Passive Crossovers -- YES part 2

[SoundMods]

Attached are the photos of the upgraded crossovers. You’ll note that all but two capacitors were exclusively polycarbonates (ear-tested for the project) with hermetic Teflon bypass caps in the tweeter networks and WIMA polypropylene used as much reduced value shunt in the mid-range networks. You will note the small form factor Electronic Concepts caps designed for Mil-spec switch-mode power supplies and glass hermetic polycarbonate and Teflon caps designed for only God knows what. And my favorite tweak is the use of clamping resistors on each every capacitor and inductor to assure that those resonant reactance’s do not have snow-ball’s chance in hell of enabling those resonances.
BTW – Wilson Audio also bought on to the clamping resistors idea back in the 1980s following a dinner that I had with Dave Wilson discussing all things audio. It works and works well.
So – my bragging aside – the owner of the B&Ws gave permission to quote him as follows:
“I first met Walt D'Ascenzo at the CAF Show a couple of years ago. While chatting about audio, I complained to him about my problem with my EAR 834P phono pre-amp, to which he offered a trio of matched 12AX7 tubes for 20 bucks. After installing them, the pre-amp came alive. After that, I bought one of his hot-rodded cartridges, and two more after that. I'm now listening to his vintage hot-rodded AKG 8E whose cantilever is filled with a stiffening compound with virtually no mass, an innovation of Walt's. It's sublime. Of course, all of the weaknesses of my components were revealed by this new clarity. I sent Walt my EAR 834P to have it checked out. A good thing I did. It came back performing like something superior to the original. He also loaded it with a couple of different tubes that were part of its overall improvement.”
“The next operation required was my B&W 802 Series 80 speakers from 1981 which have some renown from the Matrix series. I knew they were good in their day, beautiful solid cabinetry and high quality drivers, plus the midrange heads had been upgraded to make them 802F's now. They just weren't producing what I believed was possible. So, I shipped just the crossovers to Walt, and he replaced the critical components and added a few of his own. With his half-century of experience, he figured out a way to avoid the pitfalls associated with these upgrades. The speakers are finer than "new and improved" models that are more sizzle than steak. Hearing my system is thoroughly gratifying because of Walt's intervention. If you have audio components that have lost their luster, Walt is the man who will restore and improve them. You'll also get a detailed, bound account with photographs of the operation. He's rabid, alright!”
“BTW, the tweeters are reaching the stratosphere without force or distortion. Zero ear fatigue, even listening to modern composers who tend to get shrill in the upper register. The bass is rich and natural, with beautiful transparency overall.”
“There is also nothing electronic or fizzy coming out of the speakers- it sounds VERY real & live, and the presence of individual instruments is clearly delineated in one recording after another... I think I'll be at CAF this year- as an exhibitor!”
I also had to coach him to add restrictive vent holes in his mid-range modules to relieve the cavity resonance. His comment after the drilling:
“Here's my take, and hoping it's not a placebo effect, it's akin to taking the lid off a jar containing sound. The result is a completely balanced sound field. Hard to believe two quarter-inch holes could make that much of a difference. I'm thinking about my skull next!”
I also performed the same upgrades on a pair of B&W 801s 20-years ago with the same results.
The owner’s name is Matt Lass and he is available at: mattlass@hotmail.com

Attachments

[chris1973]

Hey Walt,

So I'm somewhat of a beginner, but I have designed and build several crossover networks.

I'm always looking for ways to improve my work.

Can you explain in a little more detail the concept of clamping resistors?

Thanks,

Chris

[SoundMods]

Hi Chris! Whenever you have a capacitor and coil in series or parallel you end up with a reactance that acts like a "tuned" circuit that resonates. These resonances drove me crazy some 20-years ago when I was dialing in a pair of speakers. When I did the math, the combinations created the very issues that I believed were mechanical. WTF?! So going back to basic electronics theory I decided to try to prevent those resonances from occurring. It worked. When Dave Wilson and I first discussed the damping resistor scheme he poo-pooed it. Then two weeks later I get a call. He put one his crossovers on the the bench, installed resistors, and he remarked how the ringing issues he was trying to get rid of stopped. He used signal pulses and a scope to watch the results. He gifted me with his Wilson
LPs.

[chris1973]

Hey Walt,

So this wouldn't be an issue with a 2nd order filter, because there is only a single component in series with each driver ( if you don't count resistors), except for the woofer circuit which has a zobel in it ( Essentially putting an inductor parallel with a capacitor).

