You are reading the older HTML site
Positive Feedback ISSUE 6
Goertz Crossovers: Cleaner Cleans, Louder Louds
Having recently installed a pair of loudspeaker crossover networks in my favorite speakers, crossovers using Goertz Alpha-Core "Air-Core Foil-Wound Inductors," and hearing for myself the real difference they make, I thought I'd try to get down in writing my impressions, following the old saying, "I never know what I think until I see what I say." Some serious person said that. Maybe it was M.I.T.'s Professor of Logic, Norman Crosby? I can't remember. It means, "Writing helps clarify one's thoughts."
People in the audio industry estimate it's been a few years since the better loudspeaker manufacturers "all" began using "Air-Core Foil-Wound Inductors," which I'll just call "ribbon-coils" to differentiate from conventional wire-wound types, which I'll call "wire-coils." The main reason is, "They sound better." I can vouch for that.
How do they sound? Well, they don't make any higher highs, or lower lows. They do make cleaner cleans, and louder louds, too. I guess that means a loudspeaker with a ribbon- coil crossover seems to play with less distortion, and can sustain that low distortion for a noticeable while as the music plays louder and louder. So, if you want a Do-It-Yourself project thats likely to have good results, one that just about every audiophile who knows a little about loudspeakers would agree is likely to help your speaker, you might replace your standard wire-coil crossovers with a set of new, improved, better than ever, salt-free, low-cholesterol ribbon-coil crossovers. For more information see http://www.alphacore.com.
There are many theoretical reasons why ribbon-coils should outperform wire-coils. They have much less skin effect and FM distortion, along with reduced power loss, lower stored charge, flat inductance reactance, and no saturation distortion, according to a promotional "white paper" from Alpha-Core, Inc. I'd like to go into each of these briefly, and, as a layman with no formal training in electrical engineering, I hope you'll bear in mind that each of my explanations is only an attempt at finding a better verbal approximation, or metaphor, for a tricky electrical concept. As a writer, my job is to instruct and entertain, and while I have nothing against electrical engineers, their writing can be pretty conventional and formulaic. It is not often entertaining, and if youve no shared vocabulary, it can be pretty opaque. So I'll try to put things into plain language, crack a joke or two to keep you reading, and make sense of why one kind of inductor can perform audio tasks better when compared to another type of the same value.
Negligible skin effect: below 100 kHz, ribbon-coils have many orders of magnitude less measurable skin effect when compared with conventional wire-coils of the same values. (This is according to Alpha-Core.) We know that a solid wire, like hook-up wire, has some electrons that run along the surface, or skin, of the wire and other electrons that have to push their way through the center. (If the surface is "the skin," is the center "the flesh?") Of course it takes longer for the center electrons to get through, having many collisions to endure. If you took a standard length (say, one meter) sent a pulse through, and looked at it on a scope, it would have one configuration; but if you similarly pulsed a 100 meter spool, you'd see a change in the configuration that represented a greater length of time for the center electrons to catch up with the surface electrons. This delay is called "skin effect." Most round wire-coils, or inductors, have heavy enough gauge and enough length to generate considerable skin effect. A copper ribbon foil is only 0.0015 inch thick; that's 1.5 thousandths of an inch. It can be 1.42 to 2.00 inches in height. So it is almost all "skin." There is no part that is thicker than any other; hence, negligible skin effect. Wire-coils have electrons that are always trying to catch up with the faster electrons of the pulse. We experience that as smearing the sound, or (on a scope) the wiggly tail on the pulse before it settles into a flat line.
Extremely low stored charge: this characteristic is twofold; the first being like the charge of a "storage" capacitor. Certain capacitors are designed for "storage" of current to be released when called upon by loud transients in the music. Signal path circuits often have large value, fast re-charging (photo-flash) capacitors attached to the B+ and B- "rails." Secondly, the degree to which the wire-coil inductor acts like a capacitor, due to its insulation, can be significant. The degree to which the ribbon-coil inductor acts like a capacitor is "extremely low." This is a function of the insulation which is 0.0015 inch thick polypropylene film. One of the reasons polypropylene has become one of the preferred materials for the dielectric material in signal path capacitors is its low stored charge (or Q). One physical chemist I spoke to about this said the polypropylene molecule pivoted, and in its "up" position it could receive electrons all in a bunch, while in its "down" position it could release them all at once. So in real time it could accept and release electrons very quickly compared to most others. Other common dielectrics (say, polyester or Mylar) typically release in stages: a lot at first, some more in the second instant, a little less in the third, etc., until all the electrons have passed through. Furthermore, this phenomenon can be frequency dependent. With some dielectrics it means greater high frequency sizzle, and with others greater low frequency bloom. This means slower acting dielectric capacitors create something like "tracer bullets" (that weve seen in combat footage), except the tracers are slow-to-get-away pulse electrons. These are mixed in with the next pulse, in real time. We experience this as smeared, or indistinct, sound.
