Wharfedale Super 3

While there is plenty of information on the Super 3 in this website, this additional info started with my curiosity of the condition of some of these tweeters, most of which were given new voice coils and possibly new cones way back in the 90's when a friend owned and operated a speaker repair shop. The aforementioned cones and coils were supplied by Waldom Electronics, no longer around; probably assimilated by a corporate merger. The original Super 3 voice coils were wound with aluminum wire but the replacements are wound with copper. Only two of the tweeters noted in this page are known to be all original. They are the J1 and J2 units which were acquired from a friend in new York City in a swap for two other original Super 3 units. He sent me four, one of which had an open voice coil. All needed replacement annuli. In exchange for my repairing two which I sent back to him, he gave me the other two. I was able to repair the one with the open voice coil. These are the J1 and J2 units, the J1 being that with the repaired voice coil. This was done by unwinding one turn of wire from the coil; the break was at the junction of the coil and the cone's apex. This was an easy fix, provided one has aluminum solder and flux.

Aluminum oxidizes immediately upon exposure to air, that's what keeps it shiny. The oxide layer is probably only one or two molecular thicknesses but ordinary solder flux isn't strong enough to break it down.

Anyway, after making measurements on these fellas, this page evolved.

There are ten Super 3 units here. The P1 and P2 units are being used in my Wharfedale corner cabinets. (Photo 3 at the bottom of this page.) Another is all original and is used in a three way system built by a half-brother in 1957. It is unknown if the P1 and P2 units have aluminum voice coils. There is a possibility that they have coils I wound with AWG36 copper wire and are underhung which may account for their higher sensitivity. see FIGURE 4

The four NT units are described later.

All figures are linked to a larger image.

Wharfedale/SUPER-3___again/Nutone1234/Z RED-NT1 GREEN-NT2 YELLOW-NT3 BLUE-NT-4-s.jpg

These four units were repaired way back in the early 90's, possibly earlier.

From left to right, they are NT1, NT2, NT3 and NT4

The NT1 and NT2 have the aluminum voice coils but the cones and dust caps are Waldom replacements, indicating they were restored in late 80's. Later, Waldom changed the cones to that shown in NT3, the one with the letter A on the magnet. This cone and annulus are one piece. This was not a good idea since the annulus will have resonances and induce reflections back into the cone. The result if this can be seen in the yellow and orange response curves in figure 2. The NT4 may have an original aluminum voice coil as the solder eyelets on the cone appear to be original. The purple annulus was added in July 29, 2014; it's marked on the back of the magnet. It is 15W; the dcr is 10.2W; the others are all under 6W dcr. The foam annuli on NT1 and NT2 were probably acquired from a dry cleaner when they used foam on metal hangars. Briggs also used such foam in the early years.

The variations in their resonant peaks is very likely due to the different dust caps, annuli and even the amount of glue used at the cone to voice coil junction.

FIGURE 1

Impedance curves of the four NT units.

Red-NT1; Green-NT2; Yellow-NT3; Blue-NT4

One possible reason for the variance in resonant frequencies is the inner centering disc in the voice coil. These foam discs were cut by hand and their circumferences are irregular, causing a frictional difference. This is quite obvious in the green curve with all its ripples.

These variations were not considered worthy of remedy as the speakers are second order high pass filtered at 5khz.

Wharfedale/SUPER-3___again/Nutone1234/1w1m RED-NT1 PINK-THD GRN-NT2 LtBlue-THD YEL-NT3 ORN-THD DkBlue-NT4 GREY-THD-s.jpg

FIGURE 2

1w1m response curves of the four NT units.

Red-NT1; Green-NT2; Yellow-NT3; Dark Blue-NT4

THD curves

Pink-NT1; Light Blue-NT2; Orange-NT3; Grey-NT4

As mentioned earlier, the NT3 yellow and orange curves stands out. Note the wide peak centered at 2khz and high peak in the associated orange (THD) curve.

The other three response curves are very similar.

FIGURE 3

Impedance curves of P1, P2 and P3.

The green curve of P2 parallels that of P1, red, above 1khz. These are the two used in my corner cabinets.

Again, the different resonant peaks are likely due to the amount of glue used at the voice coil to cone junction, the amount used for the dust caps and more likely due to the friction of the inner foam disc used as the centering device for the voice coil. Those foam discs were hand cut before I came up with the idea of using two one inch washers bolted together. They are described on this page, photos 5 and 6.

Wharfedale/SUPER-3___again/P1P2P3/1w1m RED-P1 PINK-THD GREEN-P2 LtBlue-THD BLACK-P3 GREY-THD-s.jpg

FIGURE 4

1w1m response curves of P1, P2 and P3.

Red-P1 Pink-THD Green-P2 LtBlue-THD Black-P3 Grey-THD

Pink is actually violet. It parallels the red curve. The two upper THD curves, green and light blue correspond to red and violet response curves. The black and grey curves are those of the P3 unit.

Again, the P1 and P2 units probably have voice coils wound with AWG36 and are underhung which may account for their 10dB higher sensitivity above 3khz.

Their nominal impedance at 5khz is 16W. This is the result of using a thinner wire with a higher resistance and is why the couls had to be underhung, where the winding height is equal to the thickness of the top plate. This keeps all the wire in the highest part of the magnetic field. This would be a problem with diaphragm excursions greater that about 0.5mm. Tweeter diaphragm excursions are extremely small.

FIGURE 5

Impedance curves of J1, red and J2, green.

Again, the difference in resonance peaks may be due to the friction of the foam centering discs. This is not a problem as the impedance of each are within 1.5W above 750hz and these units are second order high pass filtered at 5khz.