Searching \ for '[EE] Electrolytics and Voltage' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=voltage
Search entire site for: 'Electrolytics and Voltage'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Electrolytics and Voltage'
2007\03\16@172615 by Jinx

face picon face

> From what I've learned over the years although I can't put my finger
> on the source,  an electrolytic should be used close to its rated voltage
>
> Then there's also the stuff about forming the cap if it's been sitting
> for quite some time.
>
> Does anyone have anymore information on this?

If you search for tantalum in the PIClist mail archives there are all sorts
of electrolytic-related threads. I'm pretty sure Russell mentioned what
you said above, but I don't remember the actual thread

2007\03\16@192450 by John Dammeyer

flavicon
face
>
>
> > From what I've learned over the years although I can't put my finger
> > on the source,  an electrolytic should be used close to its
> rated voltage
> >
> > Then there's also the stuff about forming the cap if it's
> been sitting
> > for quite some time.
> >
> > Does anyone have anymore information on this?
>
> If you search for tantalum in the PIClist mail archives there
> are all sorts
> of electrolytic-related threads. I'm pretty sure Russell
> mentioned what
> you said above, but I don't remember the actual thread
>

Searching the archives turned out to be a waste of time as most of the
information presented wasn't backed up with any evidence.  A google
search produced this document:

http://www.nichicon-us.com/english/lib/aluminum.pdf

Section 2-9-4 Applied Voltage and Life: Provided text with a curve
demonstrating that applied voltage has very little effect on the life of
an electrolytic compared to temperature and ripple current.

So for me, the myth is busted.

John Dammeyer


2007\03\16@194129 by David VanHorn

picon face
>
>
> http://www.nichicon-us.com/english/lib/aluminum.pdf
>
> Section 2-9-4 Applied Voltage and Life: Provided text with a curve
> demonstrating that applied voltage has very little effect on the life of
> an electrolytic compared to temperature and ripple current.
>
> So for me, the myth is busted.


For that sort of aluminum electrolytic at least.

2007\03\16@194831 by John Dammeyer

flavicon
face
> >
> >
> > www.nichicon-us.com/english/lib/aluminum.pdf
> >
> > Section 2-9-4 Applied Voltage and Life: Provided text with a curve
> > demonstrating that applied voltage has very little effect
> on the life of
> > an electrolytic compared to temperature and ripple current.
> >
> > So for me, the myth is busted.
>
>
> For that sort of aluminum electrolytic at least.
>
> David VanHorn [spam_OUTdvanhornTakeThisOuTspammicrobrix.com]
> --


See! For me that's indicitive of the type of postings I found in the
PICLIST archives that take so much time to sort through.  Doesn't point
to the types of capacitors where it does matter.  Sarcastically implies
it does matter but provides absolutely no additional information.

John Dammeyer


2007\03\16@195921 by William Chops Westfield

face picon face

> an electrolytic should be used close to its rated voltage
>
I think this is more of a "don't use 400V caps in a 3V circuit",
rather than "close."  After all, electrolytic voltages below 16V
are a relatively recent occurrence, and low voltage circuits aren't.

BillW

2007\03\16@200446 by David VanHorn
picon face
>
>
> See! For me that's indicitive of the type of postings I found in the
> PICLIST archives that take so much time to sort through.  Doesn't point
> to the types of capacitors where it does matter.  Sarcastically implies
> it does matter but provides absolutely no additional information.


My point is that absence of evidence is not evidence of absence.
You can't extrapoltate "it definitely does not matter for this particular
type of cap" to "it does not matter for all caps".

I wasn't being sarcastic at all.

2007\03\16@200536 by Jinx

face picon face
John, have a look around the major players' sites, like Elna

http://www.elna-america.com/tech_info.php

As you said, ripple and temperature are parameters to watch for
if the capacitor is within the rated voltage

They describe wet aluminium electrolytics on that page, there are
pdfs too

http://www.elna-america.com/products/pdf_files/AL/al_caution.pdf

http://www.elna-america.com/products/pdf_files/AL/al_technotes.pdf

2007\03\16@201447 by James Newtons Massmind

face picon face
> See! For me that's indicitive of the type of postings I found in the
> PICLIST archives that take so much time to sort through.  
> Doesn't point
> to the types of capacitors where it does matter.  
> Sarcastically implies
> it does matter but provides absolutely no additional information.

