[Fu Knee]

Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

Thunderstorms found the weakness.

Now using 160uF 150V caps, Panasonic HFS, with one from GND to
Kathode, and a second from GND to Grid bias zero, not GND.

Seems to work.

Bypass bypass is a 1 uF 630V PP.

Wild and crazy Sat night.

Happy Ears!
Al

[Patrick Turner]

On Oct 4, 1:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

Thunderstorms found the weakness.

Mon Dieu, Bum Farts caused by Thunderstorms!


Now using 160uF 150V caps, Panasonic HFS, with one from GND to
Kathode, and a second from GND to Grid bias zero, not GND.

Seems to work.

Bypass bypass is a 1 uF 630V PP.

Wild and crazy Sat night.

I won a couple and lost a couple of chess games on saturday night.

Its not nearly as crazy as painting the town red with Jennifer
Hawking.

Patrick Turner.

Happy Ears!
Al

[Bret L]

Fixed bias works better.

[[email protected]]

On Oct 3, 10:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

50V caps have no place in tube equipment.

Peter Wieck
Melrose Park, PA

[Don Pearce[_3_]]

On Mon, 05 Oct 2009 07:17:41 -0500, flipper wrote:

On Sun, 4 Oct 2009 17:44:03 -0700 (PDT), "
wrote:

On Oct 3, 10:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

50V caps have no place in tube equipment.

Peter Wieck
Melrose Park, PA

Rk bypass

DC heater smoothing.

d

[[email protected]]

On Oct 5, 8:17*am, flipper wrote:
On Sun, 4 Oct 2009 17:44:03 -0700 (PDT), "
wrote:

On Oct 3, 10:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

50V caps have no place in tube equipment.

Peter Wieck
Melrose Park, PA

Rk bypass

Still and all (and to Don also) - I would not use a cap rated at
anything less than 150V in any location in any tube equipment - and
that includes battery-powered radios commonly using 20V caps in the
power-supply as an example - and the ones that most commonly fail as
it happens. The incremental cost is minimal and the advantages
significant. In the realm of tube audio - even more so.

Peter Wieck
Melrose Park, PA

[Don Pearce[_3_]]

On Mon, 5 Oct 2009 05:36:00 -0700 (PDT), "
wrote:

On Oct 5, 8:17*am, flipper wrote:
On Sun, 4 Oct 2009 17:44:03 -0700 (PDT), "
wrote:

On Oct 3, 10:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

50V caps have no place in tube equipment.

Peter Wieck
Melrose Park, PA

Rk bypass

Still and all (and to Don also) - I would not use a cap rated at
anything less than 150V in any location in any tube equipment - and
that includes battery-powered radios commonly using 20V caps in the
power-supply as an example - and the ones that most commonly fail as
it happens. The incremental cost is minimal and the advantages
significant. In the realm of tube audio - even more so.

Peter Wieck
Melrose Park, PA

Why is that? Do you imagine that you gain some extra safety or
performance thereby? I promise you don't.

d

[[email protected]]

On Oct 5, 8:38*am, (Don Pearce) wrote:
On Mon, 5 Oct 2009 05:36:00 -0700 (PDT), "
wrote:





On Oct 5, 8:17*am, flipper wrote:
On Sun, 4 Oct 2009 17:44:03 -0700 (PDT), "
wrote:

On Oct 3, 10:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

50V caps have no place in tube equipment.

Peter Wieck
Melrose Park, PA

Rk bypass

Still and all (and to Don also) - I would not use a cap rated at
anything less than 150V in any location in any tube equipment - and
that includes battery-powered radios commonly using 20V caps in the
power-supply as an example - and the ones that most commonly fail as
it happens. The incremental cost is minimal and the advantages
significant. In the realm of tube audio - even more so.

Peter Wieck
Melrose Park, PA

Why is that? Do you imagine that you gain some extra safety or
performance thereby? I promise you don't.

Performance? Not at all. To that end, a cap is a cap is a cap within
the most basic parameters.
Safety? Depends on how that is defined, but not really, assuming good-
quality caps in the first place. I do not expect caps to act as fuses.
Nor resistors.
Longevity? Reliability? Absolutely.

Been-there-done-that far too long and seen far too much in this hobby
to minimize parts in this sort of application.

