Crossover capacitors

J

josko

Audioholic
How exactly do crossover capacitors wear out during use? Is there a simple test to find out whether a cap. is still good or needs replacement? Also, do resistirs and inductors in crossovers wear out like capacitors do?
 
Swerd

Swerd

Audioholic Warlord
How exactly do crossover capacitors wear out during use? Is there a simple test to find out whether a cap. is still good or needs replacement? Also, do resistirs and inductors in crossovers wear out like capacitors do?
None of those components "wear out due to use".

A type of capacitor, electrolytic, contain a liquid solution that conducts electricity. They have been known to dry out over 20-30 years, causing them to no longer function correctly, but not all electrolytic capacitors do this.

If you have speakers that old, and if the crossovers have electrolytic capacitors, measure them with a multimeter that measures capacitance. If they are out of spec, assuming you know what range of values is in spec (most crossover caps are spec'ed to be ±10% of their labled value), replace them with another cap of the same value.

There are other types of caps that cannot dry out, and never deviate from their original value. Look for metalized polypropylene caps - they cost more than the cheap electrolytic caps, but they are still reasonable. There is no audible benefit from using high-priced or exotic caps in audio crossovers.

Resistors and inductors, as long as they are intact, always work.
 
DD66000

DD66000

Senior Audioholic
Although components may or may not wear out, I've found having built a number of crossovers, that the newer caps, like polypropylene, are superior to any of the caps used 30 years ago.
As such even if an old cap is still in spec, the sound can be improved by replacing them with the newer polypropylene caps.

And to go a step further, if the speakers in question were of such a good quality when new, like top of the line types, then spending the money to built new biased crossovers would be a huge improvement.
 
J

josko

Audioholic
Speakers in question are Polk Monitor 10a's, bought new in '82, and emotionally important to me. They seem to sound fine, although it could be 'cause I'm just so used to them. I've been reading about upgrading crossovers on polkaudio, and confess to being a bit skeptical about aural differences between different brands of 'boutique' crossover capacitors.
I don't mind spending $100+ on crossover components but am having a hard time figuring out at what price point the audible differences give way to hype, like silver/gold French-made $100 capacitors???

BTW, what's a 'biased' crossover? I'm assuming I just want to replace existing components with higher-quality identical-value ones. Polk is nice enough to publish a schematic of Monitor 10 crossover. Is it worth my while to research a different design, or am I better off following the Polk schematic with new components?
 
Swerd

Swerd

Audioholic Warlord
Speakers in question are Polk Monitor 10a's, bought new in '82, and emotionally important to me. They seem to sound fine, although it could be 'cause I'm just so used to them. I've been reading about upgrading crossovers on polkaudio, and confess to being a bit skeptical about aural differences between different brands of 'boutique' crossover capacitors.
I don't mind spending $100+ on crossover components but am having a hard time figuring out at what price point the audible differences give way to hype, like silver/gold French-made $100 capacitors???

BTW, what's a 'biased' crossover? I'm assuming I just want to replace existing components with higher-quality identical-value ones. Polk is nice enough to publish a schematic of Monitor 10 crossover. Is it worth my while to research a different design, or am I better off following the Polk schematic with new components?
If you have the schematic diagram of the crossover, and you want to rebuild it according to Polk's design, you can probably do it for quite a lot less money. The most expensive parts may be the inductor coils, but it depends on their size. You may very well be able to reuse the existing inductors. Can you post the diagram or list the components? With that it will be possible to estimate the cost.

I doubt if Polk used a biased crossover in their original design. Stick to what their design says.

You should be highly skeptical about $100 capacitors! The type of capacitors I prefer for new crossovers are the inexpensive metalized polypropylene (MPP) type sold by Parts Express (Dayton) or Madisound (Bennic). For example, a Dayton 10 µF cap costs $4.12 and a Bennic costs $4.50. Don’t pay extra for the Dayton ±1% MPP caps unless their standard ±5% ones are out of stock. All the standard Dayton MPP caps I have measured have always been much closer than ±1% of their specified value. If you can't find what you need in either of those brands, Solen or Jantzen are fine and don't cost much more. Prices much higher than those are a waste of money.