Dude, the datasheet is the only thing that matters. You can't make
generalizations about any sort of component with impunity, you must always
check the datasheet from the mfgr of _that_ component.

So the answer to the original question depends on the exact make of
electrolytic capacitor, and on what the datasheet says for it.

"The datasheet: its not just for PICs any more!"

Yes, it is more /likely/ that one passive will work just like the next
passive, but among passives, capacitors are more active than most. And
electrolytic are the most active class of capacitors.

So the answer is in the datasheet, not the archive.

And now this answer is in the archive.

---
James.


2007\03\16@202203 by Jinx

face picon face

> I think this is more of a "don't use 400V caps in a 3V circuit",
> rather than "close."

Hi Bill, what John is saying, I think, is that nowhere in the literature
does it say not to run a cap grossly under its rated voltage. A common
question you see in mailbags is something like "You specify a 16V
capacitor for C4 but I only have a 50V. Can I use that ?" And the
answer will be "Yes, if space permits". But I don't ever remember a
caveat with that. Whether that advice is completely true or half-true, I
don't know

My, admittedly uninformed, guess would be that any voltage with
the correct polarity will keep the capacitor polarised. I wonder
though what might happen if say you have a 400V capacitor in a
3V circuit, then have the power unconnected for some time. Would
the capacitor be more likely to de-polarise or even reverse if it
had been used just with 3V, rather than say 350V ? 3V is obviously
closer to 0V but it's still a forming polarity

2007\03\16@202530 by John Dammeyer

flavicon
face
>
>
> John, have a look around the major players' sites, like Elna
>
> http://www.elna-america.com/tech_info.php
>
> As you said, ripple and temperature are parameters to watch for
> if the capacitor is within the rated voltage
>
> They describe wet aluminium electrolytics on that page, there are
> pdfs too
>

This one here is really interesting.  It also states that low voltage
isn't much of an issue but what's really interesting is it's almost a
word for word reprint of a document from the Nichicon datasheet.

> http://www.elna-america.com/products/pdf_files/AL/al_caution.pdf

Page 101 of this next document says the same thing as my reference.  

"The effects to the life by derating of the applied voltage
etc. are neglected because they are small compared
to that by the temperature."

Essentially they ignore it.  They certainly don't say watch out for 400V
caps used at 3V.  

>
> http://www.elna-america.com/products/pdf_files/AL/al_technotes.pdf
>

There's a lot of hearsay in the archives with no evidence to back it up.
Only theological arguments that just because you can't see it doesn't
mean it doesn't exist.  

John Dammeyer



2007\03\16@203448 by David VanHorn

picon face
>
>
> My, admittedly uninformed, guess would be that any voltage with
> the correct polarity will keep the capacitor polarised. I wonder
> though what might happen if say you have a 400V capacitor in a
> 3V circuit, then have the power unconnected for some time. Would
> the capacitor be more likely to de-polarise or even reverse if it
> had been used just with 3V, rather than say 350V ? 3V is obviously
> closer to 0V but it's still a forming polarity


Capacitance can change significantly with applied voltage, kemet has some
reference to that in their docs, but I don't remember which way it went, or
which sort of caps.

I remember tantalum caps being affected most by this, with audio circuits
taking special pains to place some voltage across them, the reasoning
expressed to me was that they would greatly reduce in effective capacitance
over time.  This may be something that was true then, and not now.

2007\03\16@203507 by John Dammeyer

flavicon
face
Hi James,

> Dude, the datasheet is the only thing that matters. You can't make
> generalizations about any sort of component with impunity,
> you must always
> check the datasheet from the mfgr of _that_ component.
>
> So the answer to the original question depends on the exact make of
> electrolytic capacitor, and on what the datasheet says for it.

> Yes, it is more /likely/ that one passive will work just like the next
> passive, but among passives, capacitors are more active than most. And
> electrolytic are the most active class of capacitors.
>
> So the answer is in the datasheet, not the archive.

