a cute little amp that puts out 37w into two channels) and a used pair of B&W P5's, which I got for $250 because a tweeter wouldn't play. I figured it'd be a simple case of replacing the tweeters, putting me at about $400 for an excellent pair of speakers to get started with and giving me a chance to get my feet wet with DIY stuff.
I got them home last night and decided to run a couple of checks to make sure the tweeter was blown before replacing it. This particular model has liquid cooled tweeters that tend to go out after the liquid hardens. I swapped the tweeters between the two to verify that the tweeter was the problem, but to my surprise the tweeter started playing just fine, and the one that I'd verified as working quit. "Uh oh," I thought, "it must be the crossover." After a little while with a screwdriver and multimeter, I established that all the connections seem to be good, leaving a component problem as the only explanation I can come up with.
The crossover circuit is a fairly simple one, looking similar to a standard 18db / octave crossover I found on the internet: it has what I think is a resistor (but could be a small inductor, perhaps to act as a low-pass filter at 22khz or so): after all, it has resistance of about 1 ohm to a DC current, and is boxish and white. This is in series with a 7mfd 70wv Bennic capacitor, which is in series with two things that are in parallel: a capacitor and the tweeter in series, and an inductor. It's difficult to describe all this, so I'll draw a little ascii diagram:

(+)----R----C1-------C2---Tweeter-------(-)
                         |                         |
                         \---------L---------/


Where R is the thing that I think is a one ohm resistor, the C1 is the Bennic 7mfd 70wv capacitor, L is an unmarked inductor, and C2 is marked "B&W loudspeakers C1066".
When I take the resistance across the two terminals with my multimeter, I see a 5 MOhm resitance that slowly climbs (maybe .5MOhms/second, or a little faster). This is the same for both the working and the non-working speakers. Since for DC the circuit looks something like

(+)-----R-----C1------(-)

I'm guessing that the climbing resistance represents the charging of C1. Since the working and the non-working speakers give the same reading, I take this to mean that C1 is probably still working. My guess, then, is that C2 has failed, dropping its capacitance to levels much lower than spec'd, and thus making it a high-pass filter at some ungodly high value like 3ghz, or just breaking the circuit entirely.
My questions, then:
1. Do you agree with my diagnosis?
2. How can I go about finding a capacitor to replace it? Will any decent capacitor off parts express with the same rated capacitance and (at least) the same working voltage be a suitable replacement? Should I replace the matching capacitor in the other speaker as well?
3. Is there a way to determine what the capacitance of c2 is from the value of c1 and the knowledge that the crossover point is at 3khz?

Thanks for reading my post, and thanks in advance for your help! I've learned a lot from these forums over the last couple of weeks, and I look forward to getting to the point where I can contribute more than just questions.

 

[TLS Guy]

Some wrong assumptions

You are correct that this is a third order 18db per octave high pass filter.

Crossover failures are very rare.

You can not calculate the cap values without knowing the mechanical and above all electrical specifications of the tweeter. Also you need to know the response curve of the tweeter. Likely the values are tweaked to blend the woofer and tweeter roll offs.

You need a capacitance bridge to check capacitors.

I don't think you can tell which cap is bad by your method, as they are in series.

If you want a low tech way of analyzing this fault do this.

Take a set of headphones. Go to Radio Shack and buy a plug that fits your headphone connector and a set of jumper cables. These come in sets of colored wires with small clips on the ends.

Connect the negative of the headphone plug to the ground. Then connect one of the positives of the jack first to the plus side of R, then the C1 side of R, then the connection between C1 and C2 then the tweeter +. Listen to what happens and where the signal drops out. Compare the channels.

If the signal drops out between C1 and C2 it is just possible the inductor has a dead short and is grounding out the signal. Disconnect the inductor and see if the signal reappears between C1 and C2. If it does, the inductor needs replacing.

I think this will help you solve the problem.

Once you have the fault isolated, see if B & W will tell you the component values. You can buy good quality parts from Madisound. If they won't, then buy the replacement part from them.

 

[ratsark]

Thank you! The headphones idea is an excellent one. I'll give that a try ASAP, and try disconnecting the inductor if necessary.

 

[ratsark]

Ok, I got down and dirty with some headphones and alligator clips, and here's what I've found:
Bridged all the way from + to -: full sound
Bridged from between R and C1 to -: full sound
Bridged from after C1 to -: nothing
So it looked like C1 was bad. I bridged the capacitor with another alligator clip and lo and behold, I had a working hi-pass filter and sound coming out of the tweeter and headphones. Luckily, that's the capacitor we know about (7mfd, 70wv) so it should be easy to replace. I've never replaced anything in an audio signal path before, so I'm wondering about a couple of things:
-Should I replace the caps in both speakers?
-Will any 7mfd cap of at least 70wv work well? What price/quality point should I shoot for? Should I stick with electrolytic?

What do you all think?

Edit: in retrospect, bridging that capacitor was a bad idea because it eliminated pretty much all resistance to low frequencies. Could have blown the amp channel or burned out the inductor. But I got lucky: no damage, and diagnostic help

 

[TLS Guy]

Bad C1

Ok, I got down and dirty with some headphones and alligator clips, and here's what I've found:
Bridged all the way from + to -: full sound
Bridged from between R and C1 to -: full sound
Bridged from after C1 to -: nothing
So it looked like C1 was bad. I bridged the capacitor with another alligator clip and lo and behold, I had a working hi-pass filter and sound coming out of the tweeter and headphones. Luckily, that's the capacitor we know about (7mfd, 70wv) so it should be easy to replace. I've never replaced anything in an audio signal path before, so I'm wondering about a couple of things:
-Should I replace the caps in both speakers?
-Will any 7mfd cap of at least 70wv work well? What price/quality point should I shoot for? Should I stick with electrolytic?

What do you all think?

Edit: in retrospect, bridging that capacitor was a bad idea because it eliminated pretty much all resistance to low frequencies. Could have blown the amp channel or burned out the inductor. But I got lucky: no damage, and diagnostic help

Well done! You have made the right diagnosis without a shadow of a doubt. You are now starting your career as an audio service tech.

Now I know why C1 failed. It was naughty of B & W to use non polarized electrolytic capacitors. That is being cheap. They are not good in terms of linearity, when used in crossovers, and they fail! Replace these electrolytic caps in both speakers with polypropylene ones.

The standard value now is 6.8mfd. The 0.2mfd reduction will make no audible difference.

I would use Solen polypropylene capacitors. Here is what you need on the Madisound web site.

http://www.madisound.com/catalog/product_info.php?cPath=404_5_11&products_id=757

They carry a lot of exotic brands, with increasingly hefty price tags. In my view it is akin to exotic cables, a waste of money. I have used the French Solen Capacitors in my designs for years, and can recommend them without reservation. I can also heartily recommend Madisound. I have bought from them for 24 years.
All the Best, Mark.

 

[ratsark]

Just ordered two of the capacitors you linked to. Can't wait to get these things running! Thank you for all your help, you're way better than the B&W support guys!

 

[ratsark]

Did you know there's a 5 post minimum to allow you to send PMs? This post will allow me to respond to your message...

 

[ratsark]

Edit: in retrospect, bridging that capacitor was a bad idea because it eliminated pretty much all resistance to low frequencies. Could have blown the amp channel or burned out the inductor. But I got lucky: no damage, and diagnostic help

No wait, I forgot about the resistor... no real danger