Peter Wieck
Melrose Park, PA

[Patrick Turner]

On Oct 5, 11:36*pm, " wrote:
On Oct 5, 8:17*am, flipper wrote:

On Sun, 4 Oct 2009 17:44:03 -0700 (PDT), "
wrote:

On Oct 3, 10:49*pm, Fu Knee wrote:
Hi RATs!

So, my BlumGarts lasted over a year using 560uF 50V bypass caps.

50V caps have no place in tube equipment.

Peter Wieck
Melrose Park, PA

Rk bypass

Still and all (and to Don also) - I would not use a cap rated at
anything less than 150V in any location in any tube equipment - and
that includes battery-powered radios commonly using 20V caps in the
power-supply as an example - and the ones that most commonly fail as
it happens. The incremental cost is minimal and the advantages
significant. In the realm of tube audio - even more so.

With the plastic caps used around a tube amp circuit, sure, you would
not use 50V rated caps.

But there are plenty of areas where lower than 50V is good practice
because you don't want the rated voltage to be too much above the
working or you won't get the wanted polarisation. I would not use a
150V rated electro where the Ek was only 2Vdc.

And I don't like huge caps where there could be a perfectly adequate
little one.

The other important reason electro caps fail is that they get too hot
in a tube amp if there is inadequate ventilation under the chassis. Of
they fail in a PS because the bias has failed and ripple current plus
heat is too much for them. Many amps have a nice row of holes around
the output tubes but then have a bloomin PCB board which obstructs the
air flow vertically, so the sub-chassis volume gets quite hot,
especially in summer. I've sometimes seen the plastic covering over
electros begin to peel off in the heat, and you know that cap could
fail soon.
Its best to never use PCBs in tube amps, and only use terminal strips
so the PCB then does not blanket the real estate. Without the pcb,
electros will be strapped/bolted to the chassis sides and bedded in
silicone and they then run cooler.

Modern electros are very reliable when the temperature and ripple
current and working voltages are all within ratings.

Patrick Turner.





Peter Wieck
Melrose Park, PA

[Ian Bell[_2_]]

wrote:
On Oct 5, 8:17 am, flipper wrote:
On Sun, 4 Oct 2009 17:44:03 -0700 (PDT), "
wrote:

On Oct 3, 10:49 pm, Fu Knee wrote:
Hi RATs!
So, my BlumGarts lasted over a year using 560uF 50V bypass caps.
50V caps have no place in tube equipment.
Peter Wieck
Melrose Park, PA
Rk bypass

Still and all (and to Don also) - I would not use a cap rated at
anything less than 150V in any location in any tube equipment - and
that includes battery-powered radios commonly using 20V caps in the
power-supply as an example - and the ones that most commonly fail as
it happens. The incremental cost is minimal and the advantages
significant. In the realm of tube audio - even more so.

Peter Wieck
Melrose Park, PA


Except when you want 47,000uF.

Cheers

Ian

[Ian Iveson]

Peter wrote:

50V caps have no place in tube equipment.

I've used them for small valve cathode bypass. Less reliable
than a higher voltage, but adequate, and cheap and compact.
Temperature rating is also important for longevity,
presumably.

What's the difference, in terms of construction, between
electrolytics rated for different voltages?

Ian

[[email protected]]

On Oct 5, 5:31*pm, "Ian Iveson"
wrote:

What's the difference, in terms of construction, between
electrolytics rated for different voltages?

Primarily the robustness of the elements making it. Just as a chain
required to stretch ten feet but hold only a few grams may be of a
different sort of construction than a chain stretching the same
distance but required to hold several pounds. That also translates to
size, all other things being equal, and in applications where real-
estate is precious this could be a concern. However, it has been my
experience that tube-based equipment is inherently wasteful of real-
estate such that this is seldom at issue.

Peter Wieck
Melrose Park, PA

[Ian Iveson]

Peter wrote:

***Primarily the robustness of the elements making it. Just
as a chain
required to stretch ten feet but hold only a few grams may
be of a
different sort of construction than a chain stretching the
same
distance but required to hold several pounds. That also
translates to
size, all other things being equal, and in applications
where real-
estate is precious this could be a concern. However, it has
been my
experience that tube-based equipment is inherently wasteful
of real-
estate such that this is seldom at issue.***

Hmm. I asked because the answer isn't obvious, as it would
be for a film cap, where higher voltage ratings can use a
thicker dielectric layer.

