For the purist.

[Guest Brian_D]

The idea of restoring classic speakers (of any manufacturer) is to revive the high-quality sound of days long since overruled by ultra-high efficiency, low build cost "speakerish" boxes.

Keeping in mind the goal of restoring a classic speaker in order to enjoy sound exactly like the original equipment manufacturer intended, what relevant changes could occur in a speaker over time to change this "OEM" sound?

I'm talking about audible differences. If you want a tollerance in frequency response, say 1db, which is suposedly the smallest audible increment in audible energy. I want to stay away from theoretical affects that may affect the electromechanical or electrical responses and characteristics of components. For instance, iron core inductors will eventually realign their magnetic field if placed too close to another inductor when pure DC is present. Does this affect the inductance of the component and therefore change the crossover point of the circuit? Yes, but there isn't any pure DC signal in music, so we'll ignore that fact.

So the obvious one is foam rot, right? So cross that one off the list and let's move on.

Everyone knows a little about something other than audio... I'm hoping everyone can contribute to some degree. My contribution? My trade is printing and I happen to know a little something about paper, the material from which a great many AR drivers were produced. I'll the the topic started with that...

[mrdb]

Iron core inductors can become magnetized (DC) if mounted too close to a woofer magnet. Alnico magnets can lose gauss over time and with high AC fields from voice coils of speakers played at high levels. Non-polar electrolytic capacitors can become leaky and/or change value as they age. In addition, most speakers from the 50s, 60s, and (maybe) 70s used 50 volt caps. When used with the high-powered amps of today, the voltage rating of these caps might be exceeded, leading to accelerated failure.

Bob

[Guest Brian_D]

Paper was used in many drivers and is still used in woofers for cone material. It could be argued that a paper structure would, even under ideal environmental conditions, eventually loose it's rigidity and therefore change it's ability to perform a linear excursion as a result. The possibility of this failure would be exaserbated by poor environmental conditions, namely extreme humidity.

For all intensive purposes, paper is technically a flat product used for writing or printing. There are various other names given to formed wood celulose materials, such as fiberboard, chipboard, cardboard, etc. For the purpose of this message, I will replace the term "formed wood celulose material" with the word "paper."

Three issues can affect the rigidity of a formed paper shape. (cone, dome, whatever)

1. Celulose fiber composition: Is the fiber natural; containing lignen, chlora-based compounds and celulose which can break down over time, or is it clean; chemically treated to remove all contaminiants and leaving only pure celulose pulp?

2. Celulose fiber length: Is each fiber of celulose short; allowing for precise moldability, but with reduced structural integrity that decreases with repetitive bending, or is each fiber long; allowing for rudamentary molding while maintaining an outstanding strength-to-weight ratio that copes well with repetitive stressors?

3. Celulose fiber binder: Is the "glue" that binds the fibers a natural material; sensitive to actinic (ultraviolet) radiation and humidity, or is it one of the many synthetic binders that is impervious to even moderately poor environmental conditions including actinic radiation, humidity and extreme temperatures?