OK.  So point me to a data sheet where it does matter.  

John Dammeyer


>
> ---
> James.
> James Newtons Massmind [.....jamesnewtonKILLspamspam@spam@massmind.org]>  
>
> --

2007\03\16@205438 by David VanHorn

picon face
A bit of googling turns up this article:

www.electronicproducts.com/ShowPage.asp?Filename=itw.may2006.html

2007\03\16@210943 by Jinx

face picon face
> http://www.electronicproducts.com/ShowPage.asp?Filename=itw.may2006.html

"if operated below their rated voltage, the capacitors will reform to the
voltage at which they are being used, and any voltage surge can result
in circuit failure"

That would be something to watch out for in any application where
there's a chance Vcc might be increased in the future

It would imply then that, for cost reasons too maybe, you'd choose
the cap with the lowest voltage rating above present Vcc, ie no point
using a 50V in a 5V circuit because it'll reform to a 5V cap anyway

> http://www.elna-america.com/products/pdf_files/AL/al_technotes.pdf

. If the two terminals are short-circuited after the
recovery voltage is generated, a spark may scare the
workers working in the assembly line

Nice touch, thoughtful. And don't ffffttt the girls with air-guns

2007\03\16@211123 by David VanHorn

picon face
yarchive.net/electr/electrolytic_caps.html

This suggests that this effect was real, but only with very old parts.



Still, there are things that aren't in the data sheets, or are in only some
versions of the data sheets..

Put a 78XX series regulator in circuit with a couple thousand uF on the
output, let it run, then crowbar the input. You'll likely end up with a dead
78XX, because of current flowing backward through the regulator.  This is
documented, but many versions of the data sheets from many manufacturers do
not mention it.  (solved by a single diode reverse biased from input to
output)


I ran into a problem with some Unitrode current chopping stepper drivers
cratering.
Months of back and forth with the factory, no answers, till I found an
original data sheet from Rifa who originally designed the die (Unitrode and
others licenced their design, Cherry did their own)   Turned out we were
exceeding the max chopping speed, which was not specified on modern versions
of the data sheets!   They showed a "typical" component selection to set
chopping speed, but they didn't say what the max was.

2007\03\16@215046 by John Dammeyer

flavicon
face
>
>
> A bit of googling turns up this article:
> David VanHorn [dvanhornspamKILLspammicrobrix.com]

>
> www.electronicproducts.com/ShowPage.asp?Filename=itw.ma
> y2006.html
> --

The information is anecdotal.

"Unfortunately, if operated below their rated voltage, the capacitors
will reform to the voltage at which they are being used, and any voltage
surge can result in circuit failure."

Hmmm.  "So a cap run at 80% of its rated voltage will reform to that
voltage and a surge will damage it".  Do we then run at the rated
voltage?  We do know that over voltage is very bad and that's direct
from the manufacturers data sheets.

My statement isn't any less true or more true the one quoted.  And
there's the rub.  It sounds right but isn't backed by a data sheet or
manufacturers reference.  At least not directly.  

The electrolytic caps will reform to the lower voltage.  So if you put
it into 100V cap  into a 24V circuit the implication is that the cap
will no longer be able to handle 100V without a reforming application.
The Nichicon reference states a 1K resistor in series will do this.

But really,  think about it,  once the cap is soldered in circuit that
runs on a 24V supply,  that cap will never see 100V.  If it does on a
regular basis, then the cap stays formed to that higher voltage.  Hence,
the reason for the 100V cap is because occasionally it's hit with the
higher voltage.  That's just plain good design.  (although better to try
and remove the peaks).