With an electrolytic, once the oxide layer is complete, no
more can be deposited by electrolytic action because there
is no conduction. The thickness of dielectric between the
aluminium electrode and the electrolyte is presumably
therefore always pretty much the same.

I remember reading that higher capacitance for the same
linear area of electrode can be achieved by etching the
aluminium to give a rough matt finish, increasing the
surface area. Considering higher voltage rating caps are
bigger, I assume their electrodes must be smoother, possibly
polished.

I guess that a rougher finish leads to many points of charge
concentration ('point action'), making arcing more likely
for a given voltage.

Where your "robustness" may come in is that there will
always be the odd rough spot or defect that will get zapped
as the cap charges to its rated voltage, especially for the
first time or after a period of languor. The higher energy
of higher-voltage arcs may require a thicker foil to contain
the damage.

The upshot is that electrolytics get bigger with voltage
rating in similar proportion as do film caps, even though
the reasons are different.

I've thought of another question. What happens to the
insulating layer of oxide if the cap polarity is reversed?
Does it turn back into aluminium? How does that happen when,
initially at least, no DC current is flowing?

Ian

[[email protected]]

On Oct 6, 9:08*am, "Ian Iveson"
wrote:

The upshot is that electrolytics get bigger with voltage
rating in similar proportion as do film caps, even though
the reasons are different.

I've thought of another question. What happens to the
insulating layer of oxide if the cap polarity is reversed?
Does it turn back into aluminium? How does that happen when,
initially at least, no DC current is flowing?

I will distill the off-group reply.

"Ruggedness" applies to the thickness of the foil, pattern (if any) on
the foil, the precise chemical nature of the electrolyte, and the
tightness with which it is wound vs. the surface area of the cap.
Aluminum oxides are very, very stable. But in this application, if
reverse voltage is applied in sufficient quantity the caps will outgas
and explode. KABOOM. Try it one day - but stand back and wear goggles,
and point the top or bleed area of the cap away from yourself.

Kinda-sorta like applying reverse voltage on a diode - nothing much
happens until it does. And the next state is complete failure.

There is lots out there on capacitor construction and the various
forms of electrolytes, metals used and failure modes. A google search
will bring you more reading than you will ever care to do.

Peter Wieck
Melrose Park, PA

[Fu Knee]

Hi RATs!

Thank you for all the 'tro info. I knew I was pushing the envelope
operating the 50V units at 53V ... lasted over a year, of pretty much
24/7. Many thunderstorms passed over, without incident. Finally, one
of the four caps farted off into that Great Beyond. So, that tube
glowed red for the time it took me to get out of bed and shut it down.
I have the chassis sanding on its back, so the power cable hangs down,
and I can pull it out of the socket with minimal perceptual
processing. And I am rather too well informed on getting things done
with minimal ... sigh.

And, thank you for the tip on painting the town red There are lots
of Jennifers on this planet, every one is capable of amazing feats of
social interaction, you know, when the mood strikes ...

.... I am an old guy, 62, and not healthy, as I have had this bit of
some wretched bother for a few days, or weeks, well, OK, about twelve
years ... anyway, I cling to the sounds in my room like a huge turd
found floating in this sea of madness and sadness. You know, when I
can. Not as often as: "Once upon a time".

Think of me as a sea going dung beetle, sailing this piece of
excrement in search of the Holy B+ Rail.

We, my good, if even less healthy, wife and I, have retreated to this
preposterous huge suburban tract house to live out our lives, as the
event horizon comes rushing forward like a really low budget horror
movie scene. You will know what I mean, when you see it .. not before.

When I first saw the "Garter" circuit, I was struck by the simplicity
of that solution to get two tubes to draw the same current, at idle,
in a push-pull output stage. The two tubes are amplifying the same
signal (approximately 180 degrees out of phase). Balanced current
flow, while idle, allows perfect silence to be output to the speaker,
when there is no input signal. Then, when something happens, the tubes
take off in opposite directions from a better rest position, compared
to any DC offset while idle. And Eric Barbour has proved it works for
Parallel Push-Pull.

And, that idle current determines how well the output transformer can
pass signals. It ain't pretty, just the way it is.

It is glorious

I learned of this circuit from John Broskie's www.tubecad.com site. He
never actually built it, just did the 'simulation', or 'emulation',
and was impressed with the results. Of course, him being a real
engineer, he took the idea and ran off into the night designing
circuts that could do it even better That was about five years ago.
Recently, he went back and decided his little darlings were not
exactly as superior as he had imagined. So, he did some new
work ...more genius output

I got lucky and was listening to my eBay Heathkit A7-E a few years ago
and suddenly remembered the circuit and powered it down and soldered
it in. Wow!

