Leaky Electolytics in Philips 'scope

[Gareth Magennis]

"Ian Field" wrote in message news:g5FQw.1544614$bY7.1450638@fx02.am4...


"Ralph Mowery" <rmowery28146@earthlink.net> wrote in message
news:nNydndhe9vZroJLInZ2dnUU7-XudnZ2d@earthlink.com...

"mike" <ham789@netzero.net> wrote in message
news:men52i$925$1@dont-email.me...
T-rated fuses. In fact I was running off a Variac when the replacement
fuse blew at only about 80V!

Simple experiment.
80V x 2A = 160W.
run it at 75V and see what melts.

When fuses are blowing, that is not a good idea. You often melt the
transformer or something else in adition to what blows.

Fuses are often rated at somewhat above the normal current draw. With a
very bad overload , they will blow quickly enough, but at just a slight
overload, they will not blow and often the transformer or other components
will also overheat and go bad.

Its also a pretty good way to fry a variac - they're not as robust as
regular fixed mains transformers.



I learnt that on almost the first time I used mine.

I would never now fire up a Variac without a lightbulb in series to let me
know what is going on.


(Mine now has a rather crusty section of windings that fortunately still
works after 10 years or so)




Gareth.

 

[Ian Field]

"Gareth Magennis" <sound.service@btconnect.com> wrote in message
news:h%FQw.1282302$ci4.1157342@fx39.am4...

"Ian Field" wrote in message news:g5FQw.1544614$bY7.1450638@fx02.am4...


"Ralph Mowery" <rmowery28146@earthlink.net> wrote in message
news:nNydndhe9vZroJLInZ2dnUU7-XudnZ2d@earthlink.com...

"mike" <ham789@netzero.net> wrote in message
news:men52i$925$1@dont-email.me...
T-rated fuses. In fact I was running off a Variac when the replacement
fuse blew at only about 80V!

Simple experiment.
80V x 2A = 160W.
run it at 75V and see what melts.

When fuses are blowing, that is not a good idea. You often melt the
transformer or something else in adition to what blows.

Fuses are often rated at somewhat above the normal current draw. With a
very bad overload , they will blow quickly enough, but at just a slight
overload, they will not blow and often the transformer or other
components will also overheat and go bad.

Its also a pretty good way to fry a variac - they're not as robust as
regular fixed mains transformers.



I learnt that on almost the first time I used mine.

Its probably because they're very large for a given rating - easy to
misjudge, and then the magic smoke comes out.

 

[Arfa Daily]

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.D13776680BB7FEB7B01029BF@news.eternal-september.org...

which details exactly how the front end of SMPSs like the one in the
Philips
scope, achieve this end with the aid of a single pole voltage select
switch

Arfa

Ah, so the Philips' ps has a voltage doubler at its front end controlled
by
the "mains switch". The input to the ps *proper* is the same--240--fed by
either the 240 mains directly or the 120 mains, doubled.

Thanks,
Dave

That's about the long and the short of it, yes, but of course the input to
the "ps *proper* " as you call it, is DC from the bridge/ doubler, so
actually much higher than 240. Closer to 385 v DC when it is operating on UK
/ European mains

Arfa

 

[DaveC]

That's about the long and the short of it, yes, but of course the input to
the "ps *proper* " as you call it, is DC from the bridge/ doubler, so
actually much higher than 240. Closer to 385 v DC when it is operating on UK
/ European mains

Arfa

Input to the ps proper is either rectified (240vac) mains or rectified
(120vac) mains *doubled*, depending on switch position.

Dave

 

[Ian Field]

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.D13971E70C2EF017B01029BF@news.eternal-september.org...

That's about the long and the short of it, yes, but of course the input
to
the "ps *proper* " as you call it, is DC from the bridge/ doubler, so
actually much higher than 240. Closer to 385 v DC when it is operating on
UK
/ European mains

Arfa

Input to the ps proper is either rectified (240vac) mains or rectified
(120vac) mains *doubled*, depending on switch position.

