Quote:

Quote:

Quote:
Limited to speaker crossover networks. Yes or no.

It depends entirely on what capacitors were used in the first place and what capacitors were used with the replacement.

There is no general answer.

There are also other issues, if ESR changes substantially, you may get a difference due to simple changes in ESR, for instance, not due to any "capacitive" part of the system.

The major change that most loudspeaker capacitor upgrades apply usually relates to the simplist thing - their actual capacitance. Most of the original parts were +80 -20, so we don't know what they did for sure without measuring them pretty carefully.

In the case of the Sansui, some of the inductors look pretty skimpy. Iron cores, especially too-small iron cores are a flashing red light.

+80/-20????? Where are you getting these numbers?
That is sometimes the tolerance for a power supply electrolytic where more capacitance (+80) is better.
Most low cost NP electrolytics in speakers are +/-10%.
Some companies have them custom wound to spec.
What company are you referring to that used +80/-20 NP electrolytics?

Pete B.

Quote:

Quote:

The major change that most loudspeaker capacitor upgrades apply usually relates to the simplist thing - their actual capacitance. Most of the original parts were +80 -20, so we don't know what they did for sure without measuring them pretty carefully.

True, this is why I always tested caps before I put them into a crossover. It's not the -20 that is the usual problem, either, +80 on a tweeter crossover can be, well, annoying.

Quote:

In the case of the Sansui, some of the inductors look pretty skimpy. Iron cores, especially too-small iron cores are a flashing red light.

Especially if they are for low frequencies. For high frequencies maybe the bh integral isn't so big.

Mass can also equal delta or acceleration. Distortions can be high, with regard to comparisons under various loading levels. Inductors are tricky beasts. It's the mass of the field (dynamic geometric, ie MHD type integrative) and it's integration that is key. The complex wideband LCR of the inductor under various loads would be the resultant basic measurement scenario. These things are rarely considered outside of the basic math but are as real as it gets.

Quote:
+80/-20????? Where are you getting these numbers?
That is sometimes the tolerance for a power supply electrolytic where more capacitance (+80) is better.
Most low cost NP electrolytics in speakers are +/-10%.
Some companies have them custom wound to spec.
What company are you referring to that used +80/-20 NP electrolytics?

Pete B.

More to the point, 20 years later, what do you think the chance of making it to -20% is?

Quote:
More like 35 years later.

I was being cautious. But in my experience, electrolytic caps do not have a good history in terms of longevity.

Some newer ones do ok, but bipolars have a particular issue in that they have no constant voltage across them to keep the polarization going.

Now, I haven't tried recent bipolar electrolytics, but I know the older ones also had a capacitance that tended to vary with voltage across the capacitor, which was great fun in integrating into crossover designs, let me tell you, which is why I never used them if I could even inconveniently avoid it.

Quote:
That's a bunch of shit. I tested the caps and they still held a charge within tolerances. The newer caps were just better caps. Stop trying to show off your knowledge. It just makes you look like an "outsider".

Suit yourself, but I suspect I've measured a lot more capacitors in my time than you have. Can't say that for sure, but I've built, tested, repaired, scrapped, etc, lots of equipment, from loudspeakers to high-speed DSP's.

Btw, what was the ESR vs. original spec?

Quote:
I tested the caps and they still held a charge within tolerances. The newer caps were just better caps. Stop trying to show off your knowledge.

My BS detector is going off. Testing caps does not usually invovle directly measuring their ability to hold charges, but rather it measures their impedance and deconstructs that measurement into a capacitive part and a resistive part.

Quote:

It just makes you look like an "outsider".

A reasonable, mature person might take being thought to be an outsider to the "in-group" here, as a complement.

Quote:
This deserves a better answer. I've been familar with the BBC monitor speakers for some time. Then the technology branched out to Harbeth speakers. Their M40 monitor is also an improvement over the original BBCs. Ironically, the biggest customer base is in Japan. To make a long story short when I first started looking at the SP2500 by Sansui I noticed they were something of a knockoff of the BBC design. And the newer Harbeth's 3-way is a further improvement of that design. So, in essence the SP2500s are my poor man's Harbeth M40 Monitor Speakers.[/url]

BTW, you can see these boxes are still relevant since they [Harbeth] are advertising in Stereophile. You want to see how an efficient speaker system is designed? Read the above pdf file.