John Dammeyer



2007\03\16@215855 by Jinx

face picon face
> http://www.electronicproducts.com/ShowPage.asp?Filename=itw.may2006.html

"if operated below their rated voltage, the capacitors will reform to the
voltage at which they are being used, and any voltage surge can result
in circuit failure"

I wonder now about battery-powered circuits. Say you had a low Iq
circuit, 3 AA and components that worked all the way from 5V to under
3V. It might sit there for a considerable time with less than charged cells,
functioning properly. When you eventually put fresh batteries in, the caps
wouldn't kersplode by being over-voltaged by 50% because they'd
reformed at 3V ? Surely not ? You don't hear of remote controls and
the like failing

2007\03\16@220441 by John Dammeyer

flavicon
face
>
> I wonder now about battery-powered circuits. Say you had a low Iq
> circuit, 3 AA and components that worked all the way from 5V to under
> 3V. It might sit there for a considerable time with less than
> charged cells,
> functioning properly. When you eventually put fresh batteries
> in, the caps
> wouldn't kersplode by being over-voltaged by 50% because they'd
> reformed at 3V ? Surely not ? You don't hear of remote controls and
> the like failing
> Jinx [.....joecolquittKILLspamspam.....clear.net.nz]


My point exactly.  If it were a reliability issue there'd be far more in
the various data sheets about it.


John Dammeyer


2007\03\16@222210 by David VanHorn

picon face
>
>
> I wonder now about battery-powered circuits. Say you had a low Iq
> circuit, 3 AA and components that worked all the way from 5V to under
> 3V. It might sit there for a considerable time with less than charged
> cells,
> functioning properly. When you eventually put fresh batteries in, the caps
> wouldn't kersplode by being over-voltaged by 50% because they'd
> reformed at 3V ? Surely not ? You don't hear of remote controls and
> the like failing


The manufacturers do mention higher than normal leakage currents, and I
guess it depends on the meaning of "higher" :)   But from what I've seen
this evening, I'm coming to the conclusion that this may have been a problem
back 50 years ago, but with modern parts it's either small enough to ignore,
or nonexistent.  But I will probably ask around next week and see if the
vendors have any data on it.

Explosion isn't the only sort of failure either.. Being more than spec'd
tolerance off value, high ESR could also be a problem, and then going back
to normal operation under those conditions could cause further failure.   In
my "receiptmaster" printer, I did an SMPS that was designed to charge up
it's output caps to 32V and then dump the charge at 19A into a
printhead. This happens about 500 times a second, more or less, depending on
line voltage. The printer is limited by the current capacity of the
transformer that feeds it, to about 40W input.  I considered a lot of
factors in that design in order to select the caps, and I know that if ESR
went up significantly those caps would have a VERY short life after that
point.   They operate in a manner more like a photostrobe, and they also
have to deal with large ripple current during charge.

2007\03\16@223842 by Jinx

face picon face

> The manufacturers do mention higher than normal leakage currents,

Perhaps there are two scenarios -

1) a 6.3V cap, new, has 9V put on it. Hasta la vista

2) a 6.3V cap has possibly reformed at 3V. Because of its original
voltage rating, it can cope with a quick reform to 4.5V, but may
sustain some damage, damage that may or may not be significant in
the context of the circuit. You may conclude that "batteries don't last
like they used to", when in fact it's a leaky cap, damaged by fresh
batteries. Or the remote doesn't reach like it used to, because the
reservoir cap can't supply transients for the IR LEDs like it could
before

Fascinating

2007\03\17@112752 by Bob Axtell

face picon face
John Dammeyer wrote:
>> A bit of googling turns up this article:
>> David VanHorn [EraseMEdvanhornspam_OUTspamTakeThisOuTmicrobrix.com]
>>    
>
>  
>> www.electronicproducts.com/ShowPage.asp?Filename=itw.ma
>> y2006.html
>> --
>>    
>
> The information is anecdotal.
>
>  
I'm skeptical of this being true now; modern caps can handle spikes
well. I believe this refers to
very early wet aluminum caps.

--Bob


{Quote hidden}

2007\03\17@123029 by David VanHorn

picon face
>
> >
> I'm skeptical of this being true now; modern caps can handle spikes
> well. I believe this refers to very early wet aluminum caps.


This may well be like the "myth" that sitting a car battery on concrete will
discharge it.
According to several battery manufacturers (and common sense) this is not
true of modern batteries.
They say that it was true for very early batteries, due to their
construction, but nobody makes anything like that now.