It is so cool when an idea finds a home.

The Heathkit A7-E sold for $17.95 in the early fifties. It was the top
of the line, with potted transformes and a built in Phono preamp. I
bypassed the preamp and tied my CD box directly to the volume control.
I loved vinyl, but playing LPs from this bed is a bit too much
exercise. I really tried

----

This Just In: To Do: check out the new Nalu from Ocean Kayak - I don't
paddle, neither, but, I do dream of something nobler than clinging to
this POS life.
---

The 'Garter" circuit does work with no bypass caps, too, if you do not
need all that gain.

Happy Ears!
Al

[Bret L]

On Oct 5, 7:29*pm, " wrote:
On Oct 5, 5:31*pm, "Ian Iveson"
wrote:

What's the difference, in terms of construction, between
electrolytics rated for different voltages?

Primarily the robustness of the elements making it. Just as a chain
required to stretch ten feet but hold only a few grams may be of a
different sort of construction than a chain stretching the same
distance but required to hold several pounds. That also translates to
size, all other things being equal, and in applications where real-
estate is precious this could be a concern. However, it has been my
experience that tube-based equipment is inherently wasteful of real-
estate such that this is seldom at issue.


You do not want to run lytics at very much lower than rated value.
They de-form and then when the spike you were overbuilding against
does hit them, poof.

50 vwdc is okay for 6.3 vwdc heater regulated supplies, because much
more will take out the regulator, pass transistor and diodes as well
or before the caps. Crowbar protection is called for here. Or, is it?

There is a school of design called "strong link/weak link" or SGTG
which has existed since the steam era but reached its zenith in the
design of nuclear weapons surety systems. Basically, you want
something to break first in the chain at a location you can predict.
In nuke weapons that something should be something inaccessible
without destroying the weapon and the weapon's (presumably
unauthorized) tinkerer. In everyday applications that should be an
accessible, inexpensive, easily changed part rather than one that is
deeply embedded or expensive.

A good example is the small quill shaft used to drive the scavenge
blower-calling it a supercharger is like calling Howard Dumbbell,
Howard-on a Detroit Diesel engine. If the blower FODs out the shaft
will fail before destroying the blower.

It's better to blow up the regulator or pass transistor than the
power transformer, IOW.

It's OK to use a 450 wvdc lytic at 300 volts but at 150 you are
wasting space and money and positively harming the design at 50. If
you really need a big nominal to peak ratio use oil caps.

[Bret L]

DC current does flow, once past the 'diode like' bias voltage.
Hydrogen atoms are injected into the oxide reducing it and
'deinsulating' the electrode. The bigger problem is out gassing. They
explode under the pressure.



Oh boy do they. I've blown up a lot of old bad lytics on purpose for
fireworks. The big old juice can ones can really give a hell of a bang
especially if you epoxy over the vent hole.

[Bret L]

On Oct 7, 9:10*am, Fu Knee wrote:
Hi RATs!

Thank you for all the 'tro info. I knew I was pushing the envelope
operating the 50V units at 53V ... lasted over a year, of pretty much
24/7. Many thunderstorms passed over, without incident. Finally, one
of the four caps farted off into that Great Beyond. So, that tube
glowed red for the time it took me to get out of bed and shut it down.
I have the chassis sanding on its back, so the power cable hangs down,
and I can pull it out of the socket with minimal perceptual
processing. And I am rather too well informed on getting things done
with minimal ... sigh.

And, thank you for the tip on painting the town red There are lots
of Jennifers on this planet, every one is capable of amazing feats of
social interaction, you know, when the mood strikes ...

... I am an old guy, 62, and not healthy, as I have had this bit of
some wretched bother for a few days, or weeks, well, OK, about twelve
years ... anyway, I cling to the sounds in my room like a huge turd
found floating in this sea of madness and sadness. You know, when I
can. Not as often as: "Once upon a time".

Soon you will die.

So quit wasting time on stupid things like the Blumlein garter
circuit.

Blumlein was a genius, yes. But he designed that circuit at a
specific time and place for a specific job that no longer exists.
Using it today is stupid. Blumlein would tell you that yourself if he
were here now.

