Small Capacitors: What Numbers Critical?

[(PeteCresswell)]

I've got a couple of IP cams where the same cap failed.

It's a 15 uF, 100v radial cap spec'd to 105c.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

Can I shop for something with slightly different specs?

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

What about uF? Do I use a 20 or 30 uF at my own risk?
--
Pete Cresswell

 

John Robertson <spam@flippers.com> wrote:

On 04/07/2014 6:30 AM, (PeteCresswell) wrote:

But I'm having fits trying to find a replacement where they don't
rape me on shipping.

Order two of these
http://www.digikey.com/product-detail/en/UPW2A150MPD/493-1968-ND/589709
at 38 cents each from Digi-Key and send them a check. (If the lead
spacing or physical size isn't correct, they have three other 15 uF,
100 V, 105 C caps for 36 to 58 cents each.) If you send Digi-Key a
check with the order, they will pay for the shipping to the US and
Canada. You will be out 49 cents for the stamp.

> Where are you located that shipping is such a cost?

He neglected to say in his first post about this problem. Several
people, including me, gave him ideas, which he did not respond to.

https://groups.google.com/forum/#!topic/sci.electronics.repair/Yv9EDsOC9GQ

Part of the problem is that I think he wants to buy just two capacitors,
so the typical $5-$7 shipping charge is much greater than the parts
cost.

Matt Roberds

 

[RobertMacy]

On Mon, 07 Apr 2014 06:30:39 -0700, (PeteCresswell) <x@y.invalid> wrote:

I've got a couple of IP cams where the same cap failed.

It's a 15 uF, 100v radial cap spec'd to 105c.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

Can I shop for something with slightly different specs?

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

What about uF? Do I use a 20 or 30 uF at my own risk?

Depends a bit on failure mode.

same cap, same location? probably being overstressed either with voltage,
which isn't likely because designers understand voltage, but overstressed
with excessivae AC ripple current, since most designers don't understand
that stuff and simply spec for high temp, like 105C --> you see the
thought process here?

Therefore *IF* you replace go for same or more uF, same or more voltage
[tilt towards more voltage], and definitely LOWEST Resr you can find.
....my opinion only, without looking deeper.

 

"Yes. However, if the cap is in a circuit that requires an extremely
low ESR, you may have problems with a lower voltage substitute."

I assume you meant higher there. The old formula y'know :

+capacitance
+voltage
-size
-ESR (+temperature + ripple current rating)
=price

Just mentioning this because since the question was asked, the poster does not know.

It may sound strange to some, but in some cases it might be better to use a LOWER capacity. Really, is a switchmode is running at 100 Khz, does it really need a 2,200 uF ? Hell no, they used that value to get the best price on the ripple current/ESR. In some cases maybe they want certain supplies to fall before others on power down, but that's usually not critical enough to worry about and that esotera of designing is usually out the window when the thing is barely out of warranty.

If you get a lower value lytic at the same voltage and the same size, it is likely to have lower ESR, no ?

Of course if the thing just runs off an AC trnsformer then it's different. If that's the case, don't lower the value.

In the case of the OP, if you don't know, just use the original value but select one from Digikey with lower ESR, higher ripple current rating and you should be fine as long as it physically fits.

At any rate, here is the Digikey page on it :

http://www.digikey.com/product-search/en?FV=fff40002%2Cfff80009%2C34005d%2C402e23%2Cfc01ac%2C1140050&mnonly=0&newproducts=0&ColumnSort=1000011&page=1&stock=0&pbfree=0&rohs=0&quantity=&ptm=0&fid=0&pageSize=25

The highest priced one is 58 cents. From there, measure what you got, look at the ones that are the size you need and get the one with the highest ripple current rating.

And then of course, replace them all.

 

Now that I've seen the pictures, tell you this much, that looks like that conductive glue.

NO, it's not supposed to be conductive...

Remove it in all units, it'll eat your solder mask and create all kinds of problems if it is what I think it is.

Of course it has been known to fail like that for the last fifteen years, that's why they use it, to sell new cameras when the old ones fail.

 

"These caps may be OK - if you made the assumption that the brown goo
indicated failure - and the problem may be elsewhere... "

True, I didn't consider that.

 

[dave]

On 04/07/2014 07:08 AM, RobertMacy wrote:

On Mon, 07 Apr 2014 06:30:39 -0700, (PeteCresswell) <x@y.invalid> wrote:

I've got a couple of IP cams where the same cap failed.