For what it's worth, Madisound sells Bennic electrolytic (some times called non-polar electrolytic, NPE) caps (http://www.madisound.com/catalog/index.php?cPath=404_5_304) that are OK to use if you need a large value cap (>50 µF) for the woofer low-pass filter. I avoid the Dayton electrolytic caps because they are often more than 10% out of spec when new. I had to buy more than 10 just to get 2 that were within 10% of the right value! The Bennic NPE caps are usually better at that.

I usually buy Madisound inductors because their prices are a bit lower than the ones sold by Parts Express. I use 16 gauge inductors if they are in series with a driver such as in a woofer low-pass filter, and 18-20 guage for any that are in parallel parts of the circuit. The audio resistors sold by Parts Express or Madisound are both fine. I like the Parts Express resistors because they have their resistance value printed on them. The Madisound ones have a color-coded stripe label and I always forget the coding system.
 
DD66000

DD66000

Senior Audioholic
BTW, what's a 'biased' crossover? I'm assuming I just want to replace existing components with higher-quality identical-value ones. Polk is nice enough to publish a schematic of Monitor 10 crossover. Is it worth my while to research a different design, or am I better off following the Polk schematic with new components?

Biased crossovers have a 9v battery applied to the caps. JBL uses this type of crossovers on their high end K2, Everest II speakers. We're talking $15K & $30K speakers.

I built Charged-Coupled JBL TM(biased) for my 30 year old L212s. Which were JBL's top of the line speakers in the late '70s. A great improvement, but quite costly. The reason for the cost is you have to double the number of caps and double their size, as you replace each cap in the original design with a series pair. Its at the common point of each pair that connects to the 9v battery through 2~6 megohm resistors.

That is the reason I said in my first post, if the speakers in question are of a high enough quality, one might want to built those XOs.
For those Polks, I wouldn't spend that kind of money.

The cost is directly related to the size of the caps required to do the baising. In the case of the L212s (3-ways) it was $300 for the pair. Building three cards for each speaker (one for each driver). I used mostly Solen caps and a few Dayton caps.

If I were to built the baised XOs for the newer PT800s the cost would be at least $500/pr, as the caps required are quite a bit bigger than required for the L212s.
 
DD66000

DD66000

Senior Audioholic
Any changes between the two is a result of having different mid-range drivers, from what I could see.
 
F

Face`

Enthusiast
If you have the schematic diagram of the crossover, and you want to rebuild it according to Polk's design, you can probably do it for quite a lot less money. The most expensive parts may be the inductor coils, but it depends on their size. You may very well be able to reuse the existing inductors. Can you post the diagram or list the components? With that it will be possible to estimate the cost.
To keep it as simple and cost effective as possible, he should just replace the resistors and capacitors on the stock boards, why would he want to build "new ones"? More on that below.

You should be highly skeptical about $100 capacitors! The type of capacitors I prefer for new crossovers are the inexpensive metalized polypropylene (MPP) type sold by PartsExpress. For example, a Dayton 10 µF cap costs $4.12 and a Bennic costs $4.50. Don’t pay extra for the Dayton ±1% MPP caps unless their standard ±5% ones are out of stock. All the standard Dayton MPP caps I have measured have always been much closer than ±1% of their specified value. If you can't find what you need in either of those brands, Solen or Jantzen are fine and don't cost much more. Prices much higher than those are a waste of money.
Waste of money in general, or for a 20-30 year old pair of bookshelves? Have you ever tried any high end caps?

Here's the original crossover:
Link 1.

and here's the upgraded one recommended:
Link 2.
Which tweeter do you have now? The original diagram should be for the Peerless tweeter, the other is for the SL2000. The differences in the two diagrams are both in the HF circuit, not LF(woofer). I would remove the binding post cup and take a look at what you have now, then replace the components with new ones of the same value instead of going by the diagrams in case you have a oddball tweeter(SL1000 or SL2500).

