The restoration of a Wharfedale Co-Axial 12.  Actually, one was restored and repaired a second time.  Initially, both were installed into the 6 ft^3 corner cabinets.  All went well for a few days but one of the voice coil windings broke loose from the form.  What I had neglected to consider was the age of the adhesive.  Initially, all I did was replace the foam annulus but still the tweeter had to be removed to allow the 1.75" voice coil to be shimmed.  It took considerable thought as to how to go about this endeavour.  It was decided not to remove the wires but to temporarily fix the tweeter to the larger cone.  It worked.

 

 

 

 

 

 

 

 

Here's the cone from the unit that was damaged after replacing the annulus.  The coil form damage was done whilst disassembly as the wire was a twisted nightmare in the gap.  
The tweeter coil can be seen here with the inner foam spider.  The screw protruding also holds the plastic device to which the tweeter cone is attached.  The screw is a shoulder screw, threaded on both ends.  First, a phenolic washer is inserted onto one end, then  the foam spider and another phenolic washer.  The screw is then screwed into the threaded hole in the pole piece.  Next, the tweeter assembly is carefully mounted onto the upper threaded end followed with a washer and nut.  This was difficult as the tweeter wires were still connected to the larger cone.  Initial assembly at the factory would have been much easier.
The first coil still on the winder.  I didn't have any access to a finer wire, so I was forced to wind the coils in 8 ohms.  This wire, I think, is AWG 34.

AWG 36, if memory serves me well has since been obtained so two more coils will be wound in 16 ohms.  This will also increase the BL product as thinner wire will allow more turns in the 11 mm coil height.  The BL product is the product of the flux density times the length of wire in the field.

The coil form is 0.002" brass shim stock.

The whole coil winder.  Crude, yes but it works very well.  A coil such as this can be wound in about 30 minutes.   It's done under an illuminated magnifying glass whilst carefully counting turns. 

The aluminum core was machined at a machine shop to 1.752"; the pole piece diameter is 1.731", measured by several different methods without taking apart the magnet assembly. .  The actual coil diameter is 1.754" so shim stock was added.  The gap measures 0.0475".  This leaves about 0.012" gap between the inner surface of the coil and the outer surface of the pole piece.

One of the coils atop the magnet assembly.  The two gaps can easily be seen.
One of these coils is in one of the co-axials; the other is in a glass bottle.  If I get lucky, I may find another co-axial that has a damaged coil and use the second 8 ohm coil, resulting in two 8 ohm units.  Another option is to make both these 16 ohms but I hate to alter the original speaker as it has an aluminum coil.  It would be interesting to compare a 16 ohm unit with a copper voice coil to the original.

I may get really lucky and find a co-axial with a badly damaged cone but a good coil and use that 16 ohm coil.

The two completed co-axials.  The one on the left has the copper coil, the colour of the wires can be seen around 5 & 7 o'clock just outside the tweeter frame.
The copper wires can be seen easier here, especially the one at 4 to 5 o'clock..
The label, or what's left of it.

The letters, 'XIAL' are all that's left to describe the unit.

One of the finished 6 ft^3 corner cabinets in my living room.  The co-axials are not in these.  Instead, they contain a later model, W12RS with a ceramic magnet.

Later, they will be replaced by a W15FS, of which I have 4 and two have new original cone and coil assemblies I bought from Wharfedale in '71 and had them shipped from England.  the foam is still ok as Wharfedale changed that prior to using the half roll surround.

The restored Super 8's and Super 3's are upward mounted in the upper cabinet, which is open walled.  Essentially, it's a framework.

I design and build my own speaker cabinets.  The bass cabinets have an f3 of about 35 hz. with these 12 inch units; the W15FS will yield an f3 of just under 30 hz and with an increase of about 4 dB in its passband up to 800 hz due to the larger effective diaphragm area.  The vent on the bottom will have to be corrected.

A description of that which is in this piccy is at the bottom of this page.

 

Of the two graphs below, the upper pair of  response curves are a continuous sweep, no gating whereas the lower pair is gated.  The distance of the mic to speaker being 1/2 meter is very close to near field, hence the similarity between the two as well as the relatively smooth response in the bass below 300 hz.  At this distance, 1/2m, room reflections are less influential.

The speaker was mounted in a sealed box of 3 cubic feet internal volume.  This was to negate doublet action caused by the rear wave from the cone being 180 degrees out of phase with that from the front.  For a 12 inch speaker, this will have the greatest effect beginning at about 567 hz, whose half wavelength is 1 foot.  Under anechoic conditions, this would cause an attenuation of 6 dB/octave.  In a room, the response below 567 would be very jagged with peak to trough differences of 10 dB or greater, depending on the room dimensions.

The lower pair stops at 250hz due to the gating cycle.  At that point, the incident and reflected waves reach the microphone too close together preventing the system from differentiating between the two.  At `1. distance, this happens around 325hz.  At greater distances, the cutoff rises due to the decreasing difference in the paths of the incident and reflected waves.

1/4 watt at 1/2 meter will produce the same SPL as 1w/1m.

On a few instances, gated response curves were run on various speakers at 1/4w-1/2m and 1w-1m.

Both traces were identical above about 450 hz.

Co-Ax A (black) has the copper voice coil.  I suppose the lower high frequency response (above 1700hz) may be partially due to that, yet that same copper coil speaker has a smoother response between 1k and 2k.  The attenuator potentiometer may also have an influence here despite both being turned fully clockwise.  There may have been a difference in contact resistance at the full clockwise position.  Consideration was given to fully remove the potentiometers but it was thought to be a waste of time and effort.

 

 

 

 

This diatribe is placed here as it has nothing to do with Wharfedale but thought a description of what else is in that last piccy was warranted.

The tape recorder is an AKAI GX4000, with glass heads that purportedly last for 150,000 hours.  That's 17 years of 24/7 playing.  There's a few other AKAIs here along with a half dozen Revox A77's, two of which are half track and high speed, 15 ips.  Another is a full track.

The radio is an old Hallicrafters I restored.  It has a bandwidth of 540 kHz to 33 mHz.  The speaker cabinet under the end table houses a 10 inch woofer with a 3 inch voice coil.  There are actually two cabinets there, one under the table and one behind it, latched together. It's 3.2 ft^3 internally and is the woofer for the Dayton full range in the satellite.  That Dayton is only 3 dB down at 20 kHz and quite flat to 17 kHz.  The octagonal cabinet, also one of two, is solid 0.75" purple heart.  Originally, there were other cabinets atop these stands that housed a Trinity 6 by HiVi but I also had to try these.  The crossovers for the Trinitys is in the base; the Daytons are bi-amped by Adcoms and a Behringer electronic crossover, with 4th order filters at 200 hz.   I got away from  passive networks years ago.   There are 3 separate stereo systems in this room and a Wharfedale mono three way built in 1957 by a half-brother when I was 15.  It's powered by an EICO HF-60 mono preamp and HF-30 mono power amp he used way back then.  Both have been restored and both are TOOBZ.

 

Back to the Wharfedale Index

Back to the loudspeaker main page