[AMPS] Noises in electrolytic capacitors

Fri, 28 Sep 2001 07:48:47 EDT

--part1_29.1b65b2ec.28e5bd9f_boundary
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

>It sounds like you have a capacitor charging to too much voltage.=20
>That is probably why the old ones failed. If you did not test or=20
>replace the equalizing resistors you made a serious omission.

>Pay attention to the type of resistors you find. If you see that they=20
>use carbons, scrap them. Use metal film or wire wound resistors,=20
>but NEVER use carbon resistors in voltage equalizing applications.

>The normal leakage current tolerances in electrolytics generally=20
>requires you use 100 ohms per volt or less of equalizing=20
>resistance, if you run the capacitors near the working voltage=20
>rating.=A0=A0=20
73, Tom W8JI
W8JI@contesting.com=20

Thanks for the tip on the equalizing/bleeder resistors. I did make a serious=
=20
omission in not testing them. I guess the reason I didn't think to verify=20
their values is that they are 20K 20W units that look in flawless condition=20
and I have always run a muffin fan over the PS board to keep things cool. I=20
also run a fan over the tubes, given the double layer of perforated aluminum=
=20
over the top of the amp which does not allow reasonable airflow. I have=20
always felt that a more open mesh in that area, in particular, would be a bi=
g=20
improvement.

Anyway, I plan to pull all the caps this weekend and check the resistors. I=20
hate to got to a smaller wattage, higher resistance design even though I did=
=20
that successfully in my SB200, which clearly showed signs of heat damage to=20
the caps. The bakelite (or similar) board with the caps and equalizing=20
resistors is nicely done and a retrofit would be a jury rig. They use solder=
=20
lugs and nicely dressed solid wire for the interconnections, one of which wa=
s=20
never soldered (which I only discovered when I pulled this apart for repair.=
)

I'll check the resistor on the cap which initially failed first, but verify=20
the value of the others and report back for those interested in this saga.

73, Mark

--part1_29.1b65b2ec.28e5bd9f_boundary
Content-Type: text/html; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

<HTML>&gt;It sounds like you ha=
ve a capacitor charging to too much voltage.=20
 &gt;That is probably why the old ones failed. If you did not test or=20
 &gt;replace the equalizing resistors you made a serious omission.
 
 &gt;Pay attention to the type of resistors you find. If you see that the=
y=20
 &gt;use carbons, scrap them. Use metal film or wire wound resistors,=20
 &gt;but NEVER use carbon resistors in voltage equalizing applications.
 
 &gt;The normal leakage current tolerances in electrolytics generally=20
 &gt;requires you use 100 ohms per volt or less of equalizing=20
 &gt;resistance, if you run the capacitors near the working voltage=20
 &gt;rating.=A0=A0=20
 73, Tom W8JI
 W8JI@contesting.com=20
 </XMP>
 
 Thanks for the tip on the equalizing/bleeder resistor=
s. I did make a serious omission in not testing them. I guess the reason I d=
idn't think to verify their values is that they are 20K 20W units that look=20=
in flawless condition and I have always run a muffin fan over the PS board t=
o keep things cool. I also run a fan over the tubes, given the double layer=20=
of perforated aluminum over the top of the amp which does not allow reasonab=
le airflow. I have always felt that a more open mesh in that area, in partic=
ular, would be a big improvement.
 
 Anyway, I plan to pull all the caps this weekend and check the resistors=
. I hate to got to a smaller wattage, higher resistance design even though I=
 did that successfully in my SB200, which clearly showed signs of heat damag=
e to the caps. The bakelite (or similar) board with the caps and equalizing=20=
resistors is nicely done and a retrofit would be a jury rig. They use solder=
 lugs and nicely dressed solid wire for the interconnections, one of which w=
as never soldered (which I only discovered when I pulled this apart for repa=
ir.)
 
 I'll check the resistor on the cap which initially failed first, but ver=
ify the value of the others and report back for those interested in this sag=
a.
 
 73, Mark
 </HTML>

--part1_29.1b65b2ec.28e5bd9f_boundary--

--
FAQ on WWW:               http://www.contesting.com/FAQ/amps
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com