When I first saw the photo in this thread, I said "Hey! I recognize those! I have the same speakers!". On closer inspection, they're not. The box, the grill, and the Sansui logo are the same, and it appears identical in size to the ones in the photo. But mine are front-ported 2-ways, and the back plate has metal binding posts instead of the plastic spring type, and no xover knobs. I remember spending many months tweaking the hell out of them (but this was maybe 15 years ago, though I still have them around somewhere). They were my ex's speakers, so I could mess with them just for fun and educational purposes. The drivers were good quality (Phillips tweeter, inverted rubber surround woofer, which means they still hold up after maybe 35 years!), but lack of bass was the big problem with these as I recall. So at one point I had made like a 3 ft. "horn", sticking it out of the front port of each speaker, to try to coax some bass out of them. Certainly made the speakers a conversation piece whenever visitors passed by.

This I believe got replaced by straws. Some of the other ideas that came and passed, of which I can recall, included sticking old speaker magnets on the back of the drivers, stuffing the cabinets with rocks, soldering speaker wire directly to the xover's, ditching the xover's, adding a tweeter divider and installing 3 spikes to the bottom of the cabinets. I remember that at one point, I got these old speakers to make a really musical sound. But then I had to go and screw it up, because I can never leave well enough alone! I don't know how they would compare to the 2500 model. It's too bad Sansui didn't progress, because as I seem to recall, the speakers they made after that in the 80's, were only good for the scrap yard.

Quote:

Quote:
Lamont Sanford wrote: I tested the caps and they still held a charge within tolerances. The newer caps were just better caps. Stop trying to show off your knowledge.

Arnold Kreuger wrote: My BS detector is going off. Testing caps does not usually invovle directly measuring their ability to hold charges, but rather it measures their impedance and deconstructs that measurement into a capacitive part and a resistive part.

Glad to see you got here just in time to poison another thread, Arnold. Going out of your way to accuse a member of lying about testing his own caps?? Way to score points for your team, Arnold! How many points did you figure you won this time?

Quote:

Quote:
Lamont Sanford wrote: It just makes you look like an "outsider".

Arnold Kreuger wrote: A reasonable, mature person might take being thought to be an outsider to the "in-group" here, as a complement.

It sounds like you're admitting here that you're here as a troll. It doesn't look like you're fitting in very well with the group as a whole. I'm starting to be concerned for your future.

Quote:
It just makes you look like an "outsider".

Pssssst, Lamont, he is an "outsider" and chances are he'll always be just that.

However, I'd take his word that he has measured more caps than you have. Do you measure caps for relaxation?

Quote:
A reasonable, mature person might take being thought to be an outsider to the "in-group" here, as a complement.

And here jj and Xeno were trying so hard to fit in.

Hey, pajamas!

Guess what!

A reasonable and mature person wouldn't want a full time troll like you in their group let alone pay any attention to you.

Feel free to be a very outsider here, pajamas. So outside we don't realize you're gone.

Oh.

Sorry for the meltdown here. I guess I was "conquoered".

LOL, pj, LOL! You do provide the comic relief this forum needs.

Quote:

Quote:
+80/-20????? Where are you getting these numbers?
That is sometimes the tolerance for a power supply electrolytic where more capacitance (+80) is better.
Most low cost NP electrolytics in speakers are +/-10%.
Some companies have them custom wound to spec.
What company are you referring to that used +80/-20 NP electrolytics?

Pete B.

More to the point, 20 years later, what do you think the chance of making it to -20% is?

Uh-huh. I'm sorry to say this jj, but your "more to the point" response feels like a deflect to me. You have left readers confused now and not knowing which of you two to believe. Because you came here from the "Hydrogen Audio forum", touting yourself as a "professional", with superior knowledge to us "audiophools" on Stereophile, and that you are here to educate us. You claim particular expertise in speakers, because you claim to be a speaker builder (like maybe, 90% of the people on this forum are. cough cough). You also seem to think you have measured "a lot more capacitors in your time" than others here. So since you are always so adamant on this forum in claiming to have superior knowledge of audio than your subjects, I would think you'd be a little more insistent on setting Pete straight on his technical error, than changing the subject as you did. I'm sure you can see the problem now. Pete claims low cost NP electrolytics in speakers are going to be within +/-10% tolerance values. You OTOH, claim one should expect a figure of (and please let me know if I'm misquoting this figure) "+80/-20"!!! That's a huge discrepancy. Surely you can tell the difference between PS caps and speaker caps, right?

So the question remains: which of you is right? I'm sure you must be right, since you are coming from Hydrogen Audio, and tout yourself as a "professional audio expert" in just about every other message you write, and claim superior knowledge to Stereophilers of audio, and that everything you say about audio has the blessing of "science". All I am asking here is that you back up your claim and prove you are right, and that you posess this "superior knowledge of audio" over the rest of us on Stereophile. Just as you would ask us to prove claims to you, so this is a fair request. Please provide links to evidence that prove Pete was talking out of his a--, and that prove accepted tolerance values for low cost NP electrolytics in speakers are in the range of "+80/-20%" and not more like +/-10%, which is Pete's claim.