Depending on how much use or abuse they've been through, it can't hurt to replace the resistors while you're in there. I would also recommend replacing all capacitors with film caps. Dayton for the LF circuit, Dayton or better for the HF(tweeter) circuit. Also, does your model have fuses or poly switches? If you have poly switches, I would recommend either replacing or bypassing them. If you have a high current amp, you can safely bypass them. If not, new poly switches are available for free from Polk. For fused speakers, clean all connections. Since all inductors are air core, leave them alone.
 
DD66000

DD66000

Senior Audioholic
To keep it as simple and cost effective as possible, he should just replace the resistors and capacitors on the stock boards, why would he want to build "new ones"? More on that below.

Waste of money in general, or for a 20-30 year old pair of bookshelves? Have you ever tried any high end caps?

Which tweeter do you have now? The original diagram should be for the Peerless tweeter, the other is for the SL2000. The differences in the two diagrams are both in the HF circuit, not LF(woofer). I would remove the binding post cup and take a look at what you have now, then replace the components with new ones of the same value instead of going by the diagrams in case you have a oddball tweeter(SL1000 or SL2500).
Yes the tweeters are different, but the woofers are also different, as they're labeled with different #. 6501, 6503

Doing anything more than replacing the resistors, caps isn't worth the time.
 
Swerd

Swerd

Audioholic Warlord
New caps from Parts Express:

Dayton 12 µF part #027-430 $4.57 each buy 4

Jantzen 33 µF part #027-944 $10.96 each buy 2

The diagrams call for a 34 µF cap, but remember that in speaker building ±10% is acceptable. For 34 µF, that means from 30.6 to 37.4 µF is acceptable.

Total = $40.20 plus shipping

I'd keep the same inductor coils. Replacing them would bring up the question of what gauge wire. The schematics don't mention the inductors' DC resistance or their gauge. Although it is easy to say thicker is better, the DC resistance of the coiled wire is part of the circuit. Changing inductor gauge might then require different resistors in the circuit. So, if it ain't broke, don't fix it.

If the resistors look charred or darkened, replace them with Dayton audio grade resistors of the same value. I think they are $1.25 each.

Have fun with your project
 
WmAx

WmAx

Audioholic Samurai
There is no such thing as sound attributable to a capacitor, as long as the capacitor is operating correctly and within a stable thermal range. There is not one single credible double blinded study to demonstrate otherwise, nor any perceptual explanation of known audible properties that is rational.

Within a passive crossover, it is wise to use all film capacitors such as Mylar, polyester or metalized polypropylene, as these will remain stable in value for practically forever where as electrolytic capacitors are subject to the electrolyte inside drying up and causing the value of the part to drift substantially over time.

I can not conceive of any benefit to a biased passive crossover, either. A standard passive crossover will have no audible characteristic except those directly caused by the filtering characteristics therein and the electrical interaction of impedance with the amplifier source. That being said, a fully active crossover is superior to any passive system, as a full active system can have a large number of adjustments made on the fly to compensate for specific circumstances(room loading/acoustics, etc.) as well as allowing more corrective filters and a more aggressive type of filter.

-Chris
 
F

Face`

Enthusiast
You're kidding right?

Yes, as long as it's doing it's job correctly, they were all look the same on paper, but as far as sound quality goes, you couldn't be more wrong.
 
WmAx

WmAx

Audioholic Samurai
You're kidding right?

Yes, as long as it's doing it's job correctly, they were all look the same on paper, but as far as sound quality goes, you couldn't be more wrong.
Oh, I'm quite serious when I state: "There is not one single credible double blinded study to demonstrate otherwise, nor any perceptual explanation of known audible properties that is rational." - WmAx

Of course, if you are evaluating capacitor 'sound' with the standard A/B sighted tests with no controls, of course you are likely to perceive(imagine) a substantial difference in sound between capacitors (and just about anything else including cables, amps, etc.). This is not a reliable method for determining difference(s).

