Capacitor substitutions

[Norm Dresner]

Back to the HP 200CD. Since I haven't had any experience with tube
electronics in the last 50 years, I'm a little rusty on some of the
subtleties of component specifying. I've been looking at the electrolytic
capacitors in the circuit diagram and my stock with an eye to simply
replacing all of them in an effort to resurrect the instrument to "near new"
condition.

I have (or can get) an exact replacement for the 100 uF 100V non-polarized
capacitor. That's the easy one.

There's a 40 uF, 450 V as the final filter in the Power Supply filtering. I
can easily get either a 100 uF 330V or a 33 uF 300 V. The DC working
voltage for this is 170V so it would seem that the original 450V may have
been based in part on inventory rather than exact electronic need. Is this
reasonable?

There's a three section filter cap in the Power Supply, each section of
which is 10 uF at 450V and the working DC voltage is 310V. I have some 20
uF 400V Polypropylene capacitors that would fit. Are these reasonable
replacement or should I just order the "correct" values from somewhere?

There's a .082 uF 600V Paper capacitor that's part of an RC filter to a tube
grid that works at 20V DC. Would a .082 uF 400V Mylar be an adequate
substitute?

Finally, there's a capacitor that connects the Low output terminal to
chassis ground (there's also an optional external strap as well). The
schematic shows this as 40 pF , "adjusted at factory for optimum
performance. Average value shown. Part may be omitted." The parts list
gives this as a 1.5 pF Titanium Dioxide capacitor. There are three
strategies that I see here:
1. Leave it in place
2. Remove it
3. Remove, measure, and replace with a mica capacitor of appropriate
value
What's the best choice?

Also, assuming that I replace all 5 tubes with new ones and replace the
electrolytic capacitors, and clean the unit, are there any other
before-testing substitutions that should be made?

TIA
Norm

 

[Asimov]

"Norm Dresner" bravely wrote to "All" (26 Sep 04 07:49:55)
--- on the heady topic of "Capacitor substitutions"

ND> From: "Norm Dresner" <[email protected]>
ND> Back to the HP 200CD. Since I haven't had any experience with tube
ND> electronics in the last 50 years, I'm a little rusty on some of the
ND> subtleties of component specifying. I've been looking at the
ND> electrolytic capacitors in the circuit diagram and my stock with an eye
ND> to simply replacing all of them in an effort to resurrect the
ND> instrument to "near new" condition.

ND> I have (or can get) an exact replacement for the 100 uF 100V
ND> non-polarized capacitor. That's the easy one.

ND> There's a 40 uF, 450 V as the final filter in the Power Supply
ND> filtering. I can easily get either a 100 uF 330V or a 33 uF 300 V.
ND> The DC working voltage for this is 170V so it would seem that the
ND> original 450V may have been based in part on inventory rather than
ND> exact electronic need. Is this reasonable?

The reason for the high voltage rating may be due to the expected surge
voltage and not the operating voltage. You are certainly aware that for
a brief time, while the tube cathodes reach operating temperature, the
plate behaves as essentially an open circuit and this allows the supply
voltage to rise what may alarmingly seem to be excessive. In the tube
days electros were often stamped with a surge rating some 20 to 50% over
the working voltage. However, the 450 Volt rating here may, indeed as
you say, be simply due to inventory but that is, if the 170 Volt
measurement isn't the result of a fault!


ND> There's a three section filter cap in the Power Supply, each section
ND> of which is 10 uF at 450V and the working DC voltage is 310V. I have
ND> some 20 uF 400V Polypropylene capacitors that would fit. Are these
ND> reasonable replacement or should I just order the "correct" values from
ND> somewhere?

Restating the above about "surge voltage" rating...


ND> There's a .082 uF 600V Paper capacitor that's part of an RC filter to
ND> a tube grid that works at 20V DC. Would a .082 uF 400V Mylar be an
ND> adequate substitute?

Usually one should never downgrade the voltage rating. Often the
seemingly high voltage rating hides a completely different design
objective. Perhaps the 600 V part had a much lower DC leakage,
especially important when coupling input grids from a previous stage's
plate voltage. Or yet still, an RF circuit may require a 600 Volt part
even though the supply may only be 20 V DC, this due to resonnant
effects, etc...


ND> Finally, there's a capacitor that connects the Low output terminal to
ND> chassis ground (there's also an optional external strap as well). The
ND> schematic shows this as 40 pF , "adjusted at factory for optimum
ND> performance. Average value shown. Part may be omitted." The parts
ND> list gives this as a 1.5 pF Titanium Dioxide capacitor. There are
ND> three strategies that I see here:
ND> 1. Leave it in place
ND> 2. Remove it
ND> 3. Remove, measure, and replace with a mica capacitor of
ND> appropriate value
ND> What's the best choice?

ND> Also, assuming that I replace all 5 tubes with new ones and replace
ND> the electrolytic capacitors, and clean the unit, are there any other
ND> before-testing substitutions that should be made?

Don't second guess an original design. The parts were often chosen for
proprietary reasons which often may not be easily deterimined. However,
with sufficient experience and theory some modern parts can be adaptable
to an older circuit design. For instance all paper capacitors are open
season for mods.

The safest repair strategy is to leave well enough alone and only
replace what has clearly been identified as defective. I have radios
older than many reading this which have original good electros in them.
Often electros only need to be exercised regularly to keep them formed.

