Caps in amps

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Arny Krueger"
> "Phil Allison"


** Arny is trying to shift the context again - so I had to put it back.


> > If that 100uF cap developed 100 ohms of ESR then the channel's
> > gain drops by 3 dB.

>> Wrong.

** Nope - Arny is *dead wrong* - yet AGAIN !!!!!

> >
> >> As long as there's a few uF capacitance in the capacitor, there's a
> >> turnover point in the frequency response curve. Basic network
> >> analysis. It ain't much of a turnover point, but there will be a
> >> measurable FR change.
>
> >
> > ** ESR is the issue - the "S" in ESR stands for "series ". Any
> > ESR associated with that 100uf cap *adds* to the value of 270 ohm
> > resistor in the feedback network and reduces the amp's gain.
>
>
> Exactly.


** So Arny admits he was wrong.


> So work out the frequency response of the origional network, and
> the new network.


** I already did - it is still flat.

The extra 100 ohms only reduces the gain.



> You'll find that the small capacitance of the partially-failed shunt
> capacitor and it higher ESR can create a frequency-sensitive network.
>


** That is a really desperate attempt to move the context.

There need not be any loss of capacitance with a rise in the ESR.

What a despicable fake you are Arny.




.............. Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

In article <znr1093104445k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
>In article <cg7nfv$i4v$1@panix1.panix.com> kludge@panix.com writes:
>> The
>> nice thing about ESR testing is that it can easily be done in-circuit on
>> the fly, which makes it a great thing for repair technicians.
>
>But you have to know what it's supposed to be. It's not marked on the
>body of the capacitor like the capacitance and (often) the tolerance.
>As far as leakage, "not enough to affect the circuit that's supposed
>to not be getting any DC" is usually good enough. Generally if I can't
>easily lift one end of a capacitor that behaves suspiciously in its
>natural habitat, I won't be trying to fix it.

No, but you can use a cheat sheet to get a good notion of what it ought
to be, and usually when it is off, it is so far off that it's easy to
see that it's bad. ESR is honestly the most convenient way for finding
bad capacitors in equipment, and I have saved so much time since I bought
an ESR tester that I don't know what I did before I had it.

>> By running a short pulse through the capacitor and measuring the
>> instantaneous series resistance, you don't have to take it out of
>> circuit because the accompanying shunt resistance of the circuit is
>> going to be much lower than the often sub-ohm resistance that you
>> are measuring. This makes it very convenient to check capacitors
>> without disconnecting them.
>
>But how do you know that they're bad? Is "normal" ESR a pretty
>standard value for a known type of capacitor so the meter can have a "
>"good/?/bad" scale that can be relied on? I've never used an ESR meter
>and I'm not sure I've ever even seen one.

If you know the voltage and you know the capacitance, you know more or
less how much ESR a typical capacitor is going to have using the cheat
sheet. Higher grade switching supply types will have less, but if they
get to the point where they have more, they are bad. Usually the ESR
rise is high enough that it's easy to tell.

>> Right. But ESR has become a de-facto bench measurement for capacitors
>> because it's easy to make.
>
>And I suppose that's a useful measurement if you know the expected
>value (which I don't).

You want a Heatkit capacitor tester? I got a fancy digital one now and
I hardly ever use it these days, but it comes with the cheat sheet.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Arny Krueger" <
> "Mike Rivers"
>> Phil Allison
> >
> >> ** I gave the definition that all electro cap makers use -
> >> dickhead.
> >
> > I've never seen a capacitor maker define the end of a capacitor's
> > life. Why would they?
>
> Because the life of electrolytic capacitors is a big issue. You need some
> kind of a reliable measure of life to talk about it in a reasonable way.
>
>
> ESR increase accelerates the failure of the cap. The ESR goes up because
of
> heating due energy being dissipated in the cap. The higher ESR causes more
> energy to be dissipated in the cap, further raising the temperature of the
> cap.
>


** Increased temperature actually **reduces** the ESR of an electro - at
80 degrees C it is about 5 times less than at 20 degrees C. This is due
simply the electrolyte becoming more conductive when hot.

The cause of ESR rising ( in a given cap at a given temp) is loss of
electrolyte via imperfect seals - a process that accelerates at higher
temps.




............ Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Phil Allison" <philallison@tpg.com.au> wrote in message
news:2oon9aFb1scrU2@uni-berlin.de
> "Arny Krueger" <
>> "Mike Rivers"
>>> Phil Allison
>>>
>>>> ** I gave the definition that all electro cap makers use -
>>>> dickhead.
>>>
>>> I've never seen a capacitor maker define the end of a capacitor's
>>> life. Why would they?
>>
>> Because the life of electrolytic capacitors is a big issue. You need
>> some kind of a reliable measure of life to talk about it in a
>> reasonable way.
>>
>>
>> ESR increase accelerates the failure of the cap. The ESR goes up
>> because of heating due energy being dissipated in the cap. The
>> higher ESR causes more energy to be dissipated in the cap, further
>> raising the temperature of the cap.
>>
>
>
> ** Increased temperature actually **reduces** the ESR of an electro
> - at 80 degrees C it is about 5 times less than at 20 degrees C.
> This is due simply the electrolyte becoming more conductive when hot.


Like all generalities, this is falsified by actual vendor data:

http/www.powerdesigners.com/InfoWeb/design_center/Design_Tips/Electrolytics/Caps.shtm

shows a plot of ESR versus temperature and frequency.

First off, we see that ESR is sometimes an increasing function of
temperature and sometimes a decreasing function of temperature.

ESR is generally sharply decreasing at sub-freezing temperatures.

Examining the plot more closely, we see that at low frequencies below 100
KHz (audio & power frequencies) and above about 30C, ESR is an increasing
function of temperature. IOW under typical audio equipment temperature and
frequencies, ESR increases with temperature.

Perhaps other capacitors perform differently.
http/www.cornell-dubilier.com/applets/CDEspiceApplet/CDEspice.html
suggests they do.


> The cause of ESR rising ( in a given cap at a given temp) is loss of
> electrolyte via imperfect seals - a process that accelerates at higher
> temps.

But it seems like Phil previously said that capacitance does not decrease
significantly over the life of a cap. This begs the question of how a
capacitor has constant capacitance with decreasing electrolyte.

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

In article <2oon9aFb1scrU2@uni-berlin.de> philallison@tpg.com.au writes:

> The cause of ESR rising ( in a given cap at a given temp) is loss of
> electrolyte via imperfect seals - a process that accelerates at higher
> temps.

Does loss of electrolyte cause reduction in capacitance? Can we get
back to the reason why we care about ESR instead of that the marketing
department likes good numbers? Most of us don't go around measuring
capacitors to see if they're on the road to failure (we already know they
are, from the moment we put them into service) but we can tell when
they have failed.




--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Mike Rivers"
>

>
> > What is the audible effect of aging capacitors? Should I start to hear
low
> > level 60 Hz hum?
>
> If it's a power supply filter that isn't filtering like it should,
> yes. Actually, you'll most likely hear 120 Hz hum because just about
> every power supply uses a full wave rectifier.
>


** Not so for voltage doubler PSUs - they are full wave but each cap
sees 60 Hz ripple.




........... Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

In article <2ooncvFd86m4U1@uni-berlin.de> philallison@tpg.com.au writes:

> ** Not so for voltage doubler PSUs - they are full wave but each cap
> sees 60 Hz ripple.

Making an around-the-corner point just for the sake of argument again.
But this is typical for the loser-who-refuses-to-lose.



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Kurt Albershardt <kurt@nv.net> wrote:
>Scott Dorsey wrote:
>>
>> Both the Heathkit and the little kit ESR meter that Dick Smith sells
>
>Searching around a bit, I also found the B&K <http/bkprecision.com/www/np_searchmodel7.asp?lf=ESR+Meters>
>
>Too bad this isn't a standard DVM function yet.

The capacitor tester in my DVM actually is a relaxation oscillator with an
RC time constant, the C being the device under test, going into a frequency
counter. If the capacitor is leaky or has excessively high ESR, the value
will test as incorrect. It's actually very handy but only out of circuit.
Unfortunately it also reads in nF, which I can't think in easily.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Mike Rivers wrote:

> In article <2onkciFci068U1@uni-berlin.de> philallison@tpg.com.au writes:
>
> > ** I gave the definition that all electro cap makers use - dickhead.
>
> I've never seen a capacitor maker define the end of a capacitor's
> life. Why would they?

Because parameters change with 'load life'.

I've never seen a data sheet refer to life vs ESR though.

