Capacitors for RF7 crossover upgrade

[bbman]

I'm going to upgrade the crossover components in my RF7s and I've heard of a few different brands being used. Could anyone knowledgeable on the characteristics of these capacitors explain how they differ and make a recommendation on what I should use?

[mas]

I think blue ones are pretty.

Why do I get the feeling that this is another solution in search of a problem?

At least, assuming successful surgery, the extent of any potential harm will be limited only to your pocketbook. But, by all means, have fun.

[Deang]

"The Law of Conservation of Ignorance: A false conclusion once arrived at & widely accepted is not easily dislodged, & the less it is understood the more tenaciously it is held."

[mas]

My point, exactly.

Evidently someone needs funding to buy a second pair of Jubilees.

[cjgeraci]

My point, exactly.

Evidently someone needs funding to buy a second pair of Jubilees.

Whoosh!!

Carl.

[Deang]

Evidently someone needs funding to buy a second pair of Jubilees.

Nope.

He's obviously doing the work himself, and I don't sell capacitors. My opinion regarding this topic was formed long before I starting building for others -- which I'm now doing on a limited basis. I just think you're wrong about this issue.

[BEC]

That list really needs a "None of the Above" category so I could vote.

Bob

 

[Deang]

Why do you say that?

[mas]

OK. Since this site has been rather quiet for a while, how
about I open with the assumption that many seem to have that: Klipsch speakers are
the best and that everyone needs them - bar nonealthough some here feel that the Klipsch engineers and
golden ears (or accountants!) are too stupid to use the decent capacitors (read: they
intentionally choose to use crappy components) in their fantastic loudspeakers.

Do I have to stomp on the proverbial anthill further?

Hey, Don't get upset at me, I am merely the messenger! I am simply observing the trend that many have that you must buy the speaker and promply run out and modify the (fill in the blank) - in this case capacitors - in order t have a truly quality piece of equipment.

You know, I love tweaking.

But the older I get, I cant help but be enamored with the
trend to tweak before we identify an anomaly to be rectified. So, what most
believe is simply that change is good. And for many, by definition, change = improvement.

But is the suggestion that we should identify a problem that requires correction before 'making corrections' really so controversial? And why are so many evidently threatened by the mere suggestion? That is what really intrigues me! After all, there may be a very real problem that would definiately benefit from a real modification that would result in a real improvement! And I would be among the forst to applaud such a development.

Of course, we can expand this issue to include the oft
asinine emotional debate over purely subjective versus objective corroboration of
issues. And I expect the purely subjective crowd to promptly berate me for
holding my subjective opinion that might suggest that if indeed different caps
sound differently, which I BTW will admit, that we should be able to qualify
the issue in more objective terms than simply I like them.

Of course, what issues would caps directly affect? Hmmm.

Now, ignoring the fact that so many are enamored with
passive crossovers and their fundamental inherent limitations (which in the
Heritage line drove Dick Heyser to dump the Klipschorns) and their complete inability
to address significant signal alignment differences in acoustic center offsets often measured in
multiple feet

But lets try to keep the discussion in a tame generic level
without in depth technical gobbledygook.

In this case we could simply start with another potential difference that
would have an audible effect. And that would be passive reactive component
phase error.

Lets see, is it possible that various constructions and
manufacturers products could exhibit different phase relations for a given
value component?

And If such
variations, and their subsequent drop in incorporation into a circuit without
a comprehensive examination of the total circuit were sufficient to change the
phase of the output signal which defined
a portion of the passband defined by two sources reproducing the same passband
(hence, a crossover), just what would be the effect of the superposition of the
two audible signals which now exhibit a change in their phase relationship?

Oh yeah, one such result would be a change in the frequency
response as well as the polar distribution of the resultant acoustic wave which
would manifest itself in an apparent change in frequency at a given listening
position. Yup, its different all right, and I guess the money and effort are by
definition worthwhile.

And perhaps, if the phase relationships between
acoustical centers are initially less than 360 degrees, we might be onto
something (although this does imply that the original designers had their heads
somewhere else when they should have been paying attention, instead of tossing a bunch more passive components in in an attempt to resolve such group delay issues...But hey, few know what this is anyway...

But we dare not say that, do we? Or we may simply be making a
change. Just that, a change. And as we all know, change is good.

And that is but ONE aspect from which such a modification
can be viewed. Although actually connecting the micro electrical view with the
macro acoustical view is a trait conspicuously lacking in too much of the audio
world in general.

So, not only do we have a fundamental issue with the use of
a passive crossover to begin with, we have additional potential ramifications
of changing reactive components. And the irony is that many of these characteristics
are able to be quantified and even objectively verifiedboth electrically at
the circuit level as well as acoustically with regards to variations in comb
filtering and polar anomalies. In other words, we have easily performed tests to determine the nature of both the electrical and acoustical realms in both the frequency and time domains.

Now, there are quite a few on this site who are well versed
in electrical theory and practice. There are also a few (fewer) who know
acoustics. One might think that with the abundance of talent on the site, and
with the resources of a manufacturer with oodles of test equipment, that such
an issue could be discussed intelligently and intelligibly. And such factors that
result in real subjective listening differences
could (and do!) in fact lie within the objective realm which could correlate both
the problem and the solution.

And this is not even to mention the use of Nyquist and
Heyser spirals that would directly show the complete complex behavior of the impedance
directly in every aspect available in time, frequency, phase, potential,
energy, kinetic energy, minimum/non-minimum
phase, etc.

In other words, such a dilemma can be quantified and
qualified 100 ways from Tuesday.

There is NO reason, except the reluctance of some to get off
their posteriors that this cannot be done...that is, if there is truly Trouble in River City!

And the result could be objective reasons to not only perform
the mod, but a basis for criteria for practical decision making with regards to
component choice and circuit topology preferences.

