Maker Pro
Maker Pro

Capacitor rating question

E

Ed Chilada

Jan 1, 1970
0
I have a failed PSU where a rather smelly capacitor appears to be the
problem. It's rated 1200uf/16v but my local Maplin doesn't stock it.
However they do stock 1000uf/16v and 2200uf/16v. Since I don't really
know what the figures mean, does anyone else know whether either of
these are likely to work? I dunno whether the 2200 means it's overkill
for the application but it would work, so long as I can get it to fit
or something.

Thanks
 
P

Pete Wilcox

Jan 1, 1970
0
I have a failed PSU where a rather smelly capacitor appears to be the
problem. It's rated 1200uf/16v but my local Maplin doesn't stock it.
However they do stock 1000uf/16v and 2200uf/16v. Since I don't really
know what the figures mean, does anyone else know whether either of
these are likely to work? I dunno whether the 2200 means it's overkill
for the application but it would work, so long as I can get it to fit
or something.
The 2200uF would be the better choice, but you would probably get away
with using either value. Electrolytics are usually something like +50/-20
per cent tolerance rated, so there can be a large difference between
what's in the can and what's written on it.

Cheers,
Pete.
 
G

Gerard Bok

Jan 1, 1970
0
I have a failed PSU where a rather smelly capacitor appears to be the
problem. It's rated 1200uf/16v but my local Maplin doesn't stock it.
However they do stock 1000uf/16v and 2200uf/16v. Since I don't really
know what the figures mean, does anyone else know whether either of
these are likely to work? I dunno whether the 2200 means it's overkill
for the application but it would work, so long as I can get it to fit
or something.

That all depends on the type of PSU !
(Given the off-standard value of 1200 uF, you are likely talking
about a SMPS.)

In a SMPS (Switched mode PSU) like in a PC or in modern low
weight wall chargers, you cannot change 1200 uF into 2200 uF nor
into anything else. Even more: you cannot replace a 1200 uF
capacitor by just any 1200 uF capacitor as it then probably needs
to be a special low ESR type :)

On the other hand: if it is a classic PSU (one with a rather
heavy transformer), it is OK to replace your 1200 uF capacitor by
2200 uF. If the new one fits the available space, that is.
Or even replace it by 1000 uF.
 
J

Jamie

Jan 1, 1970
0
Ed said:
I have a failed PSU where a rather smelly capacitor appears to be the
problem. It's rated 1200uf/16v but my local Maplin doesn't stock it.
However they do stock 1000uf/16v and 2200uf/16v. Since I don't really
know what the figures mean, does anyone else know whether either of
these are likely to work? I dunno whether the 2200 means it's overkill
for the application but it would work, so long as I can get it to fit
or something.

Thanks
the 2200 will work fine how ever, ESR could be factor.
try to get a low ESR type. It will specify low ESR in the
parts search.


http://webpages.charter.net/jamie_5"
 
P

Pete Wilcox

Jan 1, 1970
0
That all depends on the type of PSU !
(Given the off-standard value of 1200 uF, you are likely talking
about a SMPS.)
Woops. I stand corrected. Have to admit, when I sent my original reply I
was just thinking in terms of a smoothing capacitor in a bog-standard
linear PSU. If it's a switcher, that's a whole different ball-game.

Cheers,
Pete.
 
E

Eeyore

Jan 1, 1970
0
Ed said:
I have a failed PSU where a rather smelly capacitor appears to be the
problem. It's rated 1200uf/16v but my local Maplin doesn't stock it.
However they do stock 1000uf/16v and 2200uf/16v. Since I don't really
know what the figures mean, does anyone else know whether either of
these are likely to work? I dunno whether the 2200 means it's overkill
for the application but it would work, so long as I can get it to fit
or something.

Will either of those values fit OK into the vailable space ?

If the 1200uF failed it may mean it was under-rated for the job and a
2200uF might well be the better choice, but as ever it's never 'quite' as
simple as that. Simply a better brand of capacitor may provide as much
improvement as you require. You can't go far wrong with Panasonic caps
btw. If this was some cheap Chinese piece of kit it's unlikely the
original cap would remotely match a Panasonic part for 'ripple current'
which is probably what's killed it. Is the cap case 'bulged' too ?

One other thing, is this a swichmode PSU or 50/60Hz with a traditional AC
transformer ? Switchmode supplies require 'special' low-ESR caps designed
for the job. Don't fit a 'standard' cap into a switchmode supply.

Graham
 
E

Ed Chilada

Jan 1, 1970
0
That all depends on the type of PSU !
(Given the off-standard value of 1200 uF, you are likely talking
about a SMPS.)

