Electrolytic cap performance / behavior vs. others in a crossover?

pkquat

I am only asking this to gain knowlwedge. First off, I know rebuilt crossovers will sound better...... moving on.

I pulled the caps from a later1990's crossover and measured the values at 120Hz, and 1kHz. I got significantly different values, but still within 10%. After doing some research, I found out this was a characteristic of electrolytics, and measuring at 100-120Hz is the more accepted value. Only 1 was out by close to 10% (high), the others were close to 1% (high).

Since the measured capacitance of electrolytics varies by frequency, how does this effect the crossover frequencies and the crossover in general vs. non electrolytics?

If the values of the older electrolytic caps measure very close to spec. are they still likely performing within spec.?

DarqueKnight

No. There are other things that go bad like the equivalent series resistance (ESR) increasing over time.

Here are some links to get you started.

http://www.nmr.mgh.harvard.edu/~reese/electrolytics/

http://www.nmr.mgh.harvard.edu/~reese/electrolytics/tec1.pdf

http://www.nmr.mgh.harvard.edu/~reese/electrolytics/tec2.pdf

https://www.allaboutcircuits.com/technical-articles/determining-equivalent-series-resistance-esr-of-capacitors/

pkquat

Thank you. I did read something about ESR, but thought it was only for higher freqnecies, MHz range. I will read up some more.

[Deleted User]

Electrolytic capacitors (both aluminum and tantalum) have series resistance and inductance and parallel resistance and inductance but also various combinations of series diodes and parallel diodes. If an electrolytic capacitor is properly polarized then most of the series and parallel diodes are biased on or off and the capacitor, more or less, behaves like a film capacitor. But not completely.
Thus using an electrolytic capacitor in a speaker crossover design with near
zero DC volts on both sides leads to various kinds of distortion, most of which are very audible.
The larger the capacitor value the more series inductance it has and the performance at higher frequencies suffer (that may have been the cause of what you measured). A fix for this is a parallel reasonable value film capacitor.

pkquat

Thanks for the additional information. The fine points get into a lot to wrap my head around regarding capacitor designs, but I am getting the gist of it.

These caps are non-polarized caps from some polks.

I did find a comment that, within the audio frquency range the electolytic caps near there nominal value should remain within their tolerance range, and not effect the performance of the crossover. I am guessing that is true.

From the info relating to ESR you both provided, and what I have read up on, dried out caps drift more at higher frequencies compared to how they perform when new.

A non polarized electrolytic cap has much less audible distortion issues right?

pitdogg2

If you're looking to replace caps in a speaker crossover go film and do not look back.

pkquat

I should have said "I know rebuilt crossovers with film caps will sound better" in my openning line. I forget the film cap part.

After doing some digging and measuring the electrolytics, I am looking to get a better understanding of them. The varying capacitance was also something I was not originally aware of. Electrolytics are used in most speakers even ones that cost thousands of dollars. I'm just trying to understand them better.

F1nut

They are cheap. Class is over.

pitdogg2

Yes cause they are cheap. Very few will use film caps and when they do they will also be the cheapest they can buy. Saving 25.00 per crossover adds up big over lets say 5000 crossovers. So if you're saving 100.00 per crossover that's half a million dollars right there.

[Deleted User]

Most of the time speaker manufacturers will use "non-polarized" electrolytic capacitors, especially in the high pass part of the crossover. Inside these are two capacitors of twice the target value connected in series "back-to-back" (+ - -+) so that the unwanted effects of a "regular" electrolytic will be somewhat reduced.
Not only is the part's cost a factor, but the size of a capacitor will be important to speaker designers. Imagine how in-wall and car audio speakers would look with large film type capacitors hanging on the back of mid-range drivers.

pkquat

@KennethSwauger Thank you. What I am gathering is that for a test bench and in general the np electrolytics perform close enough to the design spec within their tolerance range. I now have a better understanding of their performance, design, and aging.