I repeat, using the Blumlein garter circuit today is stupid.

Why don't you use your remaining time on planet to do something
people will benefit from? Because the only thing that lives forever is
the fame of a dead man's deeds.

[Ian Iveson]

flipper wrote:

Peter wrote:

***Primarily the robustness of the elements making it.
Just
as a chain
required to stretch ten feet but hold only a few grams may
be of a
different sort of construction than a chain stretching the
same
distance but required to hold several pounds. That also
translates to
size, all other things being equal, and in applications
where real-
estate is precious this could be a concern. However, it
has
been my
experience that tube-based equipment is inherently
wasteful
of real-
estate such that this is seldom at issue.***

Hmm. I asked because the answer isn't obvious, as it would
be for a film cap, where higher voltage ratings can use a
thicker dielectric layer.

Interestingly enough, it's actually 'the same' in one
sense. That is,
the voltage withstand is related to the oxide thickness.


With an electrolytic, once the oxide layer is complete, no
more can be deposited by electrolytic action because there
is no conduction. The thickness of dielectric between the
aluminium electrode and the electrolyte is presumably
therefore always pretty much the same.

It increases with the voltage.

How? Anodising requires current but the oxide is an
insulator. Do you have any actual data to back up your
hypothesis? Is the oxide layer in a 600V cap 100 times the
thickness of one rated at 6V? How much thicker is it, and
why is that thickness adequate to explain the higher voltage
rating?

When I got anodising done, I knew better, I thought, than to
ask for a thicker coating. I would have got nothing but a
withering look in return.

I said "pretty much the same", as I guess thickness may
increase slightly, because electrolytic caps have some
leakage and some of that may be due to current through the
oxide layer AFAIK, but I don't actually know.

I remember reading that higher capacitance for the same
linear area of electrode can be achieved by etching the
aluminium to give a rough matt finish, increasing the
surface area. Considering higher voltage rating caps are
bigger, I assume their electrodes must be smoother,
possibly
polished.

I guess that a rougher finish leads to many points of
charge
concentration ('point action'), making arcing more likely
for a given voltage.

Electrolytics are (within limits) 'self healing'. Any
'uninsulated'
spot gets insulated by current flow and the electrolytic
reaction, as
in "forming."

Yes, of course. I think however that it isn't quite like in
"forming", as you say, because the electrolyte in the cap is
different from the one used for anodising.

Your comment about high voltage caps and 'smooth'
electrodes is half
right but for the wrong reason. It isn't because of
'arcing' but the
oxide thickness and method of increasing foil area by
etching. As
voltage increases the oxide thickness increases limiting
the
'fineness' of etch one can use. I.E. a 'thick' coating
'covers up' a
very fine etch pattern rendering it essentially a
'non-etch'. Rather
than simply 'live with it' you use a coarser pattern but
that means
more area (hence final volume) per unit of capacitance.

I wondered about using the word "arcing". I mean local
discharge through the insulating layer, whether a spark is
involved or not. Wanted to avoid the word "discharge"
because it could easily be misunderstood wrt a capacitor.

When you say "wrong", do you mean to imply that a rough
surface does *not* make local discharges more likely? How
not? How does your capacitor defeat point action? How does
your cap prevent whiskers being melted by the higher
currents resulting from higher voltage discharges? There is
room enough in the story for us both to be right.

What happens when a cap fails due to over-voltage? My
assumption has been that local discharges become energetic
enough to damage more than they heal. Once this happens the
rate of damage increases in proportion to the damage done,
and catastrophic failure ensues.

If, however, more voltage results in the production of a
thicker layer of oxide, I would expect the normal anodising
process to continue over the whole surface area. It is
common for anodisers to use a current-limited power supply
to achieve perhaps 12A per square foot, possible with around
12V. That would overheat a cap quite quickly. OTOH, if the
current is limited carefully to a few milliamps, why can't I
make a low voltage cap into a high one just by exposure to
high voltage?

There has to be enough film to oxidize as well so it's
also thicker.

By how much? The oxide layer is very thin, and of greater
volume than the aluminium it consumes, which therefore
wouldn't get much thinner it seems to me. And anyway, the
issue was how thick the aluminium is *in the cap*, not how
thick it was before it was anodised.

Where your "robustness" may come in is that there will
always be the odd rough spot or defect that will get
zapped
as the cap charges to its rated voltage, especially for
the
first time or after a period of languor. The higher energy
of higher-voltage arcs may require a thicker foil to
contain
the damage.