It's a 15 uF, 100v radial cap spec'd to 105c.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

Can I shop for something with slightly different specs?

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

What about uF? Do I use a 20 or 30 uF at my own risk?

Depends a bit on failure mode.

same cap, same location? probably being overstressed either with
voltage, which isn't likely because designers understand voltage, but
overstressed with excessivae AC ripple current, since most designers
don't understand that stuff and simply spec for high temp, like 105C --
you see the thought process here?

Therefore *IF* you replace go for same or more uF, same or more voltage
[tilt towards more voltage], and definitely LOWEST Resr you can find.
...my opinion only, without looking deeper.

Is this an electrolytic? Polarized?

 

[John Robertson]

On 04/07/2014 6:30 AM, (PeteCresswell) wrote:

I've got a couple of IP cams where the same cap failed.

It's a 15 uF, 100v radial cap spec'd to 105c.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

Can I shop for something with slightly different specs?

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

What about uF? Do I use a 20 or 30 uF at my own risk?

Where are you located that shipping is such a cost?

As for larger values of capacitance, that may or may not work depending
on the job for the original cap. If part of a switching supply then I'd
say, no - don't put in a higher value. If part of a linear supply, then
a 20 instead of 15 should be OK, but not bigger.

Have you contacted the manufacturer to see if they have a service
bulletin out on this failure?

John :-#)#

--
(Please post followups or tech inquiries to the newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

 

[Fred McKenzie]

In article <st95k9lptec59epfmlrdg77vubrk3p7eg8@4ax.com>,
"(PeteCresswell)" <x@y.Invalid> wrote:

I've got a couple of IP cams where the same cap failed.

It's a 15 uF, 100v radial cap spec'd to 105c.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

Can I shop for something with slightly different specs?

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

What about uF? Do I use a 20 or 30 uF at my own risk?

Pete-

I saw this same question recently. I did not reply because I thought
others had good answers. Basically, the answer "depends"!

A higher voltage is acceptable unless the physical size is a problem.

Higher capacitance might be OK if it is used as a power supply filter.
If it is used as part of a timing circuit, then stay close to the
original value.

Failures of the same part on multiple devices could be the result of the
original manufacturer buying a bad batch and/or counterfeit parts. If
your replacements do not solve the problem, it could be due to a poor
design that over-stresses the part.

Fred

 

[Jeff Liebermann]

On Mon, 07 Apr 2014 09:30:39 -0400, "(PeteCresswell)" <x@y.Invalid>
wrote:

>I've got a couple of IP cams where the same cap failed.

I'll assume these were installed in different locations. Therefore,
it is unlikely that some major power glitch or overload condition
caused the cap to fail. Most likely, all the caps were defective on
arrival. If you have some of the same model IP camera where the cap
hasn't failed yet, you might consider a pre-emptive cap replacement.

>It's a 15 uF, 100v radial cap spec'd to 105c.

I would be interested in the brand name or better yet, a photo.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

<http://www.ebay.com/itm/321317467092> $3.19 plus $1.00 shipping.

>Can I shop for something with slightly different specs?

Yes. However, if the cap is in a circuit that requires an extremely
low ESR, you may have problems with a lower voltage substitute. If
this cap is the output filter for a small switching power supply
inside the camera, I would stay with the 100v rating. Otherwise, feel
free to substitute. However, I doubt that it's in an ESR sensitive
circuit. If low ESR was a requirement, then the ideal cap would have
a much higher capacitance, and a voltage rating as low as possible.

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

It probably won't fit.

>What about uF? Do I use a 20 or 30 uF at my own risk?

That depends on the circuit. If it's in a timing circuit, it's a bad
idea. If it's a power supply bypass, just about anything will work,
unless the designer got clever and used some kind of resonant scheme
to improve the ESR at the switching frequency.

Hint: The quality of the answers depends heavily on the quality of
the information you supply. You left out the maker and model of the
camera, the maker of the capacitor, where in the circuit it's used,
and under what conditions it is failing.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[(PeteCresswell)]

Per dave:

Is this an electrolytic? Polarized?

I can't find any indication of polarity.

I'm not sure what constitutes electrolytic, but when they failed a
brownish goo leaked out and solidified.

Here are a couple of pix:

http://tinyurl.com/manvbbg
http://tinyurl.com/n82vgpj


--
Pete Cresswell

 

[John Robertson]

On 04/07/2014 10:34 AM, (PeteCresswell) wrote:

Per dave:

Is this an electrolytic? Polarized?