-Chris
 
MinusTheBear

MinusTheBear

Audioholic Ninja
WmAx,

Would it be detrimental to the overall speaker design to swap out expensive up to spec parts for cheap up to spec parts (caps, resistors etc) in the crossover with everything else left constant. Also there is some talk about crossovers "breaking in", what is the logic behind that for people making those types of statements.

Thanks.
 
F

Face`

Enthusiast
Oh, I'm quite serious when I state: "There is not one single credible double blinded study to demonstrate otherwise, nor any perceptual explanation of known audible properties that is rational." - WmAx

Of course, if you are evaluating capacitor 'sound' with the standard A/B sighted tests with no controls, of course you are likely to perceive(imagine) a substantial difference in sound between capacitors (and just about anything else including cables, amps, etc.). This is not a reliable method for determining difference(s).

-Chris
So most reviews or impressions on how Solen caps are bright and grainy are just imaged? Doesn't it seem a little odd to you that most come to the same conclusion?

How you done any testing or comparisons yourself?

Can you point me towards a published study on how reviewers failed to discern a difference between capacitors in AB testing?

I've done AB testing here with friends many times. The results are sometimes surprising. Sometimes a $5 cap will be almost discernible from a $50 cap, but sometimes a $5 cap will sound vastly different than a $10 cap. It all comes down to what sounds best in your system, synergy. To make such a blanket statement as they all sound the same is untrue.

Here's some info for you: http://www.sonicstate.com/news/2008/05/27/aes08-if-the-cap-fits-it-must-sound-better/

WmAx,

Would it be detrimental to the overall speaker design to swap out expensive up to spec parts for cheap up to spec parts (caps, resistors etc) in the crossover with everything else left constant.
I'm not sure why someone would want to do that, but go right ahead and try it for yourself.
 
WmAx

WmAx

Audioholic Samurai
So most reviews or impressions on how Solen caps are bright and grainy are just imaged? Doesn't it seem a little odd to you that most come to the same conclusion?
Not odd at all. Standard audiophile mindset. You can find the same overwhelming opinion on high end cables. No difference or surprise here. People are naturally going to make huge erroneous assumptions with perceptual issues unless the set up proper controlled testing that is as free of bias as possible.

How you done any testing or comparisons yourself?

Can you point me towards a published study on how reviewers failed to discern a difference between capacitors in AB testing?
I actually had several discussions with and helped with the protocol of the double blinded capacitor testing that was carried out by Dennis Murphy and Bob a couple of years ago on the DIY loudspeaker events. I had several discussions with Bob via email. Then Bob and Dennis constructed the final test system and carried out these tests at several DIY events. The results of those tests were that no one could reliably pick out a high end film capacitor from even an electrolytic(as long as the electrolytic was within a tight tolerance of value) when knowledge of the items listened to was removed. The only positive outcomes were with very limited trial numbers and when retested, came out as random.

Such tests were not even needed, however. As there is no modified parameter measured that would suggest any audible difference in the first place. It's all audiophile bunk. If there is an audible difference, then it will show up in measurements.

-Chris
 
WmAx

WmAx

Audioholic Samurai
WmAx,

Would it be detrimental to the overall speaker design to swap out expensive up to spec parts for cheap up to spec parts (caps, resistors etc) in the crossover with everything else left constant. Also there is some talk about crossovers "breaking in", what is the logic behind that for people making those types of statements.

Thanks.
As long as the parts are within tolerance and have proper specifications(inductors within a close tolerance not just in inductance but also in DCR, for example) for the application, then it will not make an audible difference. I do recommend against electrolytic caps though, as they tend to drift in value over time and are more prone to failure.

-Chris
 
lsiberian

lsiberian

Audioholic Overlord
So most reviews or impressions on how Solen caps are bright and grainy are just imaged? Doesn't it seem a little odd to you that most come to the same conclusion?