This will be a good opportunity for you to prove that you are who you say you are. I, and I'm sure others, will use your response (or lack thereof) to judge your future claims about audio on this forum.

These are a very common low cost NPE cap, been around for probably 15 to 20 years and are used in _many_ commercial designs, +/- 10 %:
http://www.madisound.com/manufacturers/bennic/bennic-caps-electro.php

Please show me a +80/-20% cap used in a speaker crossover.

Just a few measurements that I've made are provided here and they were all well within spec:
http://baselaudiolabs.googlepages.com/ADVENT_LA_XO.html

I do have to point out that they should be measured at their working voltage since the breakdown voltage can be reduced with age. Also, I have seen leaky (resistance leakage not chemical leaks) electrolytic caps in older speakers which could result in driver damage.

Military spec for electrolytic shelf life is 10 years.

Quote:
These are a very common low cost NPE cap, been around for probably 15 to 20 years and are used in _many_ commercial designs, +/- 10 %:
http://www.madisound.com/manufacturers/bennic/bennic-caps-electro.php

Please show me a +80/-20% cap used in a speaker crossover.

Just a few measurements that I've made are provided here and they were all well within spec:
http://baselaudiolabs.googlepages.com/ADVENT_LA_XO.html

I do have to point out that they should be measured at their working voltage since the breakdown voltage can be reduced with age. Also, I have seen leaky (resistance leakage not chemical leaks) electrolytic caps in older speakers which could result in driver damage.

Military spec for electrolytic shelf life is 10 years.

The biggie I've seen was ESR, leading to a somewhat surprising disappation factor. If it's starting to leak (electrically) usually the ESR will have gone way up first.

But not always, of course.

Dear Mr. Audio Expert Impersonator (aka "Woodinville", aka "jj"),

I asked you to show me a low cost NP +80/-20% electrolytic used in a speaker xover, a very simple request, and you were unable to do so. Pete asked you the same and you were unable to do so. Which shows you had no idea what you were talking about when you told the OP that electrolytics for xovers will be in the +80/-20% tolerance range. That's what I thought. So this is the "superior intellect" that your buddy Arny was talking about, when he said that you and him, along with everyone on Hydrogen Audio, had a "superior intellect" and superior technical knowledge of audio, over anyone else here? Well something tells me you're not going to mention your performance in this thread, when you report back to your home boys on Hydrogen this evening!

As if it isn't bad enough that you are trying to mislead the public with erroneous information on audio on these forums, you won't even admit when you are proven dead wrong. Funny how it always happens "jj" (or whatever your real name is), that when you are wrong about something and someone asks you to provide evidence for your erroneous claims, which of course you are unable to, you ignore them. But whenever you feel slighted over some stupid little meaningless personal issue, which has nothing to do with audio but which you think you can prove, you're all over the guy with a vicious rant! I won't even begin to ask how you can pretend to be slighted as a professional, since you are obviously not. Maybe there is a guy named "jj" out there who is. If so, I request that you stop impersonating him.

All of your appeals to authority on this forum, your claims of intellectual or technical superiority over regular Stereophile members, your risible PR campaign to try hype your so-called "expertise", your demands that your erroneous opinions be treated as something more... all that ends now. I think what we've seen here today is "the troll who calls himself 'jj' is really The Wizard of Oz"! Little man with a big megaphone, hiding behind a moniker! You're free to share your opinions, Woodinville, or whatever your non-trolling name is. But make sure you make it clear that they are simply your personal opinions, of no more value than a hole in the wall. Do not claim here to be an "expert" in anything again. I do not want you misleading my forum members into believing the erroneous technical information on audio that comes from you, or other information. You clearly do not know what you're talking about, on audio, and nothing you say from here on out can be trusted.

Thank you for your cooperation.

- M. Frog

Quote:

Quote:
These are a very common low cost NPE cap, been around for probably 15 to 20 years and are used in _many_ commercial designs, +/- 10 %:
http://www.madisound.com/manufacturers/bennic/bennic-caps-electro.php

Please show me a +80/-20% cap used in a speaker crossover.

Just a few measurements that I've made are provided here and they were all well within spec:
http://baselaudiolabs.googlepages.com/ADVENT_LA_XO.html

I do have to point out that they should be measured at their working voltage since the breakdown voltage can be reduced with age. Also, I have seen leaky (resistance leakage not chemical leaks) electrolytic caps in older speakers which could result in driver damage.

