Recap B&W Matrix 2 1986 issues

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
In 1986 I bought a pair of B&W Matrix2 speakers and had a lot of fun with them. Until... one day they smelled like something strange was burning.
A capacitor of 100µF and a resistor of 4ohm where burned on both speakers and burned the pcb locally. I brought them back to the store where I bought them, and they guy who sold them to me tried to repair them, but had not the right capacitors of 100µF, so he put in the left speaker 12+85= 97µF and in the right one 15+82= 97µF. For the resistor he used a different value.They worked, but i felt not happy with it. Now so many years later I want to repair them properly.

- what make are the best capacitors to use? (original is ALCAP)
- must I also recap the RCL PMT/2R capacitors?
- some of the smaller resistors are oxidized, must I also replace them?

I'm new on this forum. So if somebody can answer my question would be great.
Many thancks in advance.

PS: I use a Kenwood KA1100SD to drive them. I think this amp is to powerfull for them.

Mike
 
B&W has become a Mundorf shop, so if you want to stick with the factory choices, I'd go with that. MKP is the least expensive, and also smallest for the overall size.

I like the Mills 12W resistors. They are great sounding, and small for the wattage. Make sure you leave plenty of space around them so they ventilate properly.

Best,

E
 
If the original cap is electrolytic (ALCAP is), buy an electrolytic, you are effectively restoring the original spec and sound. Moreover you'll find rather difficult to accommodate a film cap in a space designed for an electrolytic. Note that the repair done was probably effective because 97uF is equivalent to 100uF once you factor in the caps tolerances.

For the RCL PMT/2R caps, as they are film caps, they don't age so there isn't any reason to replace them.

I would replace also the resistor that was changed after the burn because it is not the original value, and restore the original 4R value.

Ralf

There is not a too powerful amp, in fact the opposite is true: you risk to damage a speaker more easily from an underpowered amp driven to clipping.
 
I looked at the schematic you attached. Maybe someone else will chime in here, but those big caps aren't even needed if you are running a solid state amp.

It seems they are all part of an impedance moderating circuit, and should be safely removed completely. This is why they overheated as well. Reducing impedance means they were subject to a lot of current.

Of course, you should still check your amp outputs for evidence of DC, but I would be very much tempted to run without them. If you want to see what happens without them, I'd suggest getting DATS or Room EQ Wizard (free but need a jig) and measuring the impedance of each driver, and simulating it all in XSim. You'll better be able to tell what's going on when you look at the transfer function and the impedance charts.

Best,

E
 
Last edited:
Hey K!

Well, it seems to be "optimized" so they can play very well with tube amps. To do this they attempt to flatten the impedance curve. There are three sets of components that run in parallel to, and before the filters. A Zobel (CR) and two resonant circuits (CLR).

The tweeter circuit also has a time-alignment component. That's the pair of 0.49uH inductors crossed with 20uF caps. Lastly, there's a protection circuit, which has the diodes, relay and transistors (I would probably remove this entirely).

Also, the drawing order is quite odd. Inputs on the right, but then flow left before coming back right.

The filters themselves are extremely simple.

Best,

E
 
So did a little plugging around. I used the default 8 ohm perfect drivers to stand in, and I examined the power dissipation of the 3 impedance lowering circuits. Sorry I did not keep the same part designations, but you'll see them in the top left of the attached schematic.

attachment.php


I then set the amplifier to 50 watts output. Obviously, continuously this is actually a lot, but it's helpful to see what's going on. Look at the power dissipation of R1, R5 and R2.

attachment.php


This seems almost careless. Due to the resonant nature of the circuits they are attached to, R5 and R2 (on my schema) will have almost 30% more power through them than input! The original 11 watt rating was WAY too low. What I suspect happened is that the heat from the resistors fried the caps. So you are left with a dilemma. You should probably use about 2x that (minimum) but.... the problem is they will still dissipate that heat! Meaning, you have to find some way to keep the resistor heat away from the caps or you'll repeat the problem.

I really would not rebuild these as-is. Leave all three out, and measure the impedance using DATS or Room EQ Wizard (or whatever your impedance measurement tool of preference is) before thinking it is worthwhile. It may very well be that these three circuits don't do a thing for you.

Edit: After further analysis, it may even be possible to rethink the circuits, and come up with values that dissipate much less power, but again, not sure these circuits are worthwhile for solid state amps.

Best,


E
 

Attachments

  • BWMatrix_schema.jpg
    BWMatrix_schema.jpg
    49.5 KB · Views: 627
  • BWMatrix.jpg
    BWMatrix.jpg
    131.3 KB · Views: 586
Last edited:
Not awake yet. 80 W / 50 W = 60% more power! So basically with about 7 watts of power you will fry those resistors.

The lower resonant circuit is tuned to 50 Hz, where there is a great deal of energy in the bass. I'm honestly surprised these speakers lasted as long as they did.

Best,


E
 
Thanks to Wolf posting on the Parts Express forum, I was able to find another thread here at DIY which discusses the crossover, especially the protection circuitry, and a related crossover the Matrix 3:

Could a kind soul please break down this horrid Xover

I did not see the power simulations that I have here however. :) But you can see the flat impedance curve. This idea does not seem to have lasted very long at B&W. I can't find evidence of it being used much before or after. I strongly suspect they were going after the mid-power tube market with this series of speaker. Take for example a much more recent 800 D:

511B800fig1.jpg


As you can see, any pretense of flattening the impedance curve is gone.

Best,


E
 
Last edited:
Higher voltage is fine, but as I wrote, the real problem is the heat of the resistors. Double the wattage, and try to place the caps so they are no longer near them.

If you cannot separate the resistors and caps, increase the temperature rating on the caps, by a lot. If you have 80C or 85C try to get 100C caps.

As I wrote though, I think they are unnecessary unless you are using a tube amplifier. I'd actually just remove them entirely. :)

Best,

E
 
Do you mean remove al the caps? Or what do you mean? I tought the caps are needed for filtering the right frequencies.
The resistors are now 11W. I can replace them by 25W resistors, I want to mount them to the pcb with a gap, so they dont damage the pcb if overheated. (that is wat happend before.
 
Hi MTH!

That's what I'm saying. Those 2 very large caps (100uF and 1,000uF) are not part of the frequency filtering.

They are there to make life easier for tube amps. If you aren't running tube amps, you may put them in the trash and forget about them. This style of design kind of came and went. In 5 more years, B&W would not have put those parts in at all.

+1 for Lojzek's idea, is to get heat sink mounted resistors if you MUST restore them.

Bet,

E
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.