2007\03\17@130247 by Cristóvão Dalla Costa

picon face
On 3/17/07, David VanHorn <dvanhornspamspam_OUTmicrobrix.com> wrote:
>
> >
> > >
> > I'm skeptical of this being true now; modern caps can handle spikes
> > well. I believe this refers to very early wet aluminum caps.
>
>
> This may well be like the "myth" that sitting a car battery on concrete
> will
> discharge it.
> According to several battery manufacturers (and common sense) this is not
> true of modern batteries.
> They say that it was true for very early batteries, due to their
> construction, but nobody makes anything like that now.


Huh? How is the material the battery is sitting on make any difference?

2007\03\17@132054 by David VanHorn

picon face
>
>
> > This may well be like the "myth" that sitting a car battery on concrete
> > will
> > discharge it.
> > According to several battery manufacturers (and common sense) this is
> not
> > true of modern batteries.
> > They say that it was true for very early batteries, due to their
> > construction, but nobody makes anything like that now.
>
>
> Huh? How is the material the battery is sitting on make any difference?


With a modern battery, it dosen't.
With the old glass jar and tar construction, it did, apparently.
But the old data persists, even though it's not relevant to modern
batteries.

2007\03\17@141327 by Rich

picon face
The battery-concrete story may be a myth.  There may be other factors
involved because I had to dispose of two new batteries that sat on the
concrete floor  for a year.  They would not hold a charge.  Coincidence?

{Original Message removed}

2007\03\17@143354 by David VanHorn

picon face
On 3/17/07, Rich <@spam@rgrazia1KILLspamspamrochester.rr.com> wrote:
>
> The battery-concrete story may be a myth.  There may be other factors
> involved because I had to dispose of two new batteries that sat on the
> concrete floor  for a year.  They would not hold a charge.  Coincidence?


http://www.uuhome.de/william.darden/carfaq14.htm#concrete

http://www.repairfaq.org/ELE/F_Car_Battery.html#CARBATTERY_026

http://www.powerstream.com/Storage.htm

http://www.interstatebatteries.com/www_2001/content/faqs/tech_talk/maintenance/faq_tech_maint.htm

  *Will storing my battery on concrete drain the charge?* No. Regarding
today's batteries, this is a myth. A battery placed on concrete will not
discharge any faster, but a battery will discharge over a period of time
wherever it is placed. If the battery has a surface layer of acid or grime
which is conductive, the battery will self-discharge more rapidly than if it
were clean and dry.

This myth does have some historical basis. Many years ago, wooden battery
cases encased a glass jar with the battery in it. Any moisture on the floor
could cause the wood to swell and possibly fracture the glass, causing it to
leak. Later came the introduction of the "hard rubber" cases, which were
somewhat porous. A current could be conducted through this container, which
had a high carbon content, if the moist concrete floor permitted the current
to find an electrical ground. The wise advise of the old days to "not store
batteries on concrete" has apparently been passed down to us today, but it
no longer applies.

2007\03\17@153651 by Thomas C. Sefranek

face picon face
Letting ANY lead acid battery (sealed or not) self discharge over 2 years
will kill it.  No concrete required.  Can you say Spongy lead SULFATION?

 *
 |  __O    Thomas C. Sefranek  KILLspamWA1RHPKILLspamspamARRL.NET
 |_-\<,_   Amateur Radio Operator: WA1RHP
 (*)/ (*)  Bicycle mobile on 145.41MHz PL74.4

ARRL Instructor, Technical Specialist, VE Contact.
hamradio.cmcorp.com/inventory/Inventory.html
http://www.harvardrepeater.org

{Original Message removed}

2007\03\17@175035 by Dr Skip

picon face
I've not heard of this reforming, but have always heard that they will
not be as true to their rated capacity (not voltage) if run
significantly lower. ie, a 10 uF/50v cap will be closest to 10uF AT 50v.
At 5v, it will deviate. There is some long term effect, which I don't
remember exactly, but involves this in that it will retain its new
'value' even at 50v after a long life at 5v.

If we look at it physically, the insulating properties of the dielectric
are what determine voltage rating, and nothing happens that would create
more conductivity over time, is there?  However, since these have
chemicals that increase their capacitance over that of just
metal-paper-metal units, the limiting case (no electrolytic compound)
would be the capacitance drops to that of a metal-paper-metal capacitor
as teh compounds deactivate.

