Jump to content

2500uFs caps do need replacing


Diamonds&Rust

Recommended Posts

Years ago I decided not to replace the 2500uF capacitors in my AR-10pi's when I replaced the other capacitors. The idea was that a 2500uF capacitor would "filter" at so low a frequency that it shouldn't be heard, so why bother with it? It was put there for amplifier protection in the AR-10pi and LST speakers (later used in an AR-9 in a slightly different way), and wasn't put in these crossovers to have any sonic effect.

After listening to my 10pi's with everything except the 2500uF capacitor replaced, I felt sure that I was right and that the 2500uF capacitor had no effect on anything. But, still, goofing-around with these is a hobby. I enthusiastically participated when Klaus offered to send some Jensen 2750uF (NPE electrolytic) capacitors from Denmark to me. He sent them quickly and then I put them aside. I really wasn't that ready to get *back* into the speakers. I thought, "It can't make any real difference."

Last night I decided to take a few minutes and do some tweeter-swapping in a pair of AR-14s. I was going back to original "made in USA" parts from later tweeters. That got more involved than it should have and so I thought that while I was at-it, with the tools pulled out, I might as well quickly change these 2500uF capacitors in my 10pi's just to satisfy myself that it really wouldn't make a difference.

I was wrong.

Again.

How do I accurately describe the difference? I really can't. It's too involved. An inaccurate, or at least incomplete, description might go something like this:

The entire "sound" of the speaker is affected in a very positive way when this capacitor is replaced, but it's hard to know if the midrange actually improved, or if the mud from the woofer abated so I could hear the midrange detail better, or if I'm hearing my amplifier having an easier time with the load.

The general impression is that a slight speaking-into-cupped-hands effect (that I didn't realize I even had) went away. The bass became much more articulate and less prone to being blatty and one-noted. Congestion has cleared and instruments seem to stay separate instead of forming a mash of noise. (again, I didn't think I had a problem - until I heard the speakers without the problem)

I've been up essentially all night listening to them because I'm finding the difference in them hard to come to terms with. I can't explain it. I don't even have a good theory. I guess any failing component in the path, no matter where it is, is going to have an effect. BTW - I tested the capacitors that came out of the speakers. They were about 2700uF and 2900uF; both within their 20% tolerance from 2500uF. So they were failing in some other way. What do we expect after 30 years, right? A lot of me has aged in the past 30 years, too.

Just as an example, one track I was listening-to has two male voices singing a baritone line. It is much easier to hear the differences in their voices now. Bass guitar has more string to it. Bass pedals have more obvious harmonics. But it's the improvement through the midrange (particularly that portion of piano notes and large stringed instruments <cello, upright bass>) that doesn't make much sense - unless the woofer (which gets up into those frequencies) is just behaving better.

I'm so encouraged, I'm going to replace the 2500uF capacitors in my AR-9s. I don't do that without trepidation, but if the 9s are suffering problems like the 10pi's were, I expect my 9s to show immediate improvement in bass clarity (not amount).

I'm sorry it took me so long to be able to post this.

On the other hand, had I recapped the speakers all at once without listening to them for years with bad 2500uF caps in them, I wouldn't have known that this capacitor has a fairly major impact on the entire cabinet even if you have already replaced all the other capacitors. I finally decided I had to install these (thanks again Klaus) and can report that the 2500uF capacitors also need to be replaced.

It was worth the time, money, and trouble.

Bret

Link to comment
Share on other sites

Hi Bret

I know what you mean, I also had troubles to describe the difference, I recappad one 10Pi at a time and compared the two. I sensed an improved agility in the bas, kind of better precision, I guess it is more or less what you describe about your bass, but in other words.

I know I am repeating myself from previous posts, but I think the 10 Ohm resistor is very important to pay attention to as well. I replaced the resistor with an audiophile one before I replaced the 2500 cap, and I had an clear improvement in the sound, especially voices, they came like several meters closer to the speaker front, opened up.