The usual scheme being the main electrolytic is split into a series pair
across the plus and minus lugs of the bridge rectifier.

For 230V operation the switch is left open - for 120V, the switch is closed
to connect the junction of the 2 electrolytics to one AC lug on the bridge,
that converts the bridge into a voltage doubling rectifier.

In text books, you often see a voltage doubler with the same layout as a
bridge rectifier, but 2 of the rectifiers are replaced by capacitors - the
original diodes of a full complement bridge don't do any harm in a doubler
even though they're not needed, but they are needed for the non-doubling
configuration and leaving them there makes for simpler switching.

 

[Cursitor Doom]

On Wed, 25 Mar 2015 21:44:43 +0000, Gareth Magennis wrote:

I would never now fire up a Variac without a lightbulb in series to let
me know what is going on.

Some clarification of this statement would be nice. I have an old Variac
and don't want to see it get damaged. It has an inline fuse within it so
why isn't that in itself sufficient protection?

 

[Cursitor Doom]

On Thu, 26 Mar 2015 01:45:08 +0000, Arfa Daily wrote:

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.D13776680BB7FEB7B01029BF@news.eternal-september.org...
which details exactly how the front end of SMPSs like the one in the
Philips scope, achieve this end with the aid of a single pole voltage
select switch

Arfa

Ah, so the Philips' ps has a voltage doubler at its front end
controlled by the "mains switch". The input to the ps *proper* is the
same--240--fed by either the 240 mains directly or the 120 mains,
doubled.

Thanks,
Dave


That's about the long and the short of it, yes, but of course the input
to the "ps *proper* " as you call it, is DC from the bridge/ doubler, so
actually much higher than 240. Closer to 385 v DC when it is operating
on UK / European mains

Arfa

And those big electrolytic storage caps are only rated at 350V. I'd have
thought 400 min would be more suitable!

 

[Cursitor Doom]

On Thu, 26 Mar 2015 19:06:07 +0000, Ian Field wrote:

The usual scheme being the main electrolytic is split into a series pair
across the plus and minus lugs of the bridge rectifier.

For 230V operation the switch is left open - for 120V, the switch is
closed to connect the junction of the 2 electrolytics to one AC lug on
the bridge,
that converts the bridge into a voltage doubling rectifier.

That's exactly what Philips have done in this particular PSU.

 

[Ron D.]

I need to remind people, that the wiper of the Variac (Autotransformer) needs to be fused. The windings are rated for current, so if it's 10 A at 120 V. It;s also 10 A at 10 V. A primary fuse won;t protect it.

 

[Arfa Daily]

"Cursitor Doom" <curd@notformail.com> wrote in message
news:mf26ih$fgh$3@dont-email.me...

On Thu, 26 Mar 2015 01:45:08 +0000, Arfa Daily wrote:

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.D13776680BB7FEB7B01029BF@news.eternal-september.org...
which details exactly how the front end of SMPSs like the one in the
Philips scope, achieve this end with the aid of a single pole voltage
select switch

Arfa

Ah, so the Philips' ps has a voltage doubler at its front end
controlled by the "mains switch". The input to the ps *proper* is the
same--240--fed by either the 240 mains directly or the 120 mains,
doubled.

Thanks,
Dave


That's about the long and the short of it, yes, but of course the input
to the "ps *proper* " as you call it, is DC from the bridge/ doubler, so
actually much higher than 240. Closer to 385 v DC when it is operating
on UK / European mains

Arfa

And those big electrolytic storage caps are only rated at 350V. I'd have
thought 400 min would be more suitable!

Think about it ...

Arfa

 

[Cursitor Doom]

On Fri, 27 Mar 2015 02:12:20 +0000, Arfa Daily wrote:

Think about it ...

I am. But the problem at present is that I need to establish if the scope
actually works before I spend money on it. That means contriving some
temporary power supply (or *supplies* more likely) capable of generating
the wide selection of voltages that the scope itself requires. One of the
rails needs 60V and my largest PSU tops out at only 40. It's a PITA.

