Wharfedale BR/3 Tweeter

These were brought to my attention by a friend in N.Y. Not having ever seen such a model, I bought them. All that was needed was a new foam annulus and foam inner spider (centering device) and a minor repair to the one on the left in the first photo. This fella had a dc resistance of >100 W. Probing around after cleaning the goop from the solder joints revealed a cold joint which was carefully remedied with a 25 watt soldering iron. That wasn't something liked or desired to do because of the possibility of breaking the thin aluminum wire in a location extremely difficult to repair but there was no other option.

These units are ready for re-assembly.

The A unit disassembled again. The impedance curve rises to about 60 W at 114hz and buzzes. Most of it was traced to insufficient adhesion of the inner circumference of the annulus to the cone. Considering the glue applied the second time makes it difficult to imagine that the annulus is still the cause. Also, the resonant point dropped to around 90hz. It is suspected that there may be a loose solder joint under the goop at the cone but it doesn't seem worth the risk of breaking the aluminum wire during the process. Removal of the dust cap made no difference. The coil to cone apex was also given a dose of glue despite that no fracture was noticed as was the coil itself.

The possibility of the coil rubbing hasn't been eliminated despite that being unusual to happen at one frequency although vibrations at resonance can be somewhat unpredictable.

Removing the basket assembly was easy; replacing it was tricky; if the basket fell to the magnet assembly, one or more of those screws could have penetrated the cone.

The situation was left alone as the buzz is far below the useable range of the speaker.

114hz is 4.72 octaves below 3khz and

5.45 octaves below 5khz.

There wasn't any point to measuring the response below 1khz as these tweeters are usually first order filtered at 5khz, although I have seen some labels on them stating 3khz as a lowest limit.If by the reference speaker you refer to the one shown on page `1, dated April 25, 2012I

Equipment used for the following spectrum analyses is:

PicoScope Model 2205A

Behringer ECM8000 microphone

ART (Applied Research & Technologies) TUBE MP Studio microphone preamp.

THD of the arbitrary waveform generator is 0.081% (measured on a Tektronix AA501 analyzer

and that of the preamp is < 0.1% (as per mfr. specs.)

These analyses were run with 1 watt to the BR3 and at a distance of 1 meter.