Suspected Ditton 33 Tweeter Problem Solved

[timewave]

First a little bit of history.

I bought my Ditton 33’s as a birthday present to myself back in 1975. I think I first used them in conjunction with my Philips N4450 reel to reel tape machine and then later a Goodmans Module 90 tuner amp. When my disco speakers went faulty they were even pressed into service to keep the party going. So they have seen hard service although I have managed to keep them looking reasonably tidy.
These days they live in our bedroom (I got away with that one because they are a good match to the rest of the furniture) connected to a 1980’s Kenwood amp fed by a Sony DVD HDD recorder, Sky box and MP3 player.

For some time now I have been adjusting the balance control on the amp to compensate for reduction in output on one channel (could it be my ears). Last weekend I decided to investigate further.
Having swapped the speakers around to see if the fault migrated I eliminated the amp as a possible culprit and realised that it was one speaker at fault. There was almost no high frequency output. I feared the worst given their misspent youth but having removed the HF unit for test, proved that it was in good shape by fitting it into the other speaker.

So the problem was in the crossover unit. New territory!
Having traced out the circuit I went for the HF associated capacitors and tested them on the capacitance range on my Fluke 179 comparing components in each speaker.
The 8uF capacitors measured 9.5uF and 9.96uF close to each other but some way off the correct value.
The 12uF capacitors measured 8.7uF and 13.6uF, not close to each other and the one in the faulty speaker was the lower value.
I then checked them on my ESR meter and the 8.7uF one measured 7 ohms and the rest measured zero ohms so with a combination of several parallel components I replaced it.
Hey Presto!! HF restored.
Ok, problem solved but curiosity tweaked. I checked all the remaining capacitors. All read virtually zero ohms ESR but all were considerably higher capacity, between 15 and 20%, of their marked values.
At this point I decided that I was going to replace all the capacitors not just because they all appeared to be out of spec but also in view of their age (and mine) I didn’t want to run the risk of having to take the speakers to bits at a later date.
The capacitors are all ELCAP 50V units and a search on the web said that ALCAP units from Falcon Acoustics would be reasonably priced replacements.
The components arrived very quickly accompanied by a note on capacitor tolerances.

“NB most hand held multimeters use the wrong frequency for capacitors, particularly for electrolytics where they can often be +10% in error”

So I checked the new components and indeed all measured about 10% higher than their stated value.
Having recapped one speaker I compared it with the other one and it sounded crisp and clear. The other one sounded mellow or dull to my wife who had no idea which one had the old capacitors in it.
Having completed the work on the other unit we then spent 3 hours rediscovering our favourite music tracks.

Conclusion
A lesson learned.
When faultfinding always keep an open mind and don’t assume that your test equipment accuracy is beyond doubt. I was suspicious of the readings I was getting because components of the same marked values were wrong by similar amounts and all but the faulty component measured high. Even so I still wanted to believe my “trusty” Fluke!!

Oh, and new is not always better. These speakers sound amazing now after a little TLC.

[camtechman]

That was a good result as the usual fault was the tweeter.

Over the years I've had to replace the tweeter (when they were still available) for customers.

[bobbyball]

What a great story, nice that a piece of "older" kit is appreciated by the "other half" too!

Robert