 

[Cursitor Doom]

On Thu, 26 Mar 2015 19:09:27 -0700, Ron D. wrote:

I need to remind people, that the wiper of the Variac (Autotransformer)
needs to be fused. The windings are rated for current, so if it's 10 A
at 120 V. It;s also 10 A at 10 V. A primary fuse won;t protect it.

My Variac (made by Zenith of London some decades ago by the look of it)
has two fuses: one in the plug as usual and another in the body of the
variac. A curious looking thing with a tight, curly filament wound round
a tiny former and a rating of 3.15A.

 

[Cursitor Doom]

On Fri, 27 Mar 2015 19:23:39 +0000, Ian Field wrote:

In UK equipment (with no doubling) the electrolytic is usually at least
385V - frequently 400V.

In doubling types, the capacitors are usually 200V each.

The ones with PFC front end don't have a reservoir after the bridge -
that comes after the PFC transistor, the most common voltage rating I've
seen is 450V.

Well, 230VAC RMS comes to 325V P-P., so Philips are pushing their luck
with the 350V ones they use in this scope. Especially since there will be
surges above this nominal level and general power station fluctuations
around 230V RMS anyway. The standard caps' voltage rating leaves precious
little allowance for these expected variations.

 

[Ian Field]

"Cursitor Doom" <curd@notformail.com> wrote in message
news:mf23q1$fgh$1@dont-email.me...

On Wed, 25 Mar 2015 21:44:43 +0000, Gareth Magennis wrote:

I would never now fire up a Variac without a lightbulb in series to let
me know what is going on.

Some clarification of this statement would be nice. I have an old Variac
and don't want to see it get damaged. It has an inline fuse within it so
why isn't that in itself sufficient protection?

You probably want the fastest rated fuse you can get away with for a variac.

 

[Ian Field]

"Cursitor Doom" <curd@notformail.com> wrote in message
news:mf26ih$fgh$3@dont-email.me...

On Thu, 26 Mar 2015 01:45:08 +0000, Arfa Daily wrote:

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.D13776680BB7FEB7B01029BF@news.eternal-september.org...
which details exactly how the front end of SMPSs like the one in the
Philips scope, achieve this end with the aid of a single pole voltage
select switch

Arfa

Ah, so the Philips' ps has a voltage doubler at its front end
controlled by the "mains switch". The input to the ps *proper* is the
same--240--fed by either the 240 mains directly or the 120 mains,
doubled.

Thanks,
Dave


That's about the long and the short of it, yes, but of course the input
to the "ps *proper* " as you call it, is DC from the bridge/ doubler, so
actually much higher than 240. Closer to 385 v DC when it is operating
on UK / European mains

Arfa

And those big electrolytic storage caps are only rated at 350V. I'd have
thought 400 min would be more suitable!

In UK equipment (with no doubling) the electrolytic is usually at least
385V - frequently 400V.

In doubling types, the capacitors are usually 200V each.

The ones with PFC front end don't have a reservoir after the bridge - that
comes after the PFC transistor, the most common voltage rating I've seen is
450V.

 

[Ian Field]

"Cursitor Doom" <curd@notformail.com> wrote in message
news:mf4g77$sp1$3@dont-email.me...

On Fri, 27 Mar 2015 19:23:39 +0000, Ian Field wrote:

In UK equipment (with no doubling) the electrolytic is usually at least
385V - frequently 400V.

In doubling types, the capacitors are usually 200V each.

The ones with PFC front end don't have a reservoir after the bridge -
that comes after the PFC transistor, the most common voltage rating I've
seen is 450V.

Well, 230VAC RMS comes to 325V P-P., so Philips are pushing their luck
with the 350V ones they use in this scope. Especially since there will be
surges above this nominal level and general power station fluctuations
around 230V RMS anyway. The standard caps' voltage rating leaves precious
little allowance for these expected variations.

