What's wrong with this speaker?

[SeanAtx]

A while ago, I bought a used & cheap PSB Alpha C to use for center channel duties. It sounded horrible and Craigslist doesn't do refunds, so I put it back in the box and stuck it in the garage. 2 years later and it's time to clean out the garage, but before I drop this thing off at the nearest dumpster, I wanted to run some measurements to see why it sounds so bad.

As you can see below, there's a few issues with the frequency response and a huge blob of distortion starting at 3kHz that coincides with another massive dip in the frequency response. I'm tempted to open her up and see what's going on, but wasn't sure where to start.

 

[ChrisG]

Well, the tweeter sure isn't happy. Might be possible to fix it, though: my suspicion is that the voicecoil is misaligned, and it might well be possible to re-center it.

Post up some pictures and we'll have a look.

You could also try running a sweep through just the tweeter (ie, to rule out the crossover), but set your sweep at a low-ish level and 3kHz upwards. The tweeter won't have any protection, and LF signals are likely to damage it.


Chris

 

[solderdude]

You could measure another speaker to see if that one also shows a 30dB dip around 6kHz.
Maybe a faulty crossover or tweeter ?

 

[terryforsythe]

2 years later and it's time to clean out the garage, but before I drop this thing off at the nearest dumpster, I wanted to run some measurements to see why it sounds so bad.

Depending on how in depth you want to go, you could run a sweep of the tweeter by itself, bypassing the crossover, starting about 2kHz. That would provide more insight into whether it is an issue with the tweeter itself. If you see the same type of response and level of THD, then the tweeter is bad. If not, there is an issue elsewhere in the speaker.

 

[mhardy6647]

Maybe too high a frequency but I wonder if it might be reversed polarity on the tweeter wiring?

 

[terryforsythe]

Maybe too high a frequency but I wonder if it might be reversed polarity on the tweeter wiring?

I thought that too, but that doesn't explain the high distortion, which peaks where the dip is in the frequency response.

 

[SeanAtx]

I actually remember replacing the tweeter back when I first got it, but it didn't make a difference. I can try reversing the wiring, but the connectors are color coded red and black to match the wiring. Worth a shot at least. Here are pics of the crossover and tweeter.

 

[audio_tony]

In two way systems, the tweeter is usually wired out of phase with the woofer. Check that first.

 

[kemmler3D]

Seems pretty obvious that something is wrong with the tweeter, but if you replaced the driver, then the suspicion turns to the crossover... I would second @terryforsythe 's suggestion to sweep the tweeter by itself. But probably start the sweep at 1500hz at the lowest to be safe, and do it pretty quietly...

 

[DVDdoug]

In two way systems, the tweeter is usually wired out of phase with the woofer. Check that first.

Right... Or the midrange in a 3-way.

The crossover introduces phase shifts, so depending on how everything is designed reversing the polarity of one of the drivers can bring them back in-phase (at the crossover frequency where it counts).

 

[Sokel]

Response is worst than it looks at the chart,see the scale.
These dips at 600Hz and specially at 1kHz shows multiple problems.

You said you replaced the crossover,but what did you replaced it with?
What I see is not only an unhappy tweeter but very unhappy mid too.

 

[SeanAtx]

Here's the tweeter directly connected to the amp, seems to be fine

 

[kemmler3D]

Here's the tweeter directly connected to the amp, seems to be fine

View attachment 404155

Yeah, seems like the issue is probably the crossover or (doubtful) maybe some real weirdness with the cabinet. Can you replace / repair the crossover?

 

[SeanAtx]

Not seeing any crossovers on eBay, so it looks like repair is the only option here. Or I can toss the crossover completely and use the drivers and cabinet as the base to some kind of DIY boombox build. I don't have a need for a center channel at this point so very tempting to go that route.

 

[Sokel]

Not seeing any crossovers on eBay, so it looks like repair is the only option here. Or I can toss the crossover completely and use the drivers and cabinet as the base to some kind of DIY boombox build. I don't have a need for a center channel at this point so very tempting to go that route.

I won't be hard to simulate your existing x-over with VituixCad or similar software.
Just follow the connections and values closely.

 

[terryforsythe]

Here's the tweeter directly connected to the amp, seems to be fine

View attachment 404155

Frequency response isn't bad, but distortion still is high (EDIT - I was looking at it on my phone screen. Now that I can see the scale, the distortion is not too bad). Try reversing the polarity of the tweeter connection and check the capacitors in the crossover.

 

[AnalogSteph]

A bit of crossover reverse-engineering seems in order, it doesn't look too complicated. I guess the tweeter gets a 3rd-order CLC highpass with the 0.2 mH and adjacent caps and the 1R/10W series resistor, and the woofers have an LC lowpass with the 0.035 mH and the big cap? Should be pretty clear with a view of the underside.

I would also suspect that one of the tweeter caps has gone open, or there's a bad solder joint. Shouldn't be too hard to pinpoint with a multimeter and some capacitors of similar value from the junk box that can be tacked on in parallel.

 

[ChrisG]

Tweeter looks fine. In that case, then, replace the electrolytic caps in the crossover.

I've seen/measured this before, albeit not quite as severe: a pair of Kef 104.2 have a lot of electrolytics in the crossover, and this pair had a weird distortion profile. Turned out some of the caps had started getting a bit semi-conductor-y and adding lots of distortion (a few percent, even at low levels). Replacing all the caps got the results you'd expect: much flatter, much cleaner.


Poly caps would be preferable for longevity, but direct replacement with other electrolytics would be better than nothing.
If you do go for electrolytics, make sure they're non-polarised.


Chris