I would suggest testing your electros rather than outright replacing
them willynilly. My rule of thumb is, if the caps have low enough esr,
have normal leakage at the rated voltage, and no obvious physical
deterioration then leave well enough alone. Get a hold of the
rec.antiques.radio+phono faq on replacing electros for another opinion.

A*s*i*m*o*v

.... Dunno if we'll get that past the CSA und UL 'owever.

 

[Norm Dresner]

"Norm Dresner" bravely wrote to "All" (26 Sep 04 07:49:55)
--- on the heady topic of "Capacitor substitutions"

ND> From: "Norm Dresner" <[email protected]>
ND> Back to the HP 200CD. Since I haven't had any experience with tube
ND> electronics in the last 50 years, I'm a little rusty on some of the
ND> subtleties of component specifying. I've been looking at the
ND> electrolytic capacitors in the circuit diagram and my stock with an eye
ND> to simply replacing all of them in an effort to resurrect the
ND> instrument to "near new" condition.

ND> I have (or can get) an exact replacement for the 100 uF 100V
ND> non-polarized capacitor. That's the easy one.

ND> There's a 40 uF, 450 V as the final filter in the Power Supply
ND> filtering. I can easily get either a 100 uF 330V or a 33 uF 300 V.
ND> The DC working voltage for this is 170V so it would seem that the
ND> original 450V may have been based in part on inventory rather than
ND> exact electronic need. Is this reasonable?

The reason for the high voltage rating may be due to the expected surge
voltage and not the operating voltage. You are certainly aware that for
a brief time, while the tube cathodes reach operating temperature, the
plate behaves as essentially an open circuit and this allows the supply
voltage to rise what may alarmingly seem to be excessive. In the tube
days electros were often stamped with a surge rating some 20 to 50% over
the working voltage. However, the 450 Volt rating here may, indeed as
you say, be simply due to inventory but that is, if the 170 Volt
measurement isn't the result of a fault!


ND> There's a three section filter cap in the Power Supply, each section
ND> of which is 10 uF at 450V and the working DC voltage is 310V. I have
ND> some 20 uF 400V Polypropylene capacitors that would fit. Are these
ND> reasonable replacement or should I just order the "correct" values from
ND> somewhere?

Restating the above about "surge voltage" rating...


ND> There's a .082 uF 600V Paper capacitor that's part of an RC filter to
ND> a tube grid that works at 20V DC. Would a .082 uF 400V Mylar be an
ND> adequate substitute?

Usually one should never downgrade the voltage rating. Often the
seemingly high voltage rating hides a completely different design
objective. Perhaps the 600 V part had a much lower DC leakage,
especially important when coupling input grids from a previous stage's
plate voltage. Or yet still, an RF circuit may require a 600 Volt part
even though the supply may only be 20 V DC, this due to resonnant
effects, etc...


ND> Finally, there's a capacitor that connects the Low output terminal to
ND> chassis ground (there's also an optional external strap as well). The
ND> schematic shows this as 40 pF , "adjusted at factory for optimum
ND> performance. Average value shown. Part may be omitted." The parts
ND> list gives this as a 1.5 pF Titanium Dioxide capacitor. There are
ND> three strategies that I see here:
ND> 1. Leave it in place
ND> 2. Remove it
ND> 3. Remove, measure, and replace with a mica capacitor of
ND> appropriate value
ND> What's the best choice?

ND> Also, assuming that I replace all 5 tubes with new ones and replace
ND> the electrolytic capacitors, and clean the unit, are there any other
ND> before-testing substitutions that should be made?

Don't second guess an original design. The parts were often chosen for
proprietary reasons which often may not be easily deterimined. However,
with sufficient experience and theory some modern parts can be adaptable
to an older circuit design. For instance all paper capacitors are open
season for mods.

The safest repair strategy is to leave well enough alone and only
replace what has clearly been identified as defective. I have radios
older than many reading this which have original good electros in them.
Often electros only need to be exercised regularly to keep them formed.

I would suggest testing your electros rather than outright replacing
them willynilly. My rule of thumb is, if the caps have low enough esr,
have normal leakage at the rated voltage, and no obvious physical
deterioration then leave well enough alone. Get a hold of the
rec.antiques.radio+phono faq on replacing electros for another opinion.

A*s*i*m*o*v

... Dunno if we'll get that past the CSA und UL 'owever.

Thanks for your thoughtful advice.

Norm

 

[JURB6006]

I dunno if you're still watching this thread, but it reminded me of something.

I recall repairing an ultrasonic cleaner. It was old, had tubes, but had
selenium rects in a voltage tripler. Something else.

Well one of the rects had fried and I had to use modern replacements. Two
diodes in series with two resistors (I don't remember the value right now) in
parallel in the position of each diode. If one section of a "stack" like that
is way more effecient, it'll stress the other components more. I tried to
minimize that.

JURB

 

[Norm Dresner]

I dunno if you're still watching this thread, but it reminded me of something.

I recall repairing an ultrasonic cleaner. It was old, had tubes, but had
selenium rects in a voltage tripler. Something else.

Well one of the rects had fried and I had to use modern replacements. Two
diodes in series with two resistors (I don't remember the value right now) in
parallel in the position of each diode. If one section of a "stack" like that
is way more effecient, it'll stress the other components more. I tried to
minimize that.

Interesting -- but unfortunately not applicable in this case: the rectifier
is a pure diode tube.

Thanks for the suggestion

Norm