A pair of 5 inch Wharfedale midrange units from the early 60's.
I believe these were used in the "W" series of floor standing and bookshelf systems.
The annuli (surrounds) have been replaced.
|Two impedance curves, with
(red) and without (blue) equalization. I highly question the
usefulness of such a circuit as the books showing it do state that the
equations will get one into the ball park, so to speak but in the end,
one needs a method of determining the driver impedance at any
frequency. If one has such a device, then equalization seems
unnecessary as the filter (crossover) components can be calculated at
crossover frequency impedance.
Achieving then right resistance and capacitance is tricky and a good ohmmeter and capacitance meter are necessary. If not available, an impedance bridge is definitely required. I have all the above with an accuracy of better than 1% and still, it took all of 30 minutes of trial and error to get the right component values to get the red curve shown at the right.
|Curves 44 & 45 are measured with a
Zobel impedance EQ circuit, which consists of a 5 uF cap in series with
a 20 ohm resistor, the series pair being in parallel with the voice
The two upper jagged traces are un-gated room responses at 1w1m, referenced @ 1khz where the load impedance is 11 ohms. (3.32v); the lower smooth traces are derived with a gated response which removes the room reflections. It simulates an anechoic field.
The speaker and mic are placed 1m apart and midway between floor and ceiling. It takes about 3 mS for a pressure wave to travel 1m. Since the first reflection is 2.44m in distance (the two equal sides of an isoceles right angle triangle, it takes 7.3 mS for a pressure wave to cover that distance. The mic is switched on and the signal applied. The mic is switched off about 5 mS later which is 2.3 mS before the first reflection reaches the mic.
The lower trace is the simulated anechoic response. The 6 dB drop from the other trace is due to room reflections as well as a little insertion loss due to the eq circuit.
Response traces were run with and without the eq circuit and no difference was noticed. However, using the driver impedance at the crossover frequency will reduce or eliminate any difference at that point. Keep in mine also that this trace is for a single driver. Had a two way network been measured, a dip or bump at the crossover point would most likely have been seen if the driver impedance as stated been used.
Further discussion on this topic will rapidly get well beyond the scope of this page.
Back to the Wharfedale Index
Back to the loudspeaker main page