It self heals 'exposed' or 'weak' insulation.

Yes, we know that. The issue here is how it fails. In normal
operation, why does a high voltage cap need thicker
aluminium than a low voltage one? Self healing is common to
all.

If a thicker foil is used for higher voltages, then
does that mean that ESR is lower for caps of higher voltage
rating? Or is most of the ESR attributable to the
electrolyte layer?

The upshot is that electrolytics get bigger with voltage
rating in similar proportion as do film caps, even though
the reasons are different.

I've thought of another question. What happens to the
insulating layer of oxide if the cap polarity is reversed?
Does it turn back into aluminium? How does that happen
when,
initially at least, no DC current is flowing?

DC current does flow, once past the 'diode like' bias
voltage.

Which doesn't answer the question of how, considering the
oxide is an insulator, which means current can't flow. Can I
assume that the voltage difference is sufficient to begin to
reduce the oxide, and that the process of reduction leads to
current, and happens at a similar rate to the process of
oxidation? Anodising at my local shop takes hours at 12A per
square foot. Even if the process of de-anodising was just
as slow, the current would fry the cap even before it
resulted in a short.

Hydrogen atoms are injected into the oxide reducing it and
'deinsulating' the electrode. The bigger problem is out
gassing. They
explode under the pressure.

OK. I blew one up once but I didn't time it. What is the
gas? Steam or oxygen? The electrolyte is presumably chosen
so that hydrogen isn't released by the process of healing,
in which case hydrogen combined somehow in the electrolyte
would be available to mop up some oxygen if the process is
reversible, but not enough to contain all the oxygen
released by wholesale reduction of the oxide layer.

Overall, your story contains some truth, but you are
underplaying the importance of continuing processes that
consume resources, are damaging by nature, and eventually
result in failure. A high voltage cap needs more resources
because the processes of failure are hastened by high
voltage.

Finally, I think you are wrong to argue that voltage rating
is not a guide to reliability. I think higher voltage caps
are less likely to suffer from electrolyte failure, for
instance, because they contain more electrolyte.

In general, as Peter originally implied, they have more
resources to cope with the inevitable and continuous
processes of decay. The details aren't simple, but the more
I consider the "overall robustness" argument, the more sense
it makes.

Ian

[Fu Knee]

On Oct 7, 1:49�pm, Bret L wrote:

�Soon you will die.

�So quit wasting time on stupid things like the Blumlein garter
circuit.

�Blumlein was a genius, yes. But he designed that circuit at a
specific time and place for a specific job that no longer exists.
Using it today is stupid. Blumlein would tell you that yourself if he
were here now.

�I repeat, using the Blumlein garter circuit today is stupid.

�Why don't you use your remaining time on planet to do something
people will benefit from? Because the only thing that lives forever is
the fame of a dead man's deeds

Hi RATs!

Relax, Burp, ignorance and hostility exactly like yours is being
reborn every few seconds. It will never end.

You are a vulgar noisemaker.

I commend your parents for showing such a fine eye for detail when
they separated you from real **** when you were an infant.

Consider this a love note.

I do not understand people who spend timing attacking others for
having fun with their hobby.

The garters are a simple idea.

They perform a simple function, very well.

You provide no function whatsoever. Spewing stupidity is not a
function

I love trading insults on this forum.

Folks can read and realize their lives are much better than some.

Happy Ears!
Al




PS Immortality is a fool's quest

[Ian Iveson]

Bret L wrote:

You do not want to run lytics at very much lower than rated
value.
They de-form and then when the spike you were overbuilding
against
does hit them, poof.


***This might have been a very good point.

***But your argument doesn't make sense. If I use a HV-rated
electro in an application that *never* exceeds 50V, i.e. in
a situation where a 50V cap would be an acceptable
alternative, why is that a bad thing to do? You say it
wastes space, but you avoid the issue at stake, which is
whether the HV-rated cap would last longer in that context.

***If, in the situation you describe of occasional HV, I
used a 50V cap, then it would fail far more certainly than
the HV rated one. Your point leads to the conclusion that
the presence of occasional HV rules out the use of *any*
electrolytic cap.

50 vwdc is okay for 6.3 vwdc heater regulated supplies,
because much
more will take out the regulator, pass transistor and diodes
as well
or before the caps. Crowbar protection is called for here.
Or, is it?