I can't find any indication of polarity.

I'm not sure what constitutes electrolytic, but when they failed a
brownish goo leaked out and solidified.

Here are a couple of pix:

http://tinyurl.com/manvbbg
http://tinyurl.com/n82vgpj

That brownish goo on the top of those capacitors is the glue to hold it
to the PCB. If the caps leaked it would be at the other end, where the
two leads come out of the (usually) black plastic sealed end, not
through the aluminum top!

These are standard electrolytic caps - if you look at the side closest
to the camera you will see the <-><-> band which is the negative.

These caps may be OK - if you made the assumption that the brown goo
indicated failure - and the problem may be elsewhere...

John :-#(#
--
(Please post followups or tech inquiries to the newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

 

[Michael Black]

On Mon, 7 Apr 2014, Jeff Liebermann wrote:

On Mon, 07 Apr 2014 09:30:39 -0400, "(PeteCresswell)" <x@y.Invalid
wrote:

I've got a couple of IP cams where the same cap failed.

I'll assume these were installed in different locations. Therefore,
it is unlikely that some major power glitch or overload condition
caused the cap to fail. Most likely, all the caps were defective on
arrival. If you have some of the same model IP camera where the cap
hasn't failed yet, you might consider a pre-emptive cap replacement.

It's a 15 uF, 100v radial cap spec'd to 105c.

I would be interested in the brand name or better yet, a photo.

But I'm having fits trying to find a replacement where they don't rape
me on shipping.

http://www.ebay.com/itm/321317467092> $3.19 plus $1.00 shipping.

Can I shop for something with slightly different specs?

Yes. However, if the cap is in a circuit that requires an extremely
low ESR, you may have problems with a lower voltage substitute. If
this cap is the output filter for a small switching power supply
inside the camera, I would stay with the 100v rating. Otherwise, feel
free to substitute. However, I doubt that it's in an ESR sensitive
circuit. If low ESR was a requirement, then the ideal cap would have
a much higher capacitance, and a voltage rating as low as possible.

i.e. is 100v just an upper limit beyond which the cap will fail... and I
could replace it with a 200v or 300v cap?

It probably won't fit.

What about uF? Do I use a 20 or 30 uF at my own risk?

That depends on the circuit. If it's in a timing circuit, it's a bad
idea. If it's a power supply bypass, just about anything will work,
unless the designer got clever and used some kind of resonant scheme
to improve the ESR at the switching frequency.

Hint: The quality of the answers depends heavily on the quality of
the information you supply. You left out the maker and model of the
camera, the maker of the capacitor, where in the circuit it's used,
and under what conditions it is failing.

And where he is.

No wonder the question sounded familiar, he's the same guy who was asking
a similar question last week.

Michael

 

>"Can you expand on this Brown Glue issue a bit or provide reference links? "

Look under Mitsubishi glue in reference to TV repair. They used it for a number of years and it caused alot of problems. It starts out white, when it turns brown is when it starts being conductive. They eventually stopped using that type of glue, probably because in the business it was becoming so widely known even some of the consumers knew about it.

Other manufacturers used it as well but none in the brownwares business as liberally as Mitsubishi.

You could literally stick ohmmeter probes in it and reason it like a resistor, and not always ten megs. I've seen glops of it read like 10 Kohms.

After they stopped using it they then went to lytics that leaked electrolyte all over the board, not only becoming conductive but actually eating the copper traces awat. In an RPTV, if that failed to kill the unit, their coolant chamber design was defective enough to leak the ethylene glycol/glycerine coolant mixture on the board and it, given a little time with voltage applied also became conductive and corrosive.

So, what Mitsubishi, as well as a few other RPTV manufacturers did, instead of desiging the pressure relief (bladder) correctly, the installed gutters in the sets because it was found that if the coolant leaked on parts of the board woith higher voltage and current, like the horizontal or high voltage, it could cause a fire. Think of what the lawyers would have said ! AQctually I have seen units in which the board actually caught fire but in those cases it did not burn the whole house down. If it had, the road men would have noticed something funny during the initial service call. They're observant that way...

As such, even though it is well known all over the industry not to use that type of glue, I am confident that some manufacturers will use that type of glue. Think of what the accountants would say.

 

Y'know, there is no way in hell those things need a 100 volt rating.

