Wharfedale Crossover Circuit Puzzle

[Chris F.[_2_]]

Someone brought me an old Wharfedale model W-90 monitor speaker, a massive
old beast with two highs, two midranges, and two 15" woofers. The problem
was that the crossover circuits are damaged; two 50 ohm potentiometers were
burned up, and someone had attempted to bypass them, getting the wiring
mixed up in the process. I've drawn up a diagram of the components,
numbering each component lead rather than show how they were hooked up. You
can view the diagram here:

http://s282.photobucket.com/albums/kk275/packrat79/?action=view&current=WharfedaleSpeaker.jpg

I'm hoping that someone can figure out the arrangement, and rather than have
to draw up a schematic for me, you can just point out which numbers to
connect.
Some notes:
-the connections between the high and midrange speakers appear to be
original, though I cannot be certain
-the woofers are presently connected in series (each measures 4 ohms)
-values of the other components are as follows:
C1 - 25 uF 50VDC
C2 - 12 uF 25VAC (much larger than C1)
L1, L2 - values unknown, but both measure about 1 ohm
VR1, VR2 - 50 ohms, 5 watts wirewound

I played around for a while and was able to achieve a decent sound with the
following configuration: 1-3-8-13, 2-18-24, 22-23, 4-5, 7-17, 9-10, 11-19,
14-21. This doesn't seem right though, since only two leads of each pot are
used, and L2 isn't used at all. In fact, it didn't seem to matter if L1 was
bypassed.
Any help with this would be appreciated.

[Swanny[_2_]]

On 2/04/2011 5:45 AM, Chris F. wrote:
> Someone brought me an old Wharfedale model W-90 monitor speaker, a massive
> old beast with two highs, two midranges, and two 15" woofers. The problem
> was that the crossover circuits are damaged; two 50 ohm potentiometers were
> burned up, and someone had attempted to bypass them, getting the wiring
> mixed up in the process. I've drawn up a diagram of the components,
> numbering each component lead rather than show how they were hooked up. You
> can view the diagram here:
>
> http://s282.photobucket.com/albums/kk275/packrat79/?action=view&current=WharfedaleSpeaker.jpg
>
> I'm hoping that someone can figure out the arrangement, and rather than have
> to draw up a schematic for me, you can just point out which numbers to
> connect.
> Some notes:
> -the connections between the high and midrange speakers appear to be
> original, though I cannot be certain
> -the woofers are presently connected in series (each measures 4 ohms)
> -values of the other components are as follows:
> C1 - 25 uF 50VDC
> C2 - 12 uF 25VAC (much larger than C1)
> L1, L2 - values unknown, but both measure about 1 ohm
> VR1, VR2 - 50 ohms, 5 watts wirewound
>
> I played around for a while and was able to achieve a decent sound with the
> following configuration: 1-3-8-13, 2-18-24, 22-23, 4-5, 7-17, 9-10, 11-19,
> 14-21. This doesn't seem right though, since only two leads of each pot are
> used, and L2 isn't used at all. In fact, it didn't seem to matter if L1 was
> bypassed.
> Any help with this would be appreciated.
>
>

1st order filters with variable attenuators.
The inductors may well be different values, so I'd suggest measuring
their inductance.

1: 3, 8, 15
2: 6, 12, 18, 20, 22, 24
4: 13
14: 10
11: 19
9: 5
7: 17
16: 21, 23

[Swanny[_2_]]

On 2/04/2011 5:45 AM, Chris F. wrote:
> Someone brought me an old Wharfedale model W-90 monitor speaker, a massive
> old beast with two highs, two midranges, and two 15" woofers. The problem
> was that the crossover circuits are damaged; two 50 ohm potentiometers were
> burned up, and someone had attempted to bypass them, getting the wiring
> mixed up in the process. I've drawn up a diagram of the components,
> numbering each component lead rather than show how they were hooked up. You
> can view the diagram here:
>
> http://s282.photobucket.com/albums/kk275/packrat79/?action=view&current=WharfedaleSpeaker.jpg
>
> I'm hoping that someone can figure out the arrangement, and rather than have
> to draw up a schematic for me, you can just point out which numbers to
> connect.
> Some notes:
> -the connections between the high and midrange speakers appear to be
> original, though I cannot be certain
> -the woofers are presently connected in series (each measures 4 ohms)
> -values of the other components are as follows:
> C1 - 25 uF 50VDC
> C2 - 12 uF 25VAC (much larger than C1)
> L1, L2 - values unknown, but both measure about 1 ohm
> VR1, VR2 - 50 ohms, 5 watts wirewound
>
> I played around for a while and was able to achieve a decent sound with the
> following configuration: 1-3-8-13, 2-18-24, 22-23, 4-5, 7-17, 9-10, 11-19,
> 14-21. This doesn't seem right though, since only two leads of each pot are
> used, and L2 isn't used at all. In fact, it didn't seem to matter if L1 was
> bypassed.
> Any help with this would be appreciated.
>
>