Military spec for electrolytic shelf life is 10 years.

The biggie I've seen was ESR, leading to a somewhat surprising disappation factor. If it's starting to leak (electrically) usually the ESR will have gone way up first.

But not always, of course.

Yes, I often see caps with elevated ESR in power supplies. Much of my 20-25+ year old test equipment has needed the power supply caps replaced due to probably elevated ESR. Seems that high ambient temp, or high current in the cap (switchers for example) causes much more rapid failure. PC motherboard caps are a good example.

And then there were also the bad cap formulations:
http://en.wikipedia.org/wiki/Capacitor_plague

http://www.badcaps.net/

http://news.cnet.com/PCs-plagued-by-bad-capacitors/2100-1041_3-5942647.html

Quote:
Dear Mr. Audio Expert Impersonator (aka "Woodinville", aka "jj"),

- M. Frog

I actually support A/B or ABX testing and use it myself.

I certainly have a lot of respect for j-j's work, take a look at the presentations at his web page - quite impressive!

Some come on a bit strong here, and they ought to cool down and show more respect and civility so that the discussions can be more productive.

Quote:
These are a very common low cost NPE cap, been around for probably 15 to 20 years and are used in _many_ commercial designs, +/- 10 %:
http://www.madisound.com/manufacturers/bennic/bennic-caps-electro.php

Please show me a +80/-20% cap used in a speaker crossover.

No, I'm not going to run through all the speaker systems at my disposal and tear their crossovers apart looking for +80/20 caps.

However, open enough speaker boxes and you will see +80/-20 NP electrolytics.

I'll even take the devil's advocate position and point out that just because a cap is marked +80/-20, doesn't mean that it (a) wasn't actually +/- 10% as built, or (b) wasn't selected for a particular value during the process of manufacturing the speaker systems.

Quote:

Just a few measurements that I've made are provided here and they were all well within spec:
http://baselaudiolabs.googlepages.com/ADVENT_LA_XO.html

Wow, there are what 4 measurements? Well that must settle the issue of real-world capacitor values for all of the billions of speakers that have ever been manufactured! ;-)

The fact is that if you check enough speaker crossover caps, you'll find some that are far enough out of spec so that they make the speaker sound significantly different.

Replace them with caps that within spec, and the speakers will probably perform more like they were designed to.

Also, a lot of speaker crossovers are designed using some cut and try. Caps are swapped in, until the an adequate level of sound quality is obtained. If that cap of course has ESR, and that too has to be accounted for in any replacement, or the speaker will of course sound different.

As seems to be the way with these things, there are don't seem to be any relevant frequency response measurements on the web page that was cited.

Of course there should be, even it they are voltage measurements taken across the loudspeaker drivers.

You make a completely BS statement, show no actual measurements, and then attempt to discredit my measurements. You seem to enjoy what you might view as intellectual debate. Let me give you a clue, this is not debate, it is a childish, I can prove I'm right - even when you are wrong attitude.

You think that I've not opened and measured many speakers, amplifiers, etc.? Oh, wait I actually design electronic equipment, forgot to mention my own designs. They are mostly Military and industrial grade where we often use better parts.

I've never seen a 80/20 cap in the signal path of any audio component. There is a simple reason, 80/20s are not used there.

So, name one, or show a picture, if you are not BSing Arny.

Arny, your original statement was about caps with an 80/20 tolerance, not with regard to them going out of spec - spin things often? You might notice that I measured the capacitance there also and none of them were out of spec.

That page was intended to document the electrical and physical construction of the crossovers. That is all.
Why do you make this comment about no measurements, do you think I don't take measurements when they are needed? If you take a look at the SPICA TC-50 link in my sig, you'll see that I take measurements when I find it interesting and necessary.

Where are your acoustical measurements of speakers Arny?

Yeah, you know it all, right, in your own mind. You make a lot of assumptions, why? I did not come here guided by Ethan. Not going to bother reading the rest of your reply.

You're welcome for the (free) advice on re-capping your speakers (post #46807), LOL! The majority of my posts on the internet are to help people not get into pissing contests with people like you, and the rest.

Quote:

Quote:

Some come on a bit strong here, and they ought to cool down and show more respect and civility so that the discussions can be more productive.

Good adult advice but that's not going to happen where JJ is concerned. JJ (Pete) came here with an agenda along with Repete. Brought here via Ethan pointing out their intended victims. But JJ could use somebody licking the wounds he received from other threads he took a shit in but we'll never be able to find it. Take this thread for example. The answer was yes/no. He wanted to turn into a class at MIT after I pulled it back out of the pile. JJ is a Seagull Troll. He likes to fly into a thread, make a lot of noise, shit all over the place, then fly out. JJ and his two butt-buddies came to this forum with an agenda to denigrate some of the other members. So, I would extend your advice to JJ and Arny (Pete & Repete) as well as Ethan. Nevertheless, I will check out JJ's (Pete's) website for a little more insight. Not that there is anything wrong with that.