Unless they use something like an old iconoscope tube or something. No way.

 

[John Robertson]

On 04/07/2014 11:15 AM, jurb6006@gmail.com wrote:

Now that I've seen the pictures, tell you this much, that looks like that conductive glue.

NO, it's not supposed to be conductive...

Remove it in all units, it'll eat your solder mask and create all kinds of problems if it is what I think it is.

Of course it has been known to fail like that for the last fifteen years, that's why they use it, to sell new cameras when the old ones fail.

OK, that's a new one for me, that brown glue can become conductive over
time? I'll have to pull out a megger and check that out, but it won't
surprise me now.

Can you expand on this Brown Glue issue a bit or provide reference links?

Thanks for the tip!

John :-#)#

--
(Please post followups or tech inquiries to the newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

 

[RobertMacy]

On Mon, 07 Apr 2014 10:55:46 -0700, John Robertson <spam@flippers.com>
wrote:

...snip...
That brownish goo on the top of those capacitors is the glue to hold it
to the PCB. If the caps leaked it would be at the other end, where the
two leads come out of the (usually) black plastic sealed end, not
through the aluminum top!

These are standard electrolytic caps - if you look at the side closest
to the camera you will see the <-><-> band which is the negative.

These caps may be OK - if you made the assumption that the brown goo
indicated failure - and the problem may be elsewhere...

John :-#(#

like put in backwards? naw, not at 100V, but I've seen such at 6V

 

[(PeteCresswell)]

Per mroberds@att.net:

Where are you located that shipping is such a cost?

He neglected to say in his first post about this problem. Several
people, including me, gave him ideas, which he did not respond to.

https://groups.google.com/forum/#!topic/sci.electronics.repair/Yv9EDsOC9GQ

Part of the problem is that I think he wants to buy just two capacitors,
so the typical $5-$7 shipping charge is much greater than the parts
cost.

Mea Culpa.

You nailed it. Ten bucks to ship two 89-cent caps that weigh less than
a half ounce? It's more a matter of principle... that plus thinking
that there *must* be a source where shipping charges are more rational.
I've had small circuit boards shipped all the way from Guandong Province
for zero shipping. OK, that's not "rational"... more like the Chinese
government subsidizing their businesses...

But at least 60 percent of the stuff I purchase via Amazon.com gets
shipped for free at a competitive price. (i.e. shipping/handling is not
rolled into the price).

But now that you've called me on it... I guess I will go the DigiKey
route.
--
Pete Cresswell

 

[(PeteCresswell)]

Per Jeff Liebermann:

Hint: The quality of the answers depends heavily on the quality of
the information you supply. You left out the maker and model of the
camera, the maker of the capacitor, where in the circuit it's used,
and under what conditions it is failing.

I didn't want to call attention to the maker because these cams work
pretty well for their cost (about $100). They are EdiMax IC-3030PoE's
as in
http://www.edimax.com/en/produce_detail.php?pd_id=360&pl1_id=8&pl2_id=91

I have one more that has not failed yet. All 3 have had different
histories - one of them having survived Hurricane Sandy mounted outdoors
in a plastic bag about 25' from the bay.

Once I have a supply of those caps - and it turns out that replacing the
cap makes the camera work again - I'll buy another IC-3030PoE in a
heartbeat.

Thanks for the analysis of what values matter when. I think I've
learned something worth knowing from all this.
--
Pete Cresswell

 

[Arfa Daily]

"(PeteCresswell)" <x@y.Invalid> wrote in message
news:8un5k9ht6a67ncrt0hbrjikbjivrjvluk3@4ax.com...

Per dave:

Is this an electrolytic? Polarized?

I can't find any indication of polarity.

I'm not sure what constitutes electrolytic, but when they failed a
brownish goo leaked out and solidified.

Here are a couple of pix:

http://tinyurl.com/manvbbg
http://tinyurl.com/n82vgpj


--
Pete Cresswell

That 'leaked goo' is just glue that the manufacturer has used to stick a
radial cap down to the PCB so that it 'pretends' to be an axial type. What
checks have you done to establish that the cap is faulty ? ESR ? This is the
main check that you always have to do on any suspect electrolytics. Value ?
Usually of secondary consideration to ESR. Physically short circuit ?

If you are basing your belief that the cap is faulty on what you perceived
to be something that had leaked out from within, then there's a very good
chance that you are mistaken.

Arfa