The other trick I have used successfully on my 10Pi and LST's is to bypass the big caps with smaller, to reduce the ESR for high frequences, in my 10Pi I have a 72 myF and a 2.2 myF polypropylenes, whereas the LST's are bypassed with a 200 myF Polypropylene and a 1 myF oilpaper/film capacitor. The bypass caps improved the airiness and definition on the tweeter, resulting in a kind of bigger and better defined soundstage.

I would not think that you would hear the same kind of improvement in the AR-9 since, it is used as a decoupling cap in an LC coupling circuit.

BRgds Klaus

Link to comment
Share on other sites

I think the 10 Ohm resistor is very important to pay attention to as well. I replaced the resistor with an audiophile one before I replaced the 2500 cap, and I had an clear improvement in the sound, especially voices, they came like several meters closer to the speaker front, opened up.

I did that when I replaced the other capacitors, too. I couldn't tell if it made a difference since I did both cabinets and the other capacitors at the same time. It's good to know I didn't completely waste that $10.00 or whatever it was.

The other trick I have used successfully on my 10Pi and LST's is to bypass the big caps with smaller, to reduce the ESR for high frequences, in my 10Pi I have a 72 myF and a 2.2 myF polypropylenes, whereas the LST's are bypassed with a 200 myF Polypropylene and a 1 myF oilpaper/film capacitor. The bypass caps improved the airiness and definition on the tweeter, resulting in a kind of bigger and better defined soundstage.

I was so sure that replacing the 2500uF cap would make no difference that I didn't think to by-pass it even though my hand was in the bag with a collection of by-pass capacitors in it. I'm not likely to go back into the cabinets for a while (another year?) but I'll keep that in mind for when I do.

The other capacitors are all by-passed with very small values, but I've told about that in many other posts so I won't bother here.

I would not think that you would hear the same kind of improvement in the AR-9 since, it is used as a decoupling cap in an LC coupling circuit.

My theory is that the circuit probably isn't working correctly anymore, so while it "shouldn't" make a difference, it probably will.

It's almost strange listening to these with all the "heaviness" and "sluggishness" removed. My listening-buddy and reality-check remarked that *now* they sound like a pair of 10pi's; that's in addition to other things he said. (we form opinions without telling one-another, then compare notes - we both heard the same thing although we described it a little differently) The more I listen the more I want to describe it as "better woofer control" but not really what I think-of as the effect of changing the damping with more-or-less stuffing.

He and I have been listening to his 10pi's since 1977 (on and off). Both of us remember needing to turn the tweeters and midrange switches down in several situations, but I couldn't imagine why we did that. I can imagine it more easily now.

Link to comment
Share on other sites

Just in case others are reading along.

Several things have been pointed-out to me and I have been reminded of others.

It would probably be an error to replace the 2500uF and 470uF capacitors in the AR-9 with a bundle of paralleled capacitors.

I was thinking of that with 5 500uf caps but what is the cost of the new 2500uf caps?

Link to comment
Share on other sites

I was thinking of that with 5 500uf caps but what is the cost of the new 2500uf caps?

I can't think of any reason not-to build a 2500uF capacitor out of smaller capacitors in the 10pi and LST. That capacitor has nothing to do with anything except amplifier preservation. It is supposed to be audibly transparent.

I don't *know* anything yet, so I hate to speculate about the AR-9. I think, no, that's too technical a term... I have an intuitive suspicion loosely based on almost no understanding that the higher ESR and inductance of a large capacitor is going to be important (on a curve - I don't think you can get the same effect from a fixed resistor and an inductor).

Since the AR-9's crossover is split and signals going to the tweeters don't ever have to pass these two big woofer capacitors, the ESR and inductance can be safely ignored above and below the woofer's operating frequencies. In theory.

Bret

Link to comment
Share on other sites

I can't think of any reason not-to build a 2500uF capacitor out of smaller capacitors in the 10pi and LST. That capacitor has nothing to do with anything except amplifier preservation. It is supposed to be audibly transparent.

I don't *know* anything yet, so I hate to speculate about the AR-9. I think, no, that's too technical a term... I have an intuitive suspicion loosely based on almost no understanding that the higher ESR and inductance of a large capacitor is going to be important (on a curve - I don't think you can get the same effect from a fixed resistor and an inductor).

