What type of capacitor should I use?

On Wednesday, 9 August 2017 20:07:16 UTC+1, Dave Platt wrote:

That word "film" bothers me, anyhow. Just what the heck is this film? Is
it just some sprayed on metalic particles? The old paper caps contained
foil, which to me means something that looks like tin foil used in the
kitchen. Maybe I'm wrong, but if my radio or (whatever) was designed for
caps made with foil, I'd prefer to use caps with foil, and whatever
plastics they used to replace the paper. In other words, I want caps
that most closely mimic the original caps, except without the paper.

Then, what you are looking for would be what's sold as "film and foil"
capacitors.

Digi-Key sells e.g. the CDE "WMF" series, which use metal foil, a
polyester dielectric film, "non-inductive" winding (which usually
means that the foil sticks out past the end of the film, and the turns
of the foil are bonded directly to the wire lead), and epoxy seals at
either end. These, or similar, might be the closest thing to a
"plastic drop-in replacement" for your old paper-and-foil caps.

The film-and-foil design is often preferred for applications where
there's a high peak current (e.g. pulse applications) as the current
carrying capability of the foil is considerable.

The other (more common) option is "metallized film". In this design,
the dielectric film (polyester, polypropylene, etc.) has a thin
conductive metal layer deposited on one side - typically via vaccum
deposition I believe. It's not discrete particles of any real size.
Two layers of this film are then spiral-wound together, with the wire
leads being bonded to the leads at the ends.

Metallized-film (sometimes "stacked film") tend to be less expensive,
and (I think) more capacity available per volume at any give voltage
rating because the conductive layer is so thin.

They are also said to have a reliability advantage, at least
potentially. If a pinhole develops in the film and two adjacent
layers short together, the short-circuit current can burn away the
conductive film right around the hole, "healing" the short.
Apparently this only works well when the film is a material with a
high oxygen content.

http://www.ecicaps.com/tech-tools/technical-papers/self-healing-affect-metallized-capacitors/

Plastic film caps (but not polystyrene) are the best choice for valve radios. There is no upside in this situation to separate foils, they destabilise capacity value some, are unable to self heal, are bulkier per farad and cost more. Pick high voltage ratings to ensure they all last a lifetime.

Trying to replicate all the defects of the original caps is pointless, there is no upside to it. That isn't true of some parts but it is of caps.


NT

 

[Michael A. Terrell]

tabbypurr@gmail.com wrote:

On Wednesday, 9 August 2017 19:50:17 UTC+1, Michael Terrell wrote:

Ceramic capacitors are mostly cheap crap. Some have a wide change in
capacitance, at different voltages. Some are microphonic, and their
normal failure mode is a dead short. About the only place they should be
used are for DC bybass, or in low power tuned circuits.

Hmmm. I've not found them crap. Excluding MLCCs, which are a different animal to the rest, ceramics are immune to the melting problems of some plastic film caps, their light weight aids robustness, they're small, they typically handle way above rated voltage without trouble if ever necessary. Slight microphony is not a problem in lots of applications - it is only very slight. The C/V gotcha only applies to some, and obviously one need be aware of it at design time and choose accordingly.

All parts have their issues, and ceramics behave fine and are very reliable in a lot of situations. As with everything, not in every situation.

If the originals were paper, I always use film capacitors. I've had
too many shorted Ceramics in HV circuits, and I've replaced too many
shorted disk capacitors in 5V circuits. Some have burnt holes in PC
boards, when they are on a high current rail. "The C/V gotcha" can cause
distortion in audio circuits, or cause an AGC circuit to 'pump'.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

On Wednesday, 9 August 2017 22:02:04 UTC+1, Michael Terrell wrote:

tabbypurr wrote:
On Wednesday, 9 August 2017 19:50:17 UTC+1, Michael Terrell wrote:

Ceramic capacitors are mostly cheap crap. Some have a wide change in
capacitance, at different voltages. Some are microphonic, and their
normal failure mode is a dead short. About the only place they should be
used are for DC bybass, or in low power tuned circuits.

Hmmm. I've not found them crap. Excluding MLCCs, which are a different animal to the rest, ceramics are immune to the melting problems of some plastic film caps, their light weight aids robustness, they're small, they typically handle way above rated voltage without trouble if ever necessary. Slight microphony is not a problem in lots of applications - it is only very slight. The C/V gotcha only applies to some, and obviously one need be aware of it at design time and choose accordingly.

All parts have their issues, and ceramics behave fine and are very reliable in a lot of situations. As with everything, not in every situation.



If the originals were paper, I always use film capacitors. I've had
too many shorted Ceramics in HV circuits, and I've replaced too many
shorted disk capacitors in 5V circuits. Some have burnt holes in PC
boards, when they are on a high current rail. "The C/V gotcha" can cause
distortion in audio circuits, or cause an AGC circuit to 'pump'.