In a SMPS (Switched mode PSU) like in a PC or in modern low
weight wall chargers, you cannot change 1200 uF into 2200 uF nor
into anything else. Even more: you cannot replace a 1200 uF
capacitor by just any 1200 uF capacitor as it then probably needs
to be a special low ESR type :)

Cheers for the help everyone. Yeah it was from a slimline PC's PSU
which is switched mode. I'll look into getting the proper replacement
part although I'm actually still wrangling with the warranty and/or
house insurance, but if it turned out to be a cheap fix I thought I
could avoid the hassle. Thanks again.
 
G

Gerard Bok

Jan 1, 1970
0
I am also working on a failed power supply, and I *believe* it is an SMPS as
it lacks he heavy transformer and other "standard" components I am familiar
with. My question: How can I tell what the ESR of a replacement (high temp)
cap is? One of the caps in the PS literally exploded,

If an electrolytic capacitor in a PSU leaks, it's probably a
production problem (on the part of the manufacturer that produced
the capacitor).
If a capacitor explodes, it's most probably because of some other
fault. (E.g. short circuit in a rectifier.)

If a capacitor exploded, you are facing several problems.
- liquid from inside the capacitor is splattered all over the
surrounding PCB. Very hard to clean. And very conductive stuff.
- the capacitor exploded for a reason (see above)
- a good replacement is often very hard to get,

So my advice would be: replace the PSU!

(Which, by the way, is allways the best advice for a failing
SMPS. Many enthousiasts don't realise that a quick and cheap
'fix' may cause them dearly. As insurers are very keen to trace
home fires back to 'inadequate modification' of electronic
equipment.)
 
E

Eeyore

Jan 1, 1970
0
Dave said:
Aaah. Okay. Well, replacing the PS PCB was what I actually did, but I was
hoping to refurbish the old one in case the new one went out too. I was
planning on replacing the missing/swollen caps with new, higher
voltage-rated ones to (in theory at least) give them better heat-dissipating
capabilities. If I still wanted to do this, couldn't I just clean the
electrolyte from the exploded cap off the PCB with one of the spray-on
cleaning solutions? Or would I need something like the
now-impossible-to-get Freon TF?

I think most modern electrolytes are water soluble. Use warm water but onbiously
avoid soaking any adjacent components.

How old is your PSU btw ? You may be a victim of the 'bad caps' problem of some
years back.

http://en.wikipedia.org/wiki/Capacitor_plague

Certain 'off-brand' caps are especially noted for the problem.

Graham
 
G

Gerard Bok

Jan 1, 1970
0
BTW, if the electroloyte is sprayed around by the exploding cap, wouldn't I
need to wash the board thoroughly, to get rid of it? And I didn't know
electroloyte was conductive. I figured it would be non-conductive, to
assiste in keeping the "plates" of the cap separated.

No. The plates have a very rough surface, increasing their
surface area. The electrolyte follows those curves on the surface
and acts as the second plate.
A thin layer of oxyde, covering the plate forms the insulator.
(That's why they sometimes only withstand some 10 volts :)

What you are looking at is (partially) 'boiled' electroyte, which
may not have the same characteristics as the original one.
And the aggressive nature of the stuff may have affected the
surrounding materials.
 
E

Eeyore

Jan 1, 1970
0
Dave said:
The PS PCB is two or three years old. I half-way suspect this may be the
problem, but that is just a gut feeling.

BTW, if the electroloyte is sprayed around by the exploding cap, wouldn't I
need to wash the board thoroughly, to get rid of it?

Depends where it went. If 'under' some components, it's best to remove those and
clean thoroughly.

And I didn't know electroloyte was conductive. I figured it would be
non-conductive, to
assiste in keeping the "plates" of the cap separated.

It's slightly conductive. Electrolytics caps are very imperfect devices. All
electrolytes have some conductivity.

Graham
 
N

Nemo

Jan 1, 1970
0
I like the description of the capacitor:
a rather smelly capacitor

I shall have to define smell in my next parts list. "C3: 10uF, 16V, a
redolent fruity little device with a hint of chives."

I have designed several SMPS's and in every case it was OK to increase
the capacitance, in fact you generally made the caps as big as you could
afford, to reduce ripple levels. So I would think 2200uF is better,
though as someone said, you want a "low ESR" type if possible (to stop
it getting too warm from ripple currents heating it). I don't understand
why Mr Bok advised against changing the capacitance, but perhaps it is
different for multi output PSU's (I did simple non isolated step down or
step up types). I'd be interested to learn more on this subject.

Nor do I understand why it is bad to wash the board thoroughly with
water. Many PCB's are washed with water to get rid of acidic flux
remnants after assembly, and provided you dry it thoroughly afterwards
(ie on a radiator or something, fairly rapidly, so nothing rusts) what
is going to get damaged by water? Obviously you don't want water on a
PCB when it's got mains voltages on it, but there's no reason not to use
water for cleaning it.
 
E

Eeyore

Jan 1, 1970
0
Nemo said:
I don't understand why Mr Bok advised against changing the capacitance,

It will change the loop response and subsequently the stability of the
feedback loop.

Graham
 
Top