DIFFUSOR

 

A couple of weeks ago, a friend made mention of this 1947 Wharfedale on auction in England.  What caught his attention was the diffusor fastened to the baffle in front of the 10" unit and wondered how effective it was.  His curiosity wasn't the only one that was aroused, and here is what was found.

Having a W10-CS on hand, itg was mounted in a 1 ft^3 enclosure and driven with 1 watt.  Gated frequency responses were run at a distance of 1 meter with and without the diffusor.  The results follow.

 

 

 

 

 

 

Unfortunately, this photo shows little of the diffusor's dimensions.  The cone diameter is known to be 7.25" so that was used as a reference scale.  This seta the diffusor diameter at just under 3".

The depth, being unknown was assumed to be at least an inch from the apex of the cone.  Had it been fitted to the pole piece by a long screw, the board across the front would not have been necessary.  This was a logical way to go as the W10-CS models didn't have a drilled and threaded hole in the pole piece and doing so would have meant disassembly of the speaker, somewhat impractical.

 

 

 

The diffusor mounted to a frontal cross beam.  The wood disc is a first attempt as those discs were laying around.  In a subsequent repetition of this experiment, it was replaced by a 2.875" disc.

The diffusor is about an inch from the cone apex.  Made of plaster, using a 5" kevlar cone from a broken speaker as a mold, the rounded end was added after the mold plaster had set.  It later broke off during sanding and was clued back on.  The cone used has a curvilinear shape.  the putty was added to finish the edge which chipped easily.  

 

 

These curves are those of a pair of SFB 10's acquired from the same friend mentioned earlier.  They are designated as SFB-10J1 and SFB-10J2.

The black and dark blue curves are those of the J1 unit  with and without the diffusor, respectively.

The grey and light blue are those of the J2 unit.

 

 

These 4 curves all belong to the W10-CSB, the same unit used in the above system.

The red and light green are with and without the diffusor, on axis and the dark blue and dark green are with and without the diffusor, 30 degrees off axis.

Oddly, the light green, 30 degrees off axis without the diffusor gives the flattest response between 1khz and 6 khz, the intended use.  However, a tweeter is definitely needed.

Considering the era, one such system would be in a room and in a corner, thus most listeners in that room would be off axis to some degree.

The seller showed two small cabinets, one being original Wharfedale and the other a duplicate.  Both contain an upward firing Super 3.  The inclusion of an original such unit, with attenuator control, certainly implies Mr. Briggs' knowledge of the necessity of a tweeter.  Perhaps it was an intended option.

The LP as we know it, didn't come to the public until 1948 prior to which, the 78 rpm was used and with quite a limited high frequency response, especially after several plays as the 3g to 5g tracking force wore the grooves fast.  Older 78's played on current (1948) phonographs were even more severely worn as they were played on acoustical phonographs with tracking forces of an ounce (26g) or so.

 

 

 

 

The following is an addendum to the above;  the diffusor is mounted to the pole piece

 

 

A close-up of the wood adapter glued to the pole piece.  The 8-32 nut is glued into a 223/64ths recess 

to accept an 8-32 machine screw, thus holding the diffusor in place. 

 

 

 

 

The diffusor in place.  The thread is a precautionary measure as the diffusor is front heavy.

 

 

 

 

 

This is a duplicate of the second one previously shown and is placed here for easy comparison to the next set of curves.

Curves 15 & 17 were run with the diffusor mounted to the front baffle.

 

The black curve is run without the diffusor and is almost identical, as it should be, to #14 (red) in the previous set.

The red curve is with the diffusor mounted to the pole piece.  The blue curve is the same as the blue one in the previous set for instant comparison.  All are run at 1w1m, gated and on-axis.

While the responses in all these curves is anything but flat, it isn't bad considering the peak to trough difference of +/- 3 dB.

Despite the dips in the blue curve (diffusor on baffle), the intended result of taming the upper mid frequencies appears to have been successful.  The dips at 2100hz and 4500hz will soften a female voice and violins & brasses, resp.

The peak at 2khz (red) looks worse that it may sound.  As to how the diffusor affects the listening experience is subjective.

 

Back to the Wharfedale Index

Back to the loudspeaker main page