Aging of capacitors within crossover networks


Hi folks, I have red that capacitors within loudspeaker's crossover networks show aging with regard to their characteristics (inductance, capacitance, reactance etc.) which has consequences for the crossover itself and finally also for the loudspeaker's sound. Are you familiar with this phenomenon? Does this mean that loudspeakers which are 10 years or older must undergo filter modification (replacement of parts)? What is your opinion?
dazzdax
What do you think of this statement by Rod Elliott of Elliott Sound Products:

"Non-polarised electrolytics are a different matter, especially when used in crossover networks. These have a tendency to lose capacitance as they age, shifting the crossover frequency with disasterous results (sonically speaking). Because the loss is gradual, you may possibly not even hear it until the tweeter has almost stopped working, as you get used to the sound over a period of time. Unless all bi-polars age at the same rate (unlikely), you will start to notice a difference between the two speakers. This is your cue to head off to the electronics shop and buy some replacements (non electrolytic, preferably)."

Because of this I started the thread.
Gee, Ed, I'm glad I never said any of those things and never recommended Blackgate.

I'll say it another way:
1. Speaker manufacturers use the best-sounding parts they can WITHIN SEVERE PRICE CONSTRAINTS.
2. There are LOTS of better-sounding caps out there that lots of us know about and have heard in our own systems that sound better than the generally inexpensive stuff most manufacturers use in their crossovers.

Stop accusing me of saying what I didn't.
Process control bsed manufacturing produces parts that are within a spec range. If the spec range for a cap is 10%, and sampling indicates they are within that range, they don't tweek the process further - it is in control. If (and this is a big if) - a speaker works better with more tightly controlled values, the manufacturer will have to test or buy selected value parts. If a copmpany is selling selected value parts, that means the stuff they sell everyone else is by design farther off nominal. I know that Spica used to measure their caps, and match caps individually to balance out their differences. That is why there are so many caps in a Spica crossover.
Dazzdax, if your crossover uses electrolytic caps, do indeed replace them, and as Mr. Elliott indicates, with better-quality caps. If the 'lytics are in the midrange or tweeter sections, replace them with the best film caps you can afford. As indicated above, the SCR/Solen/Axon 'propylene caps are good-sounding caps that are highly affordable; they're available here http://www.percyaudio.com/Catalog.pdf , and Michael has low prices and great service. If these caps are in series with the midrange or hi-frequency drivers, probably you should also bypass the SCRs with a small, higher-quality cap. To keep costs down and stay with the same supplier, I suggest you use a Reliable MultiCap RTX in 0.01 or 0.22uF value, at $3.50 to $5.95.

Another source of good-sounding, affordable film caps is North Creek, here http://www.northcreekmusic.com/CrossoverComponents.html . I used a combination of Zen and Harmony 'propylenes in my last 3 crossover projects and am VERY pleased with the results.

I've used TRT DynamiCaps for couplers in my vacuumtubed amps, but they're only 0.22s so they're still affordable. In larger values, the TRT-Ds are quite expensive IMO.
.
Honest1...NO. NO. NO. Exactly the opposite of what you said. Process control manufacturing produces product that is dead nuts on nominal, even if it doesn't need to be perfect. The process is tweeked until this happens. The old fashoned way was to 100 percent test the product and accept those which are in spec and junk or rework the others. Not only is this more expensive, but it results in an inferior product.

Go to Google, and research Taguchi. His engineering approach also involves design concepts. For example: if a circuit requires a 1 percent or a selected value resistor, the best approach is to go back to the drawing board and redesign the circuit so that it only needs 2 or 5 percent resistors. The brute force (stupid) approach is to spend a lot of time and money on precision resistors.