Inductive reactance, from 5 Hz to 50 kHz, is flat. That is to say, ribbon-coils act in a way that follows the ideal, or textbook, model of an inductor. In a ribbon-coil there is little irregularity that might cause a wire-coil to roll off at a point other than its theoretically modeled roll-off point. You might say, as Jennifer Crock has said, "A ribbon wire inductor acts like a pure inductor." That also means that it has less capacitance or resistance. As we've seen, wire-coil inductors at times act a little like a capacitor (high-pass filter), and at other times like a resistor (lowering gain). Ribbon-coil inductors, acting more like an ideal inductor (low-pass filter), make for more accurate crossover performance.
Reduced power loss. Highest space factor (ratio of conductor cross area to total cross-section) means lower DC resistance within comparable dimensions. Because its size, and the size of its conducting surfaces are large (if not thick), a ribbon-coil passes a great deal of current with negligible power loss due to skin-effect. Typically ribbon-coil inductors have lower measured D.C. resistance (in Ohms) than wire-coil inductors of the same value in mHs (milli-Henrys). Think of a ribbon coil as a spool of kitchen aluminum foil (with wax-paper on either side as insulation); and think of a wire-coil as an equal mass of foil crushed down (by hand) into as small a diameter as possible, which would resemble a pencil in diameter, and then rolled onto a spool.
This physical arrangement of ribbon-coils leads to cool operation even during prolonged high output power operation. Cool operation means less power dissipated as heat another contributing factor to negligible power loss.
High winding tension and vacuum fusing of faces provide high dimensional stability, effectively locking the conductors in place. Wire-wound inductors, which allow relative motion of conductors due to electromagnetic forces, will exhibit FM distortion by the process of reactance modulation. If youve ever placed your finger on a capacitor while music was playing through it, you should have felt it vibrating. Coils vibrate similarly. Round wire-coils can actually move, slightly; that is, the individual wires can move. This motion can cause FM distortion, as described above. Ive nothing else to add to their explanation, because I'm not sure why that would be Frequency Modulation distortion. Maybe some E.E. out there can help me out with a letter on this, and on any of my other errors.
Finally, they say: No saturation distortion due to air-core design. Inductors contain no magnetic materials. When I spoke to an audio buddy about this one he said when inductors "saturate" they just go a little crazy. It's like an inductor is doing fine, and suddenly it get filled up with electrons, and bah dah bing, bah dah boom, it starts to malfunction.
Perhaps the best analogy is magnetic tape. We know that magnetic recording tape, when used within its limits, can perform pretty well. With a tape recorder set up for a particular type of tape, say, "high-bias chromium-oxide" tape, we can expect excellent results. But even with such premium tapes, if we try to get more than the rated dB level out of a recording, the tape just starts to distort.
This reminds me of what happens to water as we heat it up. It takes one calorie of heat to raise one cubic centimeter (cc) of water one degree Centigrade (and that is defined as one BTU). At 100 degrees it takes a much larger number of calories, 400 plus, to change one cc of liquid water into steam. As we heat the water from 100 degrees liquid to steam (also at 100 degrees C) the molecules randomly bash around until their energy breaks the surface tension and they boil away as steam. The boiling point is called a "nodal point" where, though the water remains the same chemically, it changes state to a gas. Water is interesting as it has another nodal point where liquid changes to solid: ice.
The molecules in a wire-coil inductor seem to reach some "nodal point" and leap into a more energized state. Once they do that, as the current passing through them increases, they start to misbehave, and they start producing all of these various forms of distortion listed above. Before they hit their saturation point, the wire-coil inductors behave according to specification. Upon saturation, everything breaks down. I know this is basically incorrect, but it might be a helpful analogy with which to understand "saturation."
According to Alpha-Core, there is no saturation distortion due to air-core design. Inductors contain no magnetic materials. This is magnetic saturation. And, again, Im out of my depth on this one. Anyone who feels like taking a crack, please do. I'm hanging my limitations out here for public display, so please be civil. Im inviting helpful discussion, not flame-wars.
What we see in this list, if we can trust the Alpha-Core engineers, are a handful of ways in which ribbon-coil inductors outperform typical wire-coil inductors. Compared to wire-coils, ribbon coils have 1) negligible skin effect induced distortion, 2) extremely low stored charge, hence reduction of distortions generated by stored charge, 3) performance that closely follows the ideal or textbook model of an inductor, 4) much reduced power loss, 5) much cooler operation, 6) reduced FM distortion, and 7) no saturation distortion, because they never saturate. And that is why it is to your advantage to replace your old loudspeaker crossover networks with networks that use ribbon-coil inductors.
Crossovers are where loudspeaker manufacturers can employ the tricks of their trade, and so they are proprietary and closely guarded. If you own a particular manufacturers model number, you might ask them to build some new crossovers for you using Alpha-Core Air Core Foil Wound Inductors that you might install yourself. If you get a negative response to that request, you might go inside your speaker and draw up a schematic of your crossover. With a schematic, Madisound will build them for you. Or you might check on the various guys who offer mods, and see if they'll take on such a project. If you find none of this suits you, but you're interested in loudspeakers with the latest in state-of-the-art crossover design, you'll have to hunt through all the loudspeaker manufacturer's brochures until you find speakers that feature these crossovers.
Good luck! Once you get them in your speaker, you won't be disappointed. As I said before, you should get cleaner cleans, and louder louds.