Let's start at the top. Lignen (the natural binder of celulose found in a majority of relevant paper-type celulose sources, i.e. Trees) is a natural source for binder, and therefore is often left in the processed celulose to aid in the binding of fibers when formed into it's final shape. Lignen, however is sensitive to actinic radiation (ultraviolet light, or direct sunlight) and will break down, leaving only celulose and chlora-material, which does not bind. Chlora-material (the chloraphyl and other catalyst chemicals left in celulose) is what acts as a catalyst for green plants that use Carbon-dioxide as fuel for cell growth (photosynthesis). One obvious drawback to leaving the chlora-materials in celulose is that it's yellow-green in color, and is not easily dyed. It also reacts very nastily to binders other than lignen, namely synthetic binders that release chlorine gas when exposed to chlora-materials. So the ideal celulose fibers would be free from lignen and chlora-materials so they may be easily bound with synthetic binders and not discolor and/or desentegrate when exposed to direct sunlight. If you've owned cheaper, imported speakers and had them in direct sunlight, you've probably noticed that the driver cones get "spongy" and loose their color. More commonly available is newspaper. If you leave it in the sun, it yellows and gets brittle. That's the effects of leaving lignen and chlora-materials in the celulose. It's easy to get rid of lignen through chemical reaction and intense heat. Chlora-mateirals are harder to get rid of and require chemical treatment (soaked in chlorine bleach) over several stages and then rinsed several times to remove the bleach and any possible byproducts that could affect the dyes and binders used in later stages. Since the affects of poorly prepared celulose pulp are readily noticeable, it is safe to say that AR used very high quality celulose pulp in their drivers. Not to say that the cones wouldn't eventually break down if exposed to direct sunlight, but that tends to be more a function of binders than celulose at the age of pulp we're dealing with. Conclusion: If a driver's cone has not failed by this point, it is safe to assume that the celulose used was lignen-free, chlora-free and generaly devoid of contaminants that would cause breakdown over time, and therefore we can eliminate this as a cause of sound quality degradation.

Next, celulose fiber length needs to be considered. Each type of driver (woofer, midrange, tweeter) has different electromechanical goals and so changes the stressors and therefore the requirements of the build materials. Tweeters, for example, have very limited excursion requirements. The wavelengths for which they are designed to reproduce are very short, and therefore require very tight control of their movement. Also, because of their need to respond quickly to signals, reduction in mass is very important. It doesn't seem that paper is a very good choice for tweeters, but nothing could be farther from the truth. Celulose fibers are extremely lightweight, and are hollow. When combined with a light-weight binder, a very rigid structure can be formed using very short fibers. Since the mass requirements outweigh the structural intergrity requirements of the tweeter cone, the shortest, lightest celulose fibers possible should be used. Midrange drivers produce wavelengths longer than tweeters, and therefore need increased flexibility in their mechanical susspension. It is rare that a true midrange driver uses an acoustic susspension like woofers, but it does happen. More commonly, the midrange looks like an overgrown tweeter, or a woofer with no surround. Celulose fibers in these midranges should be longer, to cope with the stressors applied to it, but still short enough to allow for molding into a long-wave friendly shape. A ballance of structural rigidity and decreased mass is required in a midrange driver, so celulose fibers should be short enough to allow for moldability without comprimising structural integrity, but long enough to allow for rigid areas to susstain themselves. Woofers require very long excursions, and the voice coil exerts extreme pressure over the surface of the cone. The shape of a woofer cone is a simple cone, and any molded features exist second to the need for rigidity, therefore very long celulose fibers are used. So we see, depending on the application, celulose fibers should be chosen according to the stressors applied to the final application, as well as the type of binder to be used in molding. I examined three drivers under a microscope at work and found the following. The tweeter from an AR 17 has fibers ranging from very very short (wider than it is long!) to short (about 1mm) The midbass drivers from my AR 9 wasn't a great candidate, because it really is a woofer, but it will do. Fibers in this driver ranged from 2mm to 1cm. The 12" woofer from my AR9 yielded fibers ranging from 1cm to 3.5cm! Clearly the celulose for each driver was carefuly chosen based on it's mechanical requirements, and have therefore successfully exceeded those requirements based on their continued operation without failure or signs of stress.

Finally and perhaps most importantly is the consideration for the type of binder to be used in the adhesion of the celulose fibers. Unfortunately, I know very little about binders used in the manufacture of celulose products that are non-paper. Basically, anything that can be used as a glue can be used as a binder. I do know that resins such as common fiberglass resins are used today in paper cone woofers, but I would be speculating on anything past this. Suffice it to say that the binders used in the drivers of days past have done their job to their specifications, since they have not decintegrated! Does anyone here have any knowledge of the binders used in paper cone drivers?

Well, there you go. My contribution to the cause. I think we can safely assume from the above that paper cone drivers do not significantly degrade over the time frame to which we are subjected when restoring our classic beauties. Of course time will tell, but I think the cones are safe