Those last 2 are examples of using them where not suited. If I use ceramics as audio decoupling they need to either not have the C/V gotcha or be big enough to still have enough capacity down at 20Hz to not get in the way AND the dc component across them needs to remain steady.


NT

 

[rickman]

tabbypurr@gmail.com wrote on 8/9/2017 8:15 PM:

On Wednesday, 9 August 2017 22:02:04 UTC+1, Michael Terrell wrote:
tabbypurr wrote:
On Wednesday, 9 August 2017 19:50:17 UTC+1, Michael Terrell wrote:

Ceramic capacitors are mostly cheap crap. Some have a wide change in
capacitance, at different voltages. Some are microphonic, and their
normal failure mode is a dead short. About the only place they should be
used are for DC bybass, or in low power tuned circuits.

Hmmm. I've not found them crap. Excluding MLCCs, which are a different animal to the rest, ceramics are immune to the melting problems of some plastic film caps, their light weight aids robustness, they're small, they typically handle way above rated voltage without trouble if ever necessary. Slight microphony is not a problem in lots of applications - it is only very slight. The C/V gotcha only applies to some, and obviously one need be aware of it at design time and choose accordingly.

All parts have their issues, and ceramics behave fine and are very reliable in a lot of situations. As with everything, not in every situation.



If the originals were paper, I always use film capacitors. I've had
too many shorted Ceramics in HV circuits, and I've replaced too many
shorted disk capacitors in 5V circuits. Some have burnt holes in PC
boards, when they are on a high current rail. "The C/V gotcha" can cause
distortion in audio circuits, or cause an AGC circuit to 'pump'.

Those last 2 are examples of using them where not suited. If I use ceramics as audio decoupling they need to either not have the C/V gotcha or be big enough to still have enough capacity down at 20Hz to not get in the way AND the dc component across them needs to remain steady.

The C/V "gotcha" is best mitigated by using a part that is overrated for
voltage. The C/V effect is relative to the rated voltage so if that is
several times the circuit voltage there won't be much effect. It's when you
try to operate near the rated voltage that it causes problems. I've also
seen data that indicates the larger package parts have lower coefficients of
capacitance change with voltage.

You can also use C0G types which have a better dielectric which has a very
low coefficient of capacitance change with voltage.

--

Rick C

 

On Thursday, 10 August 2017 02:00:07 UTC+1, rickman wrote:

tabbypurr wrote on 8/9/2017 8:15 PM:
On Wednesday, 9 August 2017 22:02:04 UTC+1, Michael Terrell wrote:
tabbypurr wrote:
On Wednesday, 9 August 2017 19:50:17 UTC+1, Michael Terrell wrote:

Ceramic capacitors are mostly cheap crap. Some have a wide change in
capacitance, at different voltages. Some are microphonic, and their
normal failure mode is a dead short. About the only place they should be
used are for DC bybass, or in low power tuned circuits.

Hmmm. I've not found them crap. Excluding MLCCs, which are a different animal to the rest, ceramics are immune to the melting problems of some plastic film caps, their light weight aids robustness, they're small, they typically handle way above rated voltage without trouble if ever necessary. Slight microphony is not a problem in lots of applications - it is only very slight. The C/V gotcha only applies to some, and obviously one need be aware of it at design time and choose accordingly.

All parts have their issues, and ceramics behave fine and are very reliable in a lot of situations. As with everything, not in every situation.


If the originals were paper, I always use film capacitors. I've had
too many shorted Ceramics in HV circuits, and I've replaced too many
shorted disk capacitors in 5V circuits. Some have burnt holes in PC
boards, when they are on a high current rail. "The C/V gotcha" can cause
distortion in audio circuits, or cause an AGC circuit to 'pump'.

Those last 2 are examples of using them where not suited. If I use ceramics as audio decoupling they need to either not have the C/V gotcha or be big enough to still have enough capacity down at 20Hz to not get in the way AND the dc component across them needs to remain steady.

The C/V "gotcha" is best mitigated by using a part that is overrated for
voltage. The C/V effect is relative to the rated voltage so if that is
several times the circuit voltage there won't be much effect. It's when you
try to operate near the rated voltage that it causes problems.

Picking a random Y5V, C is down to just 20% at 40% rated V, 40% C at 20% rated V.
John Larkin tested a 4.7uF 50v cap and found 0.8uF at 30v, 1.33uF at 20v, 2..86uF at 10v. It was down over 90% at rated V, 0.42uF.

I've also
seen data that indicates the larger package parts have lower coefficients of
capacitance change with voltage.

You can also use C0G types which have a better dielectric which has a very
low coefficient of capacitance change with voltage.