Would this be a place where a clamping resistor would be any benefit?

[chris1973]

I think I got that wrong! In a second order filter, even with a zobel, there is no capacitor inductor combination (whether series or parallel) that would present that problem.

It's only when you have 3rd or 4rth order filters that it becomes relevant, right?

[SoundMods]

Actually, you are talking about a first order 6-db/octave crossover. Second order (12-db/octave) has a high-pass capacitor with a shunt inductor, and a low-pass inductor with a shunt capacitor. Just talking about a two-way speaker right now but in the case of the high-pass and low-pass sections you have a series resonant circuit in action in both the high-pass and low-pass sections. Further -- even with a first order crossover it is easy to forget the speaker voice-coil inductance.

[GaryB]

. . . You’ll note that all but two capacitors were exclusively polycarbonates (ear-tested for the project) with hermetic Teflon bypass caps in the tweeter networks and WIMA polypropylene used as much reduced value shunt in the mid-range networks. You will note the small form factor Electronic Concepts caps designed for Mil-spec switch-mode power supplies and glass hermetic polycarbonate and Teflon caps designed for only God knows what. And my favorite tweak is the use of clamping resistors on each every capacitor and inductor to assure that those resonant reactance’s do not have snow-ball’s chance in hell of enabling those resonances.

Walt,
Thanks for sharing your techniques for crossover tweaking. I have a couple of questions if you don't mind.
First off, where do you get the EC caps? I noticed a few for sale on ebay but otherwise they seem hard to get. EC does have a website and I gave them a call but they're not particularly set up for small buyers. Their price for new caps in small quantity was really high even by audiophile standards.
Second, I'm curious if you have any rules of thumb for the clamping resistors. I would imagine that one just takes the impedance of the cap or inductor at the crossover frequency and multiplies it by ~100x and that would take care of it. From the pictures, it looks like you use 470 ohms a lot with the occasional 5K ohm thrown in there.
Thanks,
---Gary

[SoundMods]

Hi Gary!

There was time -- long-long-long ago when Electronic Concepts dumped their out-of-tolerance production and the New Jersey sales office sold them for the manufacturing cost.

The DYI guys that new about it called it the "Candy Shop." That's done and gone now. Maybe the question to ask is what do they do with out-of-tolerance production? It would be sin to throw them away.

The caps you saw in the B&W crossover were scooped up from eBay. I keep a constant watch -- but I haven't seen any for some time. There are plenty fo that style in polypropylene (the EC model no. would have MP in the number as opposed to MC)

There are hermetic polycarbonate caps out there -- not EC -- but damn good. I have some large values that could be used in crossovers.

My original testing about 30-years ago (when I figured out what was going on) I determined an ideal 350-ohms across capacitors and 500-ohms across inductors.

Since then I found that it is not critical but you do not want to go lower than 350-ohms across caps or lower than 450-ohms across inductors. There were no 5-Kohm resistors used in the B&W. The gray silicon wire-wounds were actually 350-ohms. The "5" was part of the part number.

Having said that -- why polycarbonate? Polypropylene caps, even those made by EC are not as musical as polycarbonate caps. Even the best of the best still have an edge and glare to the sound that does not serve the music. Additionally, I had to be careful with the B&W crossovers since the owner attends live events on a regular basis. Mostly classical music without benefit of sound reinforcement.

Polystyrene caps can serve the music but they have to be mixed with other dielectrics or the sound can get dry with a somewhat strained quality. ??? I don't have a clue why that is, but that has been my experience.

[GaryB]

Walt,
Interesting. i should have paid more attention to polycarbonate caps. A few years ago I was restoring an old Threshold FET 10hl preamp, which used 10uf polycarbonate caps in the output. They weren't as transparent as I would have liked but bypassing them with a 1uf Hovland polypro and .1uf Vitamin Q gave a very nice balance. Flavor to taste.
---Gary

[SoundMods]

Well -- actually the output caps in the Threshold were polyester (mylar). The euro code was MKT. I looked at images on the WEB and the caps that I saw were the green Roderstein polyester caps with the MKT code. Roderstein polycarbonate caps have been typically color-coded red and their polypropylene caps have been color-coded blue.

Although mylars get the midrange right the treble can be strident and grainy. The only manufacturer producing equipment during that time, that I am aware of, that used polycarbonate coupling caps was John Curl designed Mark Levinson preamps.