Since the AR-9's crossover is split and signals going to the tweeters don't ever have to pass these two big woofer capacitors, the ESR and inductance can be safely ignored above and below the woofer's operating frequencies. In theory.

Bret

The crossover in the AR9, for most people, is not split, unless of course they are bi-amping and then it is split. A bit different, but the same class of speaker. A few years ago, I recapped my pair of AR90s, including the bass section. Simply put, the new caps in the bass crossover removed a fair amount of "mud". I'd expect the same with the ar9, but would go for an oil filled cap because of the size rather than poly.

Link to comment
Share on other sites

The crossover in the AR9, for most people, is not split, unless of course they are bi-amping and then it is split.

We've got a different way of saying the same thing, I think. The 9s crossover absolutely is electrically split. It is not necessary for the frequencies above those covered by the woofer section to "pass-through" the woofer section in order to make an A/C circuit like it is on a 90. You can completely disconnect the woofer on the 9 and the upper cabinet works just fine. If you connect the upper drivers of the right speaker to the lower drivers of the left in parallel, they both play and are still entirely independent of one-another. That's what I meant by "split."

Like you, when I recapped a 90, the woofer cap did clean it up a lot. The 9 has that capacitor, too, it's a 470uF, but the implementation is different.

If you are suggesting that there is "backlash" of some sort from garbage coming off the woofer caps that affects the upper cabinet, then I couldn't argue. I wouldn't know how to even test for that, but I've read over-and-over about the "back EMF".

Realizing that everything affects everything else, you may well be right that they will "clean-up" with new caps in them. The big cap, though, is part of this impedance compensation circuit and shouldn't be audible.

Let's see how wrong I am.

Again.

I'd rather heat the Weller than theorize, anyway.

All sorts of stuff that shouldn't make a difference seems to. I'll let you know how it goes.

But I promise I won't go all "wine aficionado" on you. "The woofer started with a slight hint of JBL, expanding rapidly to the complexity of French surround foam, and finished with a powerful International Paper Company texture, more reminiscent of Charmin than Two-ply Northern. Of course, this being the AR-9, the double roll."

Link to comment
Share on other sites

  • 9 years later...

An oldie but a goodie!  And after reading this, I'm even more excited to do a recap on my 90s.  Even though it's subjective (no slight intended), it's good to hear that the parallel caps make a different in sound quality too.  So, I hope everyone will forgive the 'bump'.  But, this deserves to be reread every decade or so!   :P

Link to comment
Share on other sites

28 minutes ago, Stimpy said:

An oldie but a goodie!  And after reading this, I'm even more excited to do a recap on my 90s.  Even though it's subjective (no slight intended), it's good to hear that the parallel caps make a different in sound quality too.  So, I hope everyone will forgive the 'bump'.  But, this deserves to be reread every decade or so!   :P

Oh no.   I hadn't seen this thread before.   Yet another component on which to invest time and money.

Link to comment
Share on other sites

7 minutes ago, DavidR said:

@Stimpy I don't believe the 2500uF is a shunt cap. Looks like a 'tank' circuit IMO. However, the only difference in sound would be from having the crossover points crossover at the intended point.

Hey David,

Thanks for the clarification.  Sorry to misidentify the 2500uF cap.  It's just my bad habit of picturing the 90 (which I own) schematic in my head, instead of the correct 9 schematic, when I think about these capacitors.  The 2 speaker designs are definitely different in the bass circuit!

Stimpy

Link to comment
Share on other sites

On 12/7/2007 at 1:59 AM, Diamonds&Rust said:

Just in case others are reading along.

Several things have been pointed-out to me and I have been reminded of others.

It would probably be an error to replace the 2500uF and 470uF capacitors in the AR-9 with a bundle of paralleled capacitors.

I'm getting ready to replace the Callins 2500uF and 470uF caps in the AR9.   The 470uF is available as a single NPE, but the 2500uF requires 3 to 5 NPEs depending on the chosen manufacturer.   Why does Bret say that it would probably be an error to install a bundle of paralleled capacitors in the AR9?   We just went through this discussion in the "Capacitor Myths" thread last month.  Pete B even said that paralleling a bunch of NPEs to replace the one of the big cans was a good idea.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...