Hi,
I got many great replies to my original post,
http://groups.google.com/group/sci.e...849027dfe5878/

With work getting in the way, I have been a bit slow at ordering and
replacing the caps in my AR 91s speakers (25 yrs old). I hope to find
some time over the holidays.

Reviewing other threads on the subject of, and talking to a couple of
friends in the my IT dept, industry, I have a few more questions
before ordering parts.

Should I replace the resistors? The speakers function, and measuring
the resistors with a multi-meter find them within tolerance. Any other
tests I can perform? The resistors and caps leads are twisted and
soldered together, so I have to "remove" them anyway.

I was thinking about use a connector bar, screw or solder type, versus
twisting and soldering the component wires together. Any comments on
that idea?

From lurking in this forum, I know that most of you that reply to
posts do not believe in high-end "snake oil" caps. Humor me, or go
enjoy a glass of egg-nog. Reading other forums, people recommend using
a quality cap in parallel with a smaller value "snake oil" cap. The
result is supposedly better sound, without spending big $$. This
doesn't quite make sense to me, since a larger value cap will roll off
at a lower frequency and thus a larger portion of the signal (for a
mid or high speaker) will be flowing through it. Is this a correct
assessment or is my "basic" understanding of caps or electronics
insufficient to understand why this recommendation has merit.


Cheers!
Christopher

Should I replace the resistors? The speakers function, and measuring
the resistors with a multi-meter find them within tolerance. Any other
tests I can perform? The resistors and caps leads are twisted and
soldered together, so I have to "remove" them anyway.

If they're in good condition, there's no reason to replace them.


I was thinking about use a connector bar, screw or solder type,
versus twisting and soldering the component wires together.
Any comments on that idea?

For reliability, all connections should be soldered, regardless of how the
components are "anchored"..


From lurking in this forum, I know that most of you that reply to
posts do not believe in high-end "snake oil" caps. Humor me, or go
enjoy a glass of egg-nog. Reading other forums, people recommend
using a quality cap in parallel with a smaller value "snake oil" cap.
The result is supposedly better sound, without spending big $$. This
doesn't quite make sense to me, since a larger value cap will roll
off at a lower frequency and thus a larger portion of the signal (for a
mid or high speaker) will be flowing through it. Is this a correct
assessment or is my "basic" understanding of caps or electronics
insufficient to understand why this recommendation has merit.

It is my opinion that switching from electrolytic to Mylar caps will effect
a noticeable improvement in the sound. Mylar caps are not horribly
expensive, so it won't cost much to experiment. You can modify one speaker
and compare it with the unmodified one to see if you agree. I will not
discuss this issue further.

You should replace the caps directly, rather than paralleling them. (Your
point is correct, at least in theory.) A 10uF electrolytic should be
replaced by a 10uF Mylar, and so on.

Mylar caps should also last longer than electrolytics, as they won't dry
out.

On Tue, 18 Dec 2007 13:52:17 -0800 (PST), cj
wrote:

Hi,
I got many great replies to my original post,
http://groups.google.com/group/sci.e...849027dfe5878/

With work getting in the way, I have been a bit slow at ordering and
replacing the caps in my AR 91s speakers (25 yrs old). I hope to find
some time over the holidays.

Reviewing other threads on the subject of, and talking to a couple of
friends in the my IT dept, industry, I have a few more questions
before ordering parts.

Should I replace the resistors? The speakers function, and measuring
the resistors with a multi-meter find them within tolerance. Any
other
tests I can perform? The resistors and caps leads are twisted and
soldered together, so I have to "remove" them anyway.

I was thinking about use a connector bar, screw or solder type,
versus
twisting and soldering the component wires together. Any comments on
that idea?

From lurking in this forum, I know that most of you that reply to
posts do not believe in high-end "snake oil" caps. Humor me, or go
enjoy a glass of egg-nog. Reading other forums, people recommend
using
a quality cap in parallel with a smaller value "snake oil" cap. The
result is supposedly better sound, without spending big $$. This
doesn't quite make sense to me, since a larger value cap will roll
off
at a lower frequency and thus a larger portion of the signal (for a
mid or high speaker) will be flowing through it. Is this a correct
assessment or is my "basic" understanding of caps or electronics
insufficient to understand why this recommendation has merit.


Cheers!
Christopher

Looking at the AR91 schematic here
http://www.classicspeakerpages.net/l...schematic.html
I notice that the original caps are all non-polarised electros. While
changing to film type caps (mylar MKT or polypropylene MKP) is thought
by many to improve the sound quality, this may prove to be a moot
point in reality. Those 24 uF and 40uF caps will take several
paralleled MKT's or MKP's to get the right value and the cost won't be
all that small. As suggested by William you might care to modify one
unit and compare it with a standard x-over. Listening test will
require a suitable speaker switchover arrangement.