When PC PSUs were part of my bread & butter work, I often found they'd used
150V MOVs to protect each of the series 200V electrolytics. Once or twice I
encountered PSUs that had suffered a cascade failure, started by one of the
under-rated MOVs breaking down.

Maybe I saw a dozen or so that blew up - there must be thousands out there
that didn't.

 

[Arfa Daily]

"Cursitor Doom" <curd@notformail.com> wrote in message
news:mf4g77$sp1$3@dont-email.me...

On Fri, 27 Mar 2015 19:23:39 +0000, Ian Field wrote:

In UK equipment (with no doubling) the electrolytic is usually at least
385V - frequently 400V.

In doubling types, the capacitors are usually 200V each.

The ones with PFC front end don't have a reservoir after the bridge -
that comes after the PFC transistor, the most common voltage rating I've
seen is 450V.

Well, 230VAC RMS comes to 325V P-P., so Philips are pushing their luck
with the 350V ones they use in this scope. Especially since there will be
surges above this nominal level and general power station fluctuations
around 230V RMS anyway. The standard caps' voltage rating leaves precious
little allowance for these expected variations.

You are not thinking about it right, though. With the PSU set to accept a
230 volt input. the two electros are in series, so the 350v rating of them
stacks up to 700 volts across the pair - more than adequate to handle peak
rectified mains plus any 'over-voltages'. You will also note that there is a
resistor across each cap - R1801 and 1802. These serve the primary purpose
of voltage sharing to make sure that the voltage across each of the two caps
is substantially the same, and the secondary purpose of discharging the caps
safely when power is turned off.

As to trying to replace the voltages coming off the PSU, to be honest, I
feel that you are over-thinking it all. I understand that you are not
comfortable with switchers and are trying to prove out the rest of the
circuitry as safely as you can, but with the best will in the world, you are
never going to sub the 1 kV and 1.5 kV outputs of the supply, so are never
going to be able to see anything on the screen anyway.

Based on a great deal of experience of switchers, I would venture to suggest
that you found the only problem that the power supply is going to have been
suffering when you located the short circuit filter, and the best way to
proceed now is to 'just go for it ...'

Arfa

 

[Arfa Daily]

"Cursitor Doom" <curd@notformail.com> wrote in message
news:mf48hd$sp1$2@dont-email.me...

On Thu, 26 Mar 2015 19:09:27 -0700, Ron D. wrote:

I need to remind people, that the wiper of the Variac (Autotransformer)
needs to be fused. The windings are rated for current, so if it's 10 A
at 120 V. It;s also 10 A at 10 V. A primary fuse won;t protect it.

My Variac (made by Zenith of London some decades ago by the look of it)
has two fuses: one in the plug as usual and another in the body of the
variac. A curious looking thing with a tight, curly filament wound round
a tiny former and a rating of 3.15A.

"T" rated fuses often have a spiral element

Arfa

 

[Cursitor Doom]

On Sat, 28 Mar 2015 01:28:57 +0000, Arfa Daily wrote:

You are not thinking about it right, though. With the PSU set to accept
a 230 volt input. the two electros are in series, so the 350v rating of
them stacks up to 700 volts across the pair - more than adequate to
handle peak rectified mains plus any 'over-voltages'. You will also note
that there is a resistor across each cap - R1801 and 1802. These serve
the primary purpose of voltage sharing to make sure that the voltage
across each of the two caps is substantially the same, and the secondary
purpose of discharging the caps safely when power is turned off.

Argh! Thanks for that, Arfa. You're right of course. I still have linear
PSU topology burned into my consiousness (large electrolyitcs in
parallel). I need to get it into my thick head that switchers use a
different approach altogether (I'm a bit slow to adapt given my advanced
age).

As to trying to replace the voltages coming off the PSU, to be honest, I
feel that you are over-thinking it all. I understand that you are not
comfortable with switchers and are trying to prove out the rest of the
circuitry as safely as you can, but with the best will in the world, you
are never going to sub the 1 kV and 1.5 kV outputs of the supply, so are
never going to be able to see anything on the screen anyway.