Here's a typical one.

http/www.sterling-comp.co.uk/PDFs/Nippon_Chemi_Con/SMH.pdf

'Endurance' is specified thus :

The following specifications shall be satisfied when the capacitors are
restored to 20C after subjected to DC voltage with the rated ripple current is
applied for 2000 hours at 85C.
Capacitance change <= +/- 20% of the initial value
D.F. (tan delta) <= +/- 200% of the initial specified value
Leakage current <= The initial specified value


Graham

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Scott Dorsey wrote:

> Kurt Albershardt <kurt@nv.net> wrote:
> >Scott Dorsey wrote:
> >>
> >> Both the Heathkit and the little kit ESR meter that Dick Smith sells
> >
> >Searching around a bit, I also found the B&K <http/bkprecision.com/www/np_searchmodel7.asp?lf=ESR+Meters>
> >
> >Too bad this isn't a standard DVM function yet.
>
> The capacitor tester in my DVM actually is a relaxation oscillator with an
> RC time constant, the C being the device under test, going into a frequency
> counter. If the capacitor is leaky or has excessively high ESR, the value
> will test as incorrect. It's actually very handy but only out of circuit.
> Unfortunately it also reads in nF, which I can't think in easily.

I tend to find the classic 0.001 uF a little tiresome myself. Don't appreciate any zeroes after the decimal
point myself.

Graham


p.s. was reading an old text book recently and had to stop to think about values in uuF !

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Pooh Bear

>
> I've never seen a data sheet refer to life vs ESR though.
>
> Here's a typical one.
>
> http/www.sterling-comp.co.uk/PDFs/Nippon_Chemi_Con/SMH.pdf
>
> 'Endurance' is specified thus :
>
> The following specifications shall be satisfied when the capacitors are
> restored to 20C after subjected to DC voltage with the rated ripple
current is
> applied for 2000 hours at 85C.
> Capacitance change <= +/- 20% of the initial value
> D.F. (tan delta) <= +/- 200% of the initial specified value
> Leakage current <= The initial specified value
>


** You have ** MISQUOTED** the DF ( tan delta) figures - there is no
+/- involved !!

DF ( dissipation factor) is proportional to the ESR.

What that spec says is that when the DF or ESR doubles the game is over.

What you claimed you had never seen was right in front of you !!!




............ Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Phil Allison wrote:

> "Pooh Bear
>
> >
> > I've never seen a data sheet refer to life vs ESR though.
> >
> > Here's a typical one.
> >
> > http/www.sterling-comp.co.uk/PDFs/Nippon_Chemi_Con/SMH.pdf
> >
> > 'Endurance' is specified thus :
> >
> > The following specifications shall be satisfied when the capacitors are
> > restored to 20C after subjected to DC voltage with the rated ripple
> current is
> > applied for 2000 hours at 85C.
> > Capacitance change <= +/- 20% of the initial value
> > D.F. (tan delta) <= +/- 200% of the initial specified value
> > Leakage current <= The initial specified value
> >
>
> ** You have ** MISQUOTED** the DF ( tan delta) figures - there is no
> +/- involved !!

Sorry mate, copying and pasting resulted in some odd characters that I had to
re-insert and I got carried away - lol..

Clearly you looked at the link I posted though.


> DF ( dissipation factor) is proportional to the ESR.

Ok, the spec goes D.F. (tan delta) <= 200% of the initial specified value

> What that spec says is that when the DF or ESR doubles the game is over.
>
> What you claimed you had never seen was right in front of you !!!

But the manufacturer doesn't spec it as ESR however. As you like quoting such
detail, I felt obliged to inform you.

Graham

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey" <
> Arny Krueger
> >
> >Since ESR is so widely used as *the* criteria for end-of-life I thought
> >that somewhere I'd find a chart of capacitance versus ESR, but no dice.
On
> >balance, capacitance seems like a poor parameter to track, because
initial
> >capacitance can vary so much.
>
> Both the Heathkit and the little kit ESR meter that Dick Smith sells come
> with charts that list minimal acceptable ESR for a given capacitance and
> voltage value. But it's really just a ballpark value, since capacitors
> all started out with very different ESR when they were new (compare a
1950's
> Mallory with a modern Panasonic FC, for instance). Can you really use
> the same standard on all of them?


** No - but you simply test another cap of the same ratings and
size/type to get a reference figure. Usually you can find such a cap in the
same piece of gear or just pick one out of your own stocks.

By the time a service person discovers a bad cap its ESR has gone way high
and there is no doubt whatsoever.


>
> >It's possible that virtually every cap that has loss significant amounts
of
> >capacitance, also has ESR problems.