On Dec 18, 4:52 pm, cj wrote:
Hi,
I got many great replies to my original post,http://groups.google.com/group/sci.e...rowse_thread/t...

With work getting in the way, I have been a bit slow at ordering and
replacing the caps in my AR 91s speakers (25 yrs old). I hope to find
some time over the holidays.

Reviewing other threads on the subject of, and talking to a couple of
friends in the my IT dept, industry, I have a few more questions
before ordering parts.

Should I replace the resistors? The speakers function, and measuring
the resistors with a multi-meter find them within tolerance. Any other
tests I can perform? The resistors and caps leads are twisted and
soldered together, so I have to "remove" them anyway.

I was thinking about use a connector bar, screw or solder type, versus
twisting and soldering the component wires together. Any comments on
that idea?

From lurking in this forum, I know that most of you that reply to
posts do not believe in high-end "snake oil" caps. Humor me, or go
enjoy a glass of egg-nog. Reading other forums, people recommend using
a quality cap in parallel with a smaller value "snake oil" cap. The
result is supposedly better sound, without spending big $$. This
doesn't quite make sense to me, since a larger value cap will roll off
at a lower frequency and thus a larger portion of the signal (for a
mid or high speaker) will be flowing through it. Is this a correct
assessment or is my "basic" understanding of caps or electronics
insufficient to understand why this recommendation has merit.

Cheers!
Christopher

That sounds right to me, but whats the point ?

Electrolytics are always plagued by tolerance, ESR, aging, and
voltage limits. A quality
speaker will be voiced by careful planning, listening, and measuring.
Anything you change can affect the sound, not necessarily for the
better.
Resistors should be Ok indefinitely, unless they are cooked or have
mechanical
fatigue. After years of fooling around with internal crossovers while
making speakers, I got smart and
put the crossover components on the outside of the cabinet, and any
changes could
be made without reopening the cabinet. You can make a change, and
check for changes in
response, polar patterns, etc.. Sometimes it could take days or weeks
to determine if I was
satisfied with the sound.

greg

I notice that the original caps are all non-polarised electros. While
changing to film type caps (Mylar MKT or polypropylene MKP)

Polypropylene is probably overkill.


is thought by many to improve the sound quality, this may prove
to be a moot point. Those 24 uF and 40uF caps will take several
paralleled MKT's or MKP's to get the right value and the cost
won't be all that small.

Mylar caps aren't horribly expensive.

I would start by replacing only those caps in series with the drivers -- the
4uF and 24uF capacitors. You can then try replacing the two that are
directly in parallel, the 8uF and 40uF. The 100uF cap seems to be part of a
Zobel filter (don't ask) and probably doesn't need replacing, unless it's
"bad", in which case you can use an electrolytic.


As suggested by William, you might care to modify one unit
and compare it with a standard x-over. A listening test will
require a suitable speaker switchover arrangement.

Not necessarily. Before modifying a speaker, listen to a mono source and try
to get a sharp central image. Modify one speaker, and repeat the listening.
If the caps have changed the sound, the mono image should be at least
"different", and (hopefully) degraded, because the speakers no longer "sound
alike".

In article , "William Sommerwerck" wrote:
I notice that the original caps are all non-polarised electros. While
changing to film type caps (Mylar MKT or polypropylene MKP)

Polypropylene is probably overkill.


is thought by many to improve the sound quality, this may prove
to be a moot point. Those 24 uF and 40uF caps will take several
paralleled MKT's or MKP's to get the right value and the cost
won't be all that small.

Mylar caps aren't horribly expensive.

I would start by replacing only those caps in series with the drivers -- the
4uF and 24uF capacitors. You can then try replacing the two that are
directly in parallel, the 8uF and 40uF. The 100uF cap seems to be part of a
Zobel filter (don't ask) and probably doesn't need replacing, unless it's
"bad", in which case you can use an electrolytic.


As suggested by William, you might care to modify one unit
and compare it with a standard x-over. A listening test will
require a suitable speaker switchover arrangement.

Not necessarily. Before modifying a speaker, listen to a mono source and try
to get a sharp central image. Modify one speaker, and repeat the listening.
If the caps have changed the sound, the mono image should be at least
"different", and (hopefully) degraded, because the speakers no longer "sound
alike".

Pink noise is the only way to go for immediate recognition. Of course. if the
component can be changed instantly outside the cabinet, you only need
one speaker to hear any change.

greg