I was going to use the existing transformers for the HT stuff and just
rig up my PSUs for the sub 60V stuff, tbh. But I admit it would be messy
and fiddly.

Based on a great deal of experience of switchers, I would venture to
suggest that you found the only problem that the power supply is going
to have been suffering when you located the short circuit filter, and
the best way to proceed now is to 'just go for it ...'

Whilst it was out of the scope I did test it and it was dead; no chopping
and no supply volts even to the PWM chip. HOWEVER, ISTR from somewhere
that these things shut down if they don't see a load, so I'll leave my
"outthinking" behind for a mo (you're not the first to say that) and
reconnect everything back in circuit and re-try.
I'll report back in due course.

 

[Cursitor Doom]

On Sat, 28 Mar 2015 17:06:23 +0000, Arfa Daily wrote:

Be very very - and I mean VERY - careful what you are measuring, and
how. For a start, you should never work on a switcher without having it
on an isolation transformer.

Indeed! I have actually ordered one (a 500W model) as I wasn't
comfortable with a mere RCD in combo with a variac between the scope and
the wall socket. In addition to the safety aspect conferred by a IT, it's
always annoying and frightening when you blow a bloody great hole in a
probe's ground shield - as I'm sure we've all done at some point - when
completing an earth loop you didn't know was there.


Secondly, it's not easy to safely measure

voltages on the primary side of the PSU without knowing what you are
going to use as a ground reference for your meter or scope. Usually,
it's the -ve side of the bridge reccy, but always check the schematic to
be sure. In the case of the Philips, that does appear to be the right
place. Bear in mind we are talking peak rectified mains here, limited in
current capacity only by the value of any primary side fuse. Skull and
crossbones stuff ... If you are not confident measuring with two hands
in this sort of environment, solder a wire to the -ve reference point
and clip your meter -ve probe to it. Then put the hand that you would
have been holding that probe with firmly in your pocket, and don't bring
it back out again until you've finished measuring and moved away, or
turned it back off. Even with it operating on an isolation transformer,
you aren't safe against hanging yourself *across* either the mains input
or the output side of the bridge. I'm sure that you probably know all
this, but better safe than sorry, and worth mentioning for any other
readers who might not be aware of the potential dangers of working
'live' on switchers ...

Quite, good point. I'm in the especially dangerous category inasmuch as I
only know enough to be dangerous. I have gaps in my knowledge. I need to
spend more time refreshing my knowledge of testing safety procedures. I
do recall the old 'hand behind your back' rule from a long way back,
though. And I don't fancy working two-handed with some of these DVMs you
see nowadays that claim to be CAT III or IV but are anything but when you
take them apart.

Supply volts for the PWM chip are on pin 1, and are supplied initially
from the rectified mains via the circuitry around V1803 and 1804. Once
the supply gets going, it is self-supplied via the diodes at V1816, 17.
I don't immediately see any active feedback from the secondary side, so
I would be surprised if it didn't at least try to start without
secondary side loads, but switchers always have the capacity to do
things that you don't expect ! There is some DC feedback from the
primary side via V1818 and 19, but I would still expect it to start
whilst monitoring for a control voltage coming back from there. It looks
as though it uses the level of voltage from there to control the PWM
output to set the base level of output volts and keep them stable
against secondary side load.

Many thanks for your observations here.

I have a copy of the datasheet for the chip if you would like me to send
it.
It details the operation of the chip pretty well, and shows some sample
implementations of it which are pretty close to how it's being used in
the Philips, if you can get your mind to re-arrange the way they have
been drawn ...

I managed to get a DS for the chip pretty quickly, it was the scope
schematics that were the big problem. I'll have to check out that site
you mentioned in your email and see what's there. I've got another 2 or 3
Philips scopes to sort out when this one's done so I'm sure it would pay
off.