Of course. But one doesn't need C0G for run of the mill decoupling, bypassing etc.


NT

 

[rickman]

tabbypurr@gmail.com wrote on 8/9/2017 9:37 PM:

On Thursday, 10 August 2017 02:00:07 UTC+1, rickman wrote:
tabbypurr wrote on 8/9/2017 8:15 PM:
On Wednesday, 9 August 2017 22:02:04 UTC+1, Michael Terrell wrote:
tabbypurr wrote:
On Wednesday, 9 August 2017 19:50:17 UTC+1, Michael Terrell wrote:

Ceramic capacitors are mostly cheap crap. Some have a wide change in
capacitance, at different voltages. Some are microphonic, and their
normal failure mode is a dead short. About the only place they should be
used are for DC bybass, or in low power tuned circuits.

Hmmm. I've not found them crap. Excluding MLCCs, which are a different animal to the rest, ceramics are immune to the melting problems of some plastic film caps, their light weight aids robustness, they're small, they typically handle way above rated voltage without trouble if ever necessary. Slight microphony is not a problem in lots of applications - it is only very slight. The C/V gotcha only applies to some, and obviously one need be aware of it at design time and choose accordingly.

All parts have their issues, and ceramics behave fine and are very reliable in a lot of situations. As with everything, not in every situation.


If the originals were paper, I always use film capacitors. I've had
too many shorted Ceramics in HV circuits, and I've replaced too many
shorted disk capacitors in 5V circuits. Some have burnt holes in PC
boards, when they are on a high current rail. "The C/V gotcha" can cause
distortion in audio circuits, or cause an AGC circuit to 'pump'.

Those last 2 are examples of using them where not suited. If I use ceramics as audio decoupling they need to either not have the C/V gotcha or be big enough to still have enough capacity down at 20Hz to not get in the way AND the dc component across them needs to remain steady.

The C/V "gotcha" is best mitigated by using a part that is overrated for
voltage. The C/V effect is relative to the rated voltage so if that is
several times the circuit voltage there won't be much effect. It's when you
try to operate near the rated voltage that it causes problems.

Picking a random Y5V, C is down to just 20% at 40% rated V, 40% C at 20% rated V.
John Larkin tested a 4.7uF 50v cap and found 0.8uF at 30v, 1.33uF at 20v, 2.86uF at 10v. It was down over 90% at rated V, 0.42uF.

Yeah, if you are going to pick the worse possible type of cap it won't work
as well. Try an X5R type and it works a lot better.

I've also
seen data that indicates the larger package parts have lower coefficients of
capacitance change with voltage.

You can also use C0G types which have a better dielectric which has a very
low coefficient of capacitance change with voltage.

Of course. But one doesn't need C0G for run of the mill decoupling, bypassing etc.

????

--

Rick C

 

[Jim Mueller]

On Wed, 09 Aug 2017 04:54:25 -0700, Big Bad Bob wrote:

On 03/09/17 17:01, oldschool@tubes.com wrote:
What type of capacitor would be the best match to replace the old
paper/wax caps in old tube gear?

I am not referring to the electrolytics, those I know need to be
electrolytic caps. I am referring to the inter stage caps, such as .01
.05 .1, .005 and so on.....

The object is to replace all the caps in an old tube radio, or any
other tube stuff.

well, if you must replace them with more modern components, a decent
ceramic capacitor should do fine. From what I have seen, ceramics were
common in gear made in the 50's and 60's because of their small relative
size and reasonable consistency in manufacturing.

You should be able to order them from companies like Digikey, Mouser,
etc.. Just make sure the voltage ratings are high enough, and you
should be ok.

I've seen on-line "photo piles" of before/after restoration efforts in
which modern components were "hidden" inside of hollowed-out components.
Maybe you can stash a modern ceramic inside of the paper/wax tube, and
seal it up with a bit of hot glue...

It will be tough to find good ceramics in the values listed with voltage
ratings high enough for tube gear. Polyester (Mylar) or Polypropylene
are better choices.

As for the quality of ceramic capacitors, it depends on what kind of
ceramic is used. C0G ceramics are about the best capacitors you can
easily get. They are very stable, have very low loss, and are excellent
in just about every other way. The one way they are not excellent is
that, except for low values (less than around 1000pF - depends on
manufacturer) they are large and expensive. Mid grade ceramics like X7R
are less stable and less expensive. They are suitable for most audio
frequency work. The really bad ceramics are the ones like Z5U. They are
truly horrible in most ways and are suitable only for non-critical
circuits like supply bypassing. They are, however, tiny and very cheap.
Go to the manufacturer's data sheet for the capacitors you want to use to
see if their characteristics match the application.