How you done any testing or comparisons yourself?

Can you point me towards a published study on how reviewers failed to discern a difference between capacitors in AB testing?

I've done AB testing here with friends many times. The results are sometimes surprising. Sometimes a $5 cap will be almost discernible from a $50 cap, but sometimes a $5 cap will sound vastly different than a $10 cap. It all comes down to what sounds best in your system, synergy. To make such a blanket statement as they all sound the same is untrue.

Here's some info for you: http://www.sonicstate.com/news/2008/05/27/aes08-if-the-cap-fits-it-must-sound-better/


I'm not sure why someone would want to do that, but go right ahead and try it for yourself.
Please note this is not intended in a mean tone, but a friendly one.:)

Face we at AH seek to dispell myths and point people to real scientific logical ways to improve loudspeaker builds and designs. I think you can acknowledge the need for science and logic to be used far more in our field. :D

:(Bose is the ultimate pinnacle of non scientific loudspeaker development. Based purely on hype and marketing. Bose would likely win many nonblind AB tests with consumers. But it would have nothing to do with the sound. :eek:

IMO, all speaker systems should be engineered using sound scientific principles not hearsay. :)Engineering focuses on the things that are proven to matter most not little truly imperceptible or unproven differences. In this case I think it's safe to say you are the party in need of the proof. ;)

Feel free to provide your documented research as it would be helpful to understanding your position and at least giving it some credibility.

Remember it's not you we have an issue with. It's simply this hypothesis we disagree with.

On a lighter note have you posted your systems pictures in the member systems gallery. :)

I've been wrong more times than I can count on this forum. In fact just today I posted without thinking something through and analyzing it more. My suggestion is don't be tied to your opinions unless they are life, death, or religious. In those cases grip tightly. :)
 
Swerd

Swerd

Audioholic Warlord
…I actually had several discussions with and helped with the protocol of the double blinded capacitor testing that was carried out by Dennis Murphy and Bob a couple of years ago on the DIY loudspeaker events. I had several discussions with Bob via email. Then Bob and Dennis constructed the final test system and carried out these tests at several DIY events. The results of those tests were that no one could reliably pick out a high end film capacitor from even an electrolytic(as long as the electrolytic was within a tight tolerance of value) when knowledge of the items listened to was removed. The only positive outcomes were with very limited trial numbers and when retested, came out as random…
I was one of the participants in that blind listening test. It took place over 4 years ago, and I posted about it here. About 40 people in all, mostly DIY speaker builders, participated in that test. Some were newbies, like myself, and others had many years of experience.

Each person took a test which consisted of 12 individual trials listening to short selections of music over very high quality speakers that contained two crossovers inside. The crossovers were identical except for one capacitor - one used an expensive film capacitor and the other used either a non-polar electrolytic or cheap polypropylene. A listener could instantly switch between the two crossovers while listening, and record whether they heard a difference or not.

The result was that no one there could successfully identify the different crossovers by sound. The results were no different than if people were guessing randomly.

A similar version of this whole test was repeated at the Dayton Ohio DIY meeting about a month or two later. It had nearly identical results.

Now those who wish can quibble that this was never published in a scientific journal. True, but for a variety of different reasons, that will probably never happen. Real science journals aren't interested in this (for good reason). There is no scientific reason to explain why capacitors made with different materials or construction should behave differently, nor is there any credible listening test data to suggest that people CAN hear a difference. End of story.

It would be wise to ignore unsubstantiated claims that passive devices such as capacitors (or wires, etc.) can generate different sounds, and spend more time reading the reports of such real scientists as Floyd Toole or Sean Olive who have spent their careers studying just how listeners impressions (usually of speakers) can be influenced by such things as appearance, brand identity, price, etc. It is a genuine psychological phenomena not limited to hearing. Too many people in audio just plain ignore this or pretend that they are immune to such influences.
 

Latest posts

newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top