** Guaranteed in fact.


............ Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

In article <2oqd84Fd8e3mU1@uni-berlin.de> philallison@tpg.com.au writes:

> By the time a service person discovers a bad cap its ESR has gone way high
> and there is no doubt whatsoever.

Hey! That's what I've wanted to say all along. Thanks for figuring it
out.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey"

> Both the Heathkit and the little kit ESR meter that Dick Smith sells come
> with charts that list minimal acceptable ESR for a given capacitance and
> voltage value.


** Do you have the Dick Smith ESR meter ??

I think nearly every tech in Australia has one by now. The designer ( Bob
Parker) lives not far from me and we are in regular contact.

Did you know that ESR meter can be used to test cells and batteries for
internal resistance ( as a guide to remaining capacity ) as well ??

Ideal for checking all those button cells and memory back up lithium's since
it places no load on the DUT !




........... Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Phil Allison <philallison@tpg.com.au> wrote:
>"Scott Dorsey"
>
>> Both the Heathkit and the little kit ESR meter that Dick Smith sells come
>> with charts that list minimal acceptable ESR for a given capacitance and
>> voltage value.
>
> ** Do you have the Dick Smith ESR meter ??

Yes, it has radically reduced my time to find bad tantalums on memory boards,
and it was worth the price a hundred times over just for that.

>I think nearly every tech in Australia has one by now. The designer ( Bob
>Parker) lives not far from me and we are in regular contact.
>
>Did you know that ESR meter can be used to test cells and batteries for
>internal resistance ( as a guide to remaining capacity ) as well ??
>
>Ideal for checking all those button cells and memory back up lithium's since
>it places no load on the DUT !

No load at all?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey"
> Phil Allison
>
> No load at all?



** The tiny AC test signal does not load the cell - ie it takes no current
from it.


BTW Can you verify ( for Arny's benefit) that heating an electro reduces
the ESR reading ?

Using a hot air gun on a cap ( say 47 uF, 350 volt) while connected to the
meter.




........... Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Phil Allison <philallison@tpg.com.au> wrote:
>
> BTW Can you verify ( for Arny's benefit) that heating an electro reduces
>the ESR reading ?
>
> Using a hot air gun on a cap ( say 47 uF, 350 volt) while connected to the
>meter.

Sure it does. But it'll also reduce the measured capacitance value with
a DMM too.

That's why Philips makes high temperature caps rated for 125'C use.
Better Tc, not just longer life at high temperature.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

"Pooh Bear"
> Phil Allison wrote:
> > >
> > > I've never seen a data sheet refer to life vs ESR though.
> > >
> > > Here's a typical one.
> > >
> > > http/www.sterling-comp.co.uk/PDFs/Nippon_Chemi_Con/SMH.pdf
> > >
> > > 'Endurance' is specified thus :
> > >
> > > The following specifications shall be satisfied when the capacitors
are
> > > restored to 20C after subjected to DC voltage with the rated ripple
> > current is
> > > applied for 2000 hours at 85C.
> > > Capacitance change <= +/- 20% of the initial value
> > > D.F. (tan delta) <= +/- 200% of the initial specified value
> > > Leakage current <= The initial specified value
> > >
> >
> > ** You have ** MISQUOTED** the DF ( tan delta) figures - there is
no
> > +/- involved !!
>
>
> > DF ( dissipation factor) is proportional to the ESR.
>
> Ok, the spec goes D.F. (tan delta) <= 200% of the initial specified value
>
> > What that spec says is that when the DF or ESR doubles the game is
over.
> >
> > What you claimed you had never seen was right in front of you !!!
<
>
> But the manufacturer doesn't spec it as ESR however. As you like quoting
such
> detail, I felt obliged to inform you.
>


** So I had the intelligence to realise they are the same and post correct
info while you did NOT.

You squirming and lies are as bad as ever.




......... Phil

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

Scott Dorsey wrote:

> Kurt Albershardt <kurt@nv.net> wrote:
>
>> Scott Dorsey wrote:
>>
>>> Both the Heathkit and the little kit ESR meter that Dick Smith sells
>>
>> Searching around a bit, I also found the B&K <http/bkprecision.com/www/np_searchmodel7.asp?lf=ESR+Meters>
>>
>> Too bad this isn't a standard DVM function yet.
>
>
> The capacitor tester in my DVM actually is a relaxation oscillator with an
> RC time constant, the C being the device under test, going into a frequency
> counter. If the capacitor is leaky or has excessively high ESR, the value
> will test as incorrect. It's actually very handy but only out of circuit.

Ditto for those in several DVMs here.