failure of the low-ESR electrolytic capacitor in second winding

[legg]

The OP really needs to make some measurements. He should get a scope photo
of the secondary winding current, and he should get some of the temperature
indicating labels another poster mentioned and stick them on the caps.

Otherwise, we're all just speculating.

Can't say it was a waste of time, for me.

I think Simon had his answers a while ago and posted his regards.

RL

 

[Eeyore]

Find refresher penance attached on a.b.s.e.

Exactly.

The ripple current is NOT the same as the load current. It's very strongly affected
by the conduction angle/period.

Graham

 

[legg]

This expression does not imply that "...the ripple current in the capacitor
is not greater than the output current...". It merely implies that both
the ripple current in the cap and the output current are less than the
secondary RMS current.

For example, suppose Isecrms = 5. Then either (Icaprms = 3 AND Ioutdc = 4)
or (Icaprms = 4 AND Ioutdc = 3) will satisfy the expression.

Having recently revisited rms summation, I see your point.

RL

 

[The Phantom]

Exactly.

The ripple current is NOT the same as the load current.

Legg's calculations, and mine, show that for the discontinuous conduction
case with constant output current, they are the same (value) for a duty
cycle somewhere between 60% and 70%.

It's very strongly affected by the conduction angle/period.

So it is. And, in fact, it's so strongly affected, that below a certain
duty cycle, the capacitor ripple current is greater than the output
current, and above that duty cycle, the capacitor ripple current is less
than the output current.

 

[Eeyore]

Legg's calculations, and mine, show that for the discontinuous conduction
case with constant output current, they are the same (value) for a duty
cycle somewhere between 60% and 70%.

SUBJECT to an assumption that the current pulse is perfectly rectangular if I understand
correctly.

That seems to be a very big assumption especially in the case of a flyback supply.

So it is. And, in fact, it's so strongly affected, that below a certain
duty cycle, the capacitor ripple current is greater than the output
current,

Which is entirely normal for ac line frequency power supplies for example at all power
levels..

and above that duty cycle, the capacitor ripple current is less
than the output current.

Yes, that bit did actually surprise me, but it's clear from examination of the
mathematical analysis, subject again to the assumption of a rectangular current
waveform.

What would interest me more still is just how valid (or otherwise) that assumption is. I
feel some modelling coming on.

Graham

 

[Jamie]

Jamie wrote:




More gibberish from Jamie (the perennially clueless) as ever.

Moron you are.
You just have no clue on proper designs..

You're out of your league..

 

[The Phantom]

SUBJECT to an assumption that the current pulse is perfectly rectangular if I understand
correctly.

I said just above your reply, "...discontinuous conduction..."; the
current pulse is not rectangular in that mode. See the image of the
current waveform for this mode in my most recent post on ABSE, but actually
it doesn't matter if the current pulse is perfectly rectangular.

That seems to be a very big assumption especially in the case of a flyback supply.

So it is. And, in fact, it's so strongly affected, that below a certain
duty cycle, the capacitor ripple current is greater than the output
current,

Which is entirely normal for ac line frequency power supplies for example at all power
levels..

and above that duty cycle, the capacitor ripple current is less
than the output current.

Yes, that bit did actually surprise me, but it's clear from examination of the
mathematical analysis, subject again to the assumption of a rectangular current
waveform.]

Again, I was talking about discontinuous conduction mode, but that doesn't
change the conclusion.

What would interest me more still is just how valid (or otherwise) that assumption is. I
feel some modelling coming on.

For all three modes which I analyzed, it's true that "...below a certain
duty cycle, the capacitor ripple current is greater than the output
current, and above that duty cycle, the capacitor ripple current is less
than the output current.".

 

[Simon]

Dear Eeyore,

I just came back to this subject, i didn't know it's been more
complicated.

I thought i was clear, seems not now.

The swollen capacitor happened at different customer site, one
application, the amibient temperature is about 40 degree due to its
poor air flow.

but others seems should be more cooler due to the appliation.


Thanks

 

[Winfield]

Dear Eeyore,

I just came back to this subject, i didn't know it's been more
complicated.

I thought i was clear, seems not now.

The swollen capacitor happened at different customer site, one
application, the amibient temperature is about 40 degree due to its
poor air flow.

But others seems should be more cooler due to the application.

Thanks

There was only one supply with a failed cap?
Or did many fail at that high-temperature site?