--
Jim Mueller wrongname@nospam.com

To get my real email address, replace wrongname with eggmen.
Then replace nospam with expressmail. Lastly, replace com with dk.

 

[Big Bad Bob]

On 08/09/17 19:47, Jim Mueller wrote:

Go to the manufacturer's data sheet for the capacitors you want to use to
see if their characteristics match the application.

ack on that. it's always good advice to see the data sheet.

Somewhat recently, I bought some 500V 0.047uF ceramics that I use for
high freq bypass in a 400V power supply. I checked, it's X7R. no
problems noted. (got them from digikey along with others)

but yeah, the part selectors don't give you a whole lot of choice for
capacitors at voltages about 100V, sometimes. Still, I've never had a
problem with what I've purchased that way.


--
your story is so touching, but it sounds just like a lie
"Straighten up and fly right"

 

[Big Bad Bob]

On 03/10/17 02:04, oldschool@tubes.com wrote:
> Not much of anything made in China is a quality item

actually, I'd think that you just have to be careful and pick a
reputable manufacturer.


--
your story is so touching, but it sounds just like a lie
"Straighten up and fly right"

 

[Big Bad Bob]

On 03/10/17 05:34, Foxs Mercantile wrote:
> There's that magic word "feelings" again. Based on nothing.

yeah, 'Feel' the 'F' word. so many bad decisions are made in the name
of 'Feel'.

 

[Big Bad Bob]

On 08/09/17 19:47, Jim Mueller wrote:

It will be tough to find good ceramics in the values listed with voltage
ratings high enough for tube gear. Polyester (Mylar) or Polypropylene
are better choices.

that might be in a lot of cases. I haven't had that much trouble, but
price DOES go up rather fast as the rated voltage increases. I'll
consider mylar/poly capacitors also, test, see what happens.

 

[rickman]

Big Bad Bob wrote on 8/15/2017 11:21 PM:

On 03/10/17 05:34, Foxs Mercantile wrote:
There's that magic word "feelings" again. Based on nothing.

yeah, 'Feel' the 'F' word. so many bad decisions are made in the name of
'Feel'.

Once you learn about human reasoning you will find it *always* comes down to
feelings. You may have conditioned yourself to weight your feelings on
"logic", but it still comes down to feelings and we have ways of working
around the logic when we want to.

As you say, too often we take a bypass based on some limited experience or
even anecdote we've heard that gives us a "feeling" with little rational
process behind it. See a couple of bad capacitors from any given company or
category of supplier and a strong bias develops which works to bypass that
source.

--

Rick C

 

On Wednesday, 16 August 2017 04:20:34 UTC+1, Big Bad Bob wrote:

On 03/10/17 02:04, oldschool@tubes.com wrote:
Not much of anything made in China is a quality item

actually, I'd think that you just have to be careful and pick a
reputable manufacturer.

A useful percentage of it (unknown China brands) is ok quality. A fair bit is pretty abysmal. Electronics seems to be worse than mechanical tools.


NT

 

[pfjw@aol.com]

On Wednesday, August 16, 2017 at 7:44:33 PM UTC-4, tabb...@gmail.com wrote:
Electronics seems to be worse than mechanical tools.


Not possible. Chinese tools are killers.

Peter Wieck
Melrose Park, PA

 

On Thursday, 17 August 2017 12:56:15 UTC+1, pf...@aol.com wrote:

On Wednesday, August 16, 2017 at 7:44:33 PM UTC-4, tabby wrote:

Electronics seems to be worse than mechanical tools.


Not possible. Chinese tools are killers.

I had a close call with one, but a lot of them are fine. Not really so with electronics.


NT

 

[Big Bad Bob]

On 08/15/17 22:53, rickman wrote:

Big Bad Bob wrote on 8/15/2017 11:21 PM:
On 03/10/17 05:34, Foxs Mercantile wrote:
There's that magic word "feelings" again. Based on nothing.

yeah, 'Feel' the 'F' word. so many bad decisions are made in the name of
'Feel'.

Once you learn about human reasoning you will find it *always* comes
down to feelings.

Uh, NO.


--
your story is so touching, but it sounds just like a lie
"Straighten up and fly right"

 

[rickman]

Big Bad Bob wrote on 8/24/2017 11:51 AM:

On 08/15/17 22:53, rickman wrote:
Big Bad Bob wrote on 8/15/2017 11:21 PM:
On 03/10/17 05:34, Foxs Mercantile wrote:
There's that magic word "feelings" again. Based on nothing.

yeah, 'Feel' the 'F' word. so many bad decisions are made in the name of
'Feel'.

Once you learn about human reasoning you will find it *always* comes down
to feelings.

Uh, NO.

Uh, YES!

--

Rick C

Viewed the eclipse at Wintercrest Farms,
on the centerline of totality since 1998