Please take note of asking for civility towards JJ (Pete) and Repete (Arny) flies into this thread and takes a shit on you.

Ah, I see, well I was accused of being associated with Ethan from the start. I'm asking for civility from everyone, there are certainly a number of offenders on the subjective side.

Quote:
You not Pete of Pete and Repeat. You're Pete B. Totally different animal. Sorry about the confusion.

"Pete" = JJ
"Repete" = Arny

Quote:
Krueger wrote: Wow, there are what 4 measurements? Well that must settle the issue of real-world capacitor values for all of the billions of speakers that have ever been manufactured! ;-)

"Wow, there are what ____ ABX tests? Well that must settle the issue of _______ for all of the billions of people that have ever heard _______! ;-)"

If anyone wants to have another crack at Arnie, I've made it easy for you: just fill in the blanks. Blank field #1 is the number of ABX tests reported on Blank Field #2: which is the article in question that the Kruegeris happens to be arguing against this time (ie. cables, amps, cd players, Furutech deMagnetizers, power cords, polarity, 256k mp3's... you know all that "magic audio stuff" we delude ourselves with). Blank Field #3 could simply be what they heard in Blank Field #2. Yes, debating audio "Krueger-Style" is no more difficult than that!

As for me, I got some to some fresh air and some museums to take in. And Kroogie? It wouldn't kill you to go out and see the light of day once a year. Your ABX/DBT religion wil still be where you left it. Hmmm.. on second thought... maybe it will. No, you're better to stay where you are. That's the safe bet.

Quote:
You're not Pete of Pete and Repeat. You're Pete B. Totally different animal. Sorry about the confusion.

"Pete" = JJ
"Repete" = Arny

BTW, I checked Pete's (JJ's) website. Only a loser would copyright a Comcast homepage. It's the equivalent of copyrighting a geocities homepage. I think I'll copy his entire site, give it new url and persona, and see how far he gets with his copyright infringement.

Oh that's nothing. Pete (JJ) used to add a copyright warning to every single one of his messages on usenet (aka newsgroups aka "Google groups"), warning people against copying or using his messages. I mean usenet, for crying out loud! I used to post his messages all over the internet, just to show him how much his copyright warning meant. And because I always thought his act was even funnier when peppered with inflated indignation and rage, but I admit, you don't really have to do much, if anything, to see that from him. Now the internet does all that for me; as usenet postings are automatically culled and regularly re-published (even edited) on all sorts of different websites, without anyone having to lift a finger. So it's just not the same anymore. (sigh).

Quote:
Where are your acoustical measurements of speakers Arny?

Arny Krueger has been too busy making recordings - see http://www.hydrogenaudio.org/forums/inde...mp;#entry636388 - with a microphone technique that results in not much more than "fat mono" - see http://forum.stereophile.com/photopost/showphoto.php/photo/2021 . Not only does this recording have very little stereo stage width, it has audible LF noise, hiss, and distortion, none of which Mr. Krueger appears to have heard.

I would be inclined to take his comments on audio engineering with a large grain of salt. :-)

John Atkinson
Editor, Stereophile

I had a bad case of fat mono once. I was lethargic, my swollen spleen was easily vented and I was prone to distorting facts and throwing hissy fits. And music sounded crappy, too.

I used to work for Kondo San and when we needed to test a capacitors sonic attributes, we always tested them in speaker crossovers as this location is quite sensitive.

With the crossover dangeling behind the speaker it was also easy!

Quote:
I used to work for Kondo San and when we needed to test a capacitors sonic attributes, we always tested them in speaker crossovers as this location is quite sensitive.

With the crossover dangeling behind the speaker it was also easy!

Very interesting!

For capacitors of the same value, did you notice large differences? Only subtle?

Quote:
I used to work for Kondo San and when we needed to test a capacitors sonic attributes, we always tested them in speaker crossovers as this location is quite sensitive.

Well, it certainly is an application where the tolerance, linearity, and ESR are all very important, which is not generally true in most non-analog-filter apps. Basically, you're building a high-powered analog filter, something that has been known to be "fun" for several generations, so, yeah, it is a good way to test most caps.

Quote:

With the crossover dangeling behind the speaker it was also easy!

***chuckle***

Welcome to the club All the modelling in the world won't replace actual measurements, there's no doubt about that.