The bottom line, at the risk of making lots of folks mad:

Identifying a problem to be fixed is prerequisite to fixing
the problem. And to date, I have read few well defined problems, and even fewer
well defined solutions that effectively address the problem without introducing
a slew of new ones.

And I find it fascinating that to suggest that more effort
be placed in this area prior to our running about solving that which we do not
quite understand in the name of different is better, seems like such
a threatening suggestion.

Does it really?????

[PrestonTom]

Gee the poor guy asked a question and look at what has happened. Let me toss in a comment (and not vote) that is meant to be helpful.

You have listed a number of expensive caps. The answers you get will probably not relate to a physical description of the cap (e.g., ESR, reliability, power handling, distortion etc). You will get a number of subjective comments about "detail" "opening the sound stage" & "removing the veil" etc. I can not imagine that another persons subjective description (using different music, different set ups, different sources, different rooms, different playback levels, different aspirations etc) will be that meaningful. For instance, one person's "too forward sounding" may be another person's "wonderfully clean & accurate". But that is the price for a question posed in this manner.

Now the helpful part, finally.

My experience was to replace the caps on a Klipschorn (AA network) without changing the values or doing any sort of re-design. The caps were about 25 years old. I replaced them with Solen metal polyprops (about $20 total). The results were that there was a bit more energy in the highs (consistent with the notion that old caps develop some series resistance). While the results were not dramatic, the change was noticeable. I had no measurement equipment avaliable, but I suspect I could have measured a small amount of boost in the high end.

For $20 and a couple of hours of work (good time to clean all the oxidized surfaces) it was good value. This is the most accurate and objective way I can describe it. My suspicion is that "fancy" caps will make little, if any, further difference (a difference does not always mean an improvement). But I am a skeptic, or maybe my Klipschorns and solid state equipment were not "revealing" enough, or maybe I have a bad set of ears.

Good Luck,

-Tom

[capo72]

Tom, you did a good job of not making me feel dumb............Thanx!

Jeremy

[PrestonTom]

Something is bothering me....

There was a suggestion above that Dean's comments about fancy caps is driven by a possible financial interest on his part (he has a side venture of re-building/modifying crossovers). That bothers me. While I disagree with Dean about the benefit of fancy caps, I have never known him to try and sell something to someone that they did not need or try an BS someone about the "mystique" of his products. In fact, he has been fairly laid back about trying to drum up business on the forum. Personally, I find that very refreshing.

that's of my chest ...

-Tom

[mas]

Funny, I have the same feeling about an atmosphere that has new folks feeling that they have to necessarily replace new caps in a new piece of equipment for it to sound good.

If you simply want to change them, great. But don't do it simply because you have gotten the idea that you must and that any change will necessarily be beneficial. And I am not referring to changing out 25+ year old caps that most likely have denigrated in quality over time.

[bbman]

Wow, I didn't realize the topic was so controversial. I really do appreciate hearing everyones point of view though.

[Deang]

I suggest you move forward using Kimbers. They are of the same quality as the Auricaps, but have two advantages with an RF-7 rebuild, 1) the leads are thin enough that they fit into the existing holes in the PCB, and 2) the size allows placement without using longer spacers and screws when you stack the boards back together. The new 12uF will occupy the space presently utilized by both of the 5uF and 12uF small epoxy coated ovals. Place the new 5uF on top of the two inductors. Lightly scuff all contact surfaces with 220 sandpaper before applying the hot melt. If, as some, you feel the HF output is just a bit much and running over the cones, you can bring down the horn's resonant peak by paralleling a 10 ohm resistor with the 2 ohm adjacent to the 18uF cap. Speaking of resistors -- replace the HF resistors with Mills non-inductives, and reverse mount them (under the board). The 18uF cap previously mentioned doesn't have to be a Kimber -- I use a Janzten Cross-cap in that position. If you go with Kimbers, you'll have to parallel two 9's -- it's a very tight fit. I agree that different doesn't always mean better, but this is a case where it simply is.

Thanks Tom. Keep in mind though that what constitutes "fancy" -- is higher quality film and lead terminations, which translates to better measurements and sound. One brand of cap I use are hermetically sealed in brass containers with soldered terminations, I imagine those will out-live the purchaser, so not that expensive if viewed from that perspective.

[Southern]

When I purchased my RF-7's used they came with DeanG crossovers. While I can't AB compare them to stock RF-7's they sound fantastic. Since I have a stock RC-7 it does not blend well with the RF-7's with music. I plan on recapping the RC-7 in the future.

[bbman]

Dean, thanks for the advice. I think I may just lay everything out on a piece of baltic birch. Do you have any pictures I could go of to get an idea of how to organize everything?

[psg]

My experience was to replace the caps on a Klipschorn (AA network) without changing the values or doing any sort of re-design. The caps were about 25 years old. I replaced them with Solen metal polyprops (about $20 total). The results were that there was a bit more energy in the highs (consistent with the notion that old caps develop some series resistance). While the results were not dramatic, the change was noticeable. I had no measurement equipment avaliable, but I suspect I could have measured a small amount of boost in the high end.

For $20 and a couple of hours of work (good time to clean all the oxidized surfaces) it was good value. This is the most accurate and objective way I can describe it. My suspicion is that "fancy" caps will make little, if any, further difference (a difference does not always mean an improvement). But I am a skeptic, or maybe my Klipschorns and solid state equipment were not "revealing" enough, or maybe I have a bad set of ears.

Same here, with the same result as well, except that I used Auricaps for the cheaper caps and still kept the cost very reasonable. However, I don't doubt that a new crossover design will change the sound quite a bit. Some day I'll get myself some DeanG or ALK crossover to try out. There's even one for us Bob Crite Tweeter users.