If we look at another limiting case - 0v for a long time (on the shelf).
One can apply voltage within rating and be fine in real life, however,
it may not have the capacitance it is labelled as (the staling effect of
old caps)... Therefore, this also supports this argument. It is
possible, barring physical damage, to cycle up old caps to full capacity
by charge/discharge cycles in some cases.

-Skip

2007\03\17@200847 by Jinx

face picon face
> If we look at it physically, the insulating properties of the dielectric
> are what determine voltage rating, and nothing happens that would
> create more conductivity over time, is there ?

What the data sheet caution says is that a surge into a de-formed
capacitor may result in higher leakage. That suggests perhaps some
punch-through, although electrolytics can self-heal if the damage is
relatively minor

It could be that "important" electrolytics in a circuit be provided
with surge protection. That's often the case in, for example, PSUs,
but not in lower current applications. Maybe something as simple
as a current-limiting resistor on V+ input. That doesn't address the
time needed to reform though

That's assuming reformation at lower voltages is true

2007\03\19@002710 by PICLIST

flavicon
face
My understanding of the lead-acid battery "myth" (I do not remember where I dug
this up, but I remember the source appearing to be reliable) is that there is in
fact in most cases an increase in self-discharge rate in a lead-acid battery
that is left sitting on a concrete floor.

The issue is not electrical but thermal. The voltage of a lead-acid battery
changes with temperature (higher temp = higher voltage). An unheated concrete
floor (most of them) is usually cooler than ambient in a northern climate. That
causes the cell voltage at the bottom of the cell to be a bit lower than the
voltage at the top of the cell (where the temperature is a bit higher). Since
the top and the bottom of the cell are connected by low impedance lead plates, a
current flows that discharges the battery. E=IR - E is small, but so is R, so I
ends up being significant.

The current will tend to discharge the warmer upper portion of the battery a
bit, and charge the cooler lower portion of the battery a bit. The problem is
that the charge/discharge cycle is not 100% efficient and entropy takes its
inevitable toll.

The recommendation I've seen is to store the battery thermally insulated from a
concrete (or any thermally conductive or thermally massed) floor. I generally
use a 1.5" block of Styrofoam for various batteries I have in storage. I also
make sure each receives a trickle charge (in the range of C/100) every few
months to keep them from sulfating.

=================================================================

This may well be like the "myth" that sitting a car battery on concrete will
discharge it.
According to several battery manufacturers (and common sense) this is not
true of modern batteries.
They say that it was true for very early batteries, due to their
construction, but nobody makes anything like that now.

2007\03\20@062350 by Russell McMahon

face
flavicon
face
> My point exactly.  If it were a reliability issue there'd be far
> more in
> the various data sheets about it.

If that were true then all tantalum capacitor data sheets would come
with a large red skull and cross bone sprinted on them and be locked
in hermetically sealed boxes :-).

ie tantalums are superb capacitors, but take them hardly above rated
voltage with a voltage spike of even trivially low energy content, or
connect then reverse polarity and, if there's substantial energy
available in the circuit (as opposed to eg the spike which triggered
the avalanche)  then you'll usually get some or all of shrieking,
smoke (very noxious), flame or explosion and end with a very very very
hard short circuit across the terminals. I've seen a capacitor that
gave ALL these in sequence. Great fun. For good measure in a metal can
tantalum (rare) you may get a hole in the side where you can view the
now solid bead of tantalum rolling to and fro.

ie data sheets may be somewhat mealy mouthed about the extent of
hazards which exist outside the specified operating range. Stay in
spec and all is well. Exceed spec and, oh dear. This applies in more
than just EE - eg try and find solid official comment  in MSDSs or
elsewhere on the toxic or lethal dosage of acetaminophen/panadol. It
can be as little as < twice the daily max recommended dose but try and
find that written down somewhere.

ie the job of the spec sheet is to tell you about in spec operation.
Not, alas, how far you can go outside spec. Sometimes its grey. What?
You ran our 100 volt caps at 5 volts? Why would you do that ?

:-)


       Russell


More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...