***A fuse is better in series than in parallel, wouldn't you
say? A cap may blow to either a short or an open circuit.
One would take out the tranny, which could be very expensive
if the windings are all on one transformer, the other
wouldn't protect anything.

There is a school of design called "strong link/weak link"
or SGTG
which has existed since the steam era but reached its zenith
in the
design of nuclear weapons surety systems. Basically, you
want
something to break first in the chain at a location you can
predict.
In nuke weapons that something should be something
inaccessible
without destroying the weapon and the weapon's (presumably
unauthorized) tinkerer. In everyday applications that
should be an
accessible, inexpensive, easily changed part rather than one
that is
deeply embedded or expensive.

A good example is the small quill shaft used to drive the
scavenge
blower-calling it a supercharger is like calling Howard
Dumbbell,
Howard-on a Detroit Diesel engine. If the blower FODs out
the shaft
will fail before destroying the blower.

It's better to blow up the regulator or pass transistor
than the
power transformer, IOW.

***How does a cap with a low voltage rating achieve that?

It's OK to use a 450 wvdc lytic at 300 volts but at 150
you are
wasting space and money and positively harming the design at
50. If
you really need a big nominal to peak ratio use oil caps.

***That's what I do, but all the same I think higher voltage
rated caps might generally be more reliable. It's just I
don't think the extra reliability is worth the downsides.

Ian

[Bret L]

***That's what I do, but all the same I think higher voltage
rated caps might generally be more reliable. It's just I
don't think the extra reliability is worth the downsides.

They are but only to a point. That point is roughly twice rated
voltage.

If you want more reliability you need a higher rated cap. Lytics come
in several ratings and cost levels, but none of them are the equal of
oil filled types by nature. Oil caps are the best for any considerable
voltage but the size and weight for needed values get unwieldy
quickly.

Marantz used oil caps on its first amplifiers. Those oil caps usually
still work after 50+ years. Later Marantz amps used a "telco grade"
lytic. Those hold up MUCH better than the commodisumo lytics in
Fenders and McIntoshes but still some have failed. Any Marantz 8B
(besides reissues) is at least 42-43 and most were made before '63.
They don't make this grade of lytic anymore as far as I am able to
determine.

[Fu Knee]

On Oct 8, 1:06�am, flipper wrote:

Far be it for me to intrude upon your myths. After all, the folks
making a living in reliability and life analysis probably don't have
the 'in depth' understanding of 'resources' you do.
Ian




Hi RATs!

We are meerly typing to pass some time.

Nobody is learning, nor preaching, Great Truths.

Not even me!

Bret is a bright boy. And, like all precocious children, is full of
himself.

Could we type about tube audio, and try to minimize pointing out each
other's foibles?

I am an old man. I slap errant children. And, like that drag queen
"mother" on Monty Python, sometimes - just for the FUN of it.

I am not perfect.

Perfection is a whole other newsgroup ...

People type nonsense. Not some people, all ... well, except Our Holy
Father Andre.

Happy Ears!
Al

[Arny Krueger]

"Ian Iveson" wrote in
message news
Peter wrote:

50V caps have no place in tube equipment.

I've used them for small valve cathode bypass. Less
reliable than a higher voltage, but adequate, and cheap
and compact. Temperature rating is also important for
longevity, presumably.

What's the difference, in terms of construction, between
electrolytics rated for different voltages?

http://archive.chipcenter.com/eexper...kruger006.html

"Dielectric Thickness.

"Several metals, such as tantalum, aluminum, niobium, zirconium and zinc,
can be coated with an oxide film by electrochemical means. For example, by
placing aluminum in appropriate solution and passing a current though the
circuit, a thin layer of aluminum oxide (Al2O3) forms on the aluminum anode.
It is this thin oxide film that is the dielectric in an electrolytic
capacitor, and the electrolytic process used for creating the dielectric
gives the name "electrolytic capacitor". The thickness of the oxide layer
depends on the formation voltage, which is typically 3-4 times higher than
the rated voltage, but it is very thin: less than a micrometer.
"

Note:

"The thickness of the oxide layer depends on the formation voltage, which is
typically 3-4 times higher than the rated voltage, but it is very thin: less
than a micrometer."

Anodizing aluminum for 450 volt electrolytics would appear to be something
that takes a lot of care due to the safety issues.