Arrival after a long resistor journey
97.83.185.226 |
||
| Posted on September 29, 2020 at 18:27:06 | ||
|
Posts: 1587
Location: midwest Joined: September 6, 2006 |
I have built many speaker systems over the past 50 years, in many different configurations and at many different levels of cost. I am not qualified or equipped to measure them, so I have always depended on professional services for the crossover design and my own ears to evaluate them. A few were flops, but most have performed satisfactorily, some admirably, a few better than those commercial brands I've owned, such as Yamaha NS1000M's, B&W DM6's and B&W 802's retrofitted with very expensive outboard xovers. My current design has gone through several iterations involving two brands of midrange and two brands of tweeter over a period of five or so years. I moved up from the modest Peerless 830870 poly cone 4" midrange to the superb Peerless NE123W 4" paper cone midrange. Then I replaced the formerly Peerless/now Scanspeak D2608 soft dome tweeter with the excellent Satori TW29RN. These comprise the MTM tops of a 4-way system, but the bass bins and sub haven't changed much. So you can see I have invested considerable expense in selection of drivers. The same is true of crossover components. All of the versions were executed with superior parts. The current xovers employ film and foil capacitors such as Jantzen Alumen Z and Audyn True Copper, along with a few poly caps such as Mundorf Silver/Oil, Sonicaps, and so on. During all these trials I have also used many types and brands of resistors, in tweeter series and parallel connections. I have never been entirely satisfied with the sound of any of them. I tried everything from Mills to Mundorf MOX to Caddock MP930 (with heatsinks) to Duelund graphite. I even built huge bundles of Vishay/Dale 1/2 watt to achieve sufficient power handling. They all varied from a grey wash to muffled to etched. I suspect that all wirewound resistors employ the same resistive element because it's cheap and readily available: fine gauge nichrome wire. If anyone knows of one that doesn't, please let us know. I don't think any self-respecting audiophile would intentionally introduce a material composed primarily of nickel and chrome into the signal path, yet all these resistors contain many inches of those very metals. The graphite and thick film resistors also left me unsatisfied. Recently I tried a simple piece of Constantan wire sourced on eBay. It was inexpensive but well made with enamel insulation and a jacket of silk tubing. This alloy contains some copper but is still primarily nickel, confirmed by its silvery sheen after removing the enamel. Aside from the strategic difficulties of working with 32 gauge wire, it was also difficult to solder unless the iron was very hot and held in place for many seconds. I persevered and got better sound, but the high frequencies still had a coarse edge to them. Just in the past few days I received a small spool of 34 gauge Manganin wire. I have now converted the series tweeter resistors to Manganin, and I think I've found the best sounding resistor of all. The high frequency edge has completely vanished, leaving pristine clarity and smooth tones. This wire contains 84% copper, as displayed by its golden sheen, and is easy to solder. It is still somewhat difficult to work with because it's so very thin, but it's tough and resistant to breaking while being extremely flexible. The resistance is appx .49 ohms per inch, so the typically small resistance values for tweeter padding are only a few inches long, and I just left the wire hanging bare in the crossovers, which are outside the speaker enclosure and not subject to much vibration. I suppose I could try damping them with something, but for right now I'm enjoying the clearest highs I've ever heard from my system. Gone is the artificial edge and hash of high frequencies that I've heard from too many speakers, including my own. This is not a tweak for those who like to improve their system by replacing outlet covers, and it will not transform your system. It is a refinement which requires some courage, some knowledge, some basic soldering skill, and a lot of patience. It does not involve much money. Ten meters of wire from Bulgaria cost me $12 and, with planets perfectly aligned, it arrived in only 16 days. I've used about two feet of 33. I have plenty left over if anyone wants to try it for a dollar. If you've read this far, you deserve a reward. Send me an email with your address and how much you need, and solemnly promise that you will actually use the stuff and report the results back here. Peace,
|
|