Correction, I missed the bit about the woofers being 4R each

1: 3, 8, 15
2: 6, 12, 18, 20, 24
4: 13
14: 10
11: 19
9: 5
7: 17
16: 21
22: 23

[Peter Larsen[_3_]]

Chris F. wrote:

> Someone brought me an old Wharfedale model W-90 monitor speaker, a
> massive old beast with two highs, two midranges, and two 15" woofers.
> The problem was that the crossover circuits are damaged; two 50 ohm
> potentiometers were burned up, and someone had attempted to bypass
> them, getting the wiring mixed up in the process. I've drawn up a
> diagram of the components, numbering each component lead rather than
> show how they were hooked up. You can view the diagram here:

> http://s282.photobucket.com/albums/kk275/packrat79/?action=view&current=WharfedaleSpeaker.jpg

> I'm hoping that someone can figure out the arrangement, and rather
> than have to draw up a schematic for me, you can just point out which
> numbers to connect.

This requieres that you understand it, painting by numbers alone doesn't do
that. However we have loudspeaker items BASS, MID and TREBLE. We have
cross-over components L1, L2, C1, C2. As suggested by Swanny that seems to
suggest a 1'st order three way. If you want to cure the patient, the
loudspeaker box, from recurring failure you need to do something more
drastic than getting the old cross-over working again. Transform it into a
LO, HI electonic x-over with HI containing MID and TREBLE and a _suitable_
cross-over, ie. 6/12, 12/12 or 12/18 and it starts looking like a sane
concept.

> Some notes:
> -the connections between the high and midrange speakers appear to be
> original, though I cannot be certain
> -the woofers are presently connected in series (each measures 4 ohms)
> -values of the other components are as follows:
> C1 - 25 uF 50VDC

in series with MID makes sense.

> C2 - 12 uF 25VAC (much larger than C1)

in series with ATT in series with TREBLE makes sense. Signal into attenuator
and out via wiper makes sense, that would add resistance to the TREBLE
impedance and then the component would suggest a reasonable treble
cross-over, with 8 ohms it would be at some 1500 to 2000 Hz, for a high
power system using a first order it is mightly lowish and too close to
probably unit resonance for proper functionality without compensation of the
impedance unlinearity at unit resonance. Units with ferrofluids are nice
because they can have near linear impedance in the resonance region, imo the
strongest point in favour of them.

> L1, L2 - values unknown, but both measure about 1 ohm

Describe them better, size, weight, use a tone generator and determine
matching resistance at a couple of frequencies.

> VR1, VR2 - 50 ohms, 5 watts wirewound

Replace with ceramic 11 watts fixed resistors in a suitable configuration.

> I played around for a while and was able to achieve a decent sound
> with the following configuration: 1-3-8-13, 2-18-24, 22-23, 4-5,
> 7-17, 9-10, 11-19, 14-21. This doesn't seem right though, since only
> two leads of each pot are used, and L2 isn't used at all. In fact, it
> didn't seem to matter if L1 was bypassed.
> Any help with this would be appreciated.

Are you doing this for someone else or for yourself/your own band? - even if
there is insistence on a passive cross-over the technically best approach is
to (re)design a more suitable cross-over, as is it is quite possibly by
design that the attenuators blew. 12/12 with zobel networks is about the
simplest you can get away with if you also want decent total system
powerhandling. The current cross-over is either non-original or designed by
Wharfedales accountant either directly or indirectly.

Kind regards

Peter Larsen

[Scott Dorsey]

Peter Larsen > wrote:
>
>This requieres that you understand it, painting by numbers alone doesn't do
>that. However we have loudspeaker items BASS, MID and TREBLE. We have
>cross-over components L1, L2, C1, C2. As suggested by Swanny that seems to
>suggest a 1'st order three way. If you want to cure the patient, the
>loudspeaker box, from recurring failure you need to do something more
>drastic than getting the old cross-over working again. Transform it into a
>LO, HI electonic x-over with HI containing MID and TREBLE and a _suitable_
>cross-over, ie. 6/12, 12/12 or 12/18 and it starts looking like a sane
>concept.

I never looked inside a W-90. However, the W-70 had the midrange run
full range, and first order filters on the woofer and tweeter. The
crossover points were sadly obvious in spite of the wide filters.

Note that the original controls on the W-90 may have used L-pads instead
of pots, especially if they did a second-order filter.

W-90 manuals should be out there, there were certainly enough of them
made although I don't think they arrived on these shores much.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."