Selenium rectifiers
Unrevealed Source
rectifiers that are NOS, and I'm sondering if it's the use (current, heat,
etc.) that breaks them down, or is it also just age? I've been using the
standard diode+resistor, but wouldn't mind using NOS selenium rectifiers, if
they'll last the 40-50 years that the originals did.
Jeff
TwoMuttHeads
wrote:
My understanding is that both age and use contribute to their failure
rate. With modern silicon diodes so cheap, why waste the time to put
in units that are known to fail? Many of those selenium units failed
within a few years. I think it's highly unlikely that the silicon
units will ever fail when used in old tube radios. Or at least their
failure rate will be far less than any selenium units.
philo
I used to work on selenium diode devices and they don't usually fail
outright...their internal resistance simply keeps going up as they age
with use.
Kenneth Scharf
on metal plates that are then stacked. Eventually, the joints fail
which can lead to the high resistance, or a short circuit. Also as the
material ages, it's current carrying ability decreases, which will
eventually lead to overload and the usual 'smell'. If stored in a
humidity and temperature controlled environment the shelf life of unused
units should be reasonable, if they were stored in a garage all bets are
off. Leave the original in place out of the circuit for 'looks' and
replace it with a si diode.
Unrevealed Source
"Kenneth Scharf" <wa2...@bellsouth.net> wrote in message
news:hg321n$ef6$1...@news.eternal-september.org...
>>
>> I used to work on selenium diode devices and they don't usually fail
>> outright...their internal resistance simply keeps going up as they age
>> with use.
> That's one way they fail. The rectifiers consist of selenium deposited on
> metal plates that are then stacked. Eventually, the joints fail which can
> lead to the high resistance, or a short circuit. Also as the material
> ages, it's current carrying ability decreases, which will eventually lead
> to overload and the usual 'smell'. If stored in a humidity and
> temperature controlled environment the shelf life of unused units should
> be reasonable, if they were stored in a garage all bets are off. Leave
> the original in place out of the circuit for 'looks' and replace it with a
> si diode.
Maybe I should have clarified why I would even asked the question.
Obviously modern diodes are going to be more reliable and long-lasting.
But when I find myself restoring a nice little radio that I'm going to keep,
I want it to be reliable and safe but I also know that I'm not going to put
many hours on it. Over the next 40 years, it may get 10 hours of play.
My motivation would be simply for ease of installation - a direct swap,
instead of arranging the diode and big resistor under the chassis, maybe
mounting a terminal strip, all that. I'm wondering if a NOS selenium
rectifier might be "good enough" for this type of situation. Simple,
quick, easy. Of course anything that I intended for real daily use would
get a diode replacement.
But if they've been degrading over the years just sitting there unused, then
I wouldn't go that route. Just like we wouldn't use NOS paper/wax
capacitors - just sitting there drying out makes them ineligible.
jughead
wrote:
> I want it to be reliable and safe but I also know that I'm not going to put
> many hours on it. Over the next 40 years, it may get 10 hours of play.
>
Even then, I would definitely do the silicon diode+resistor, leaving
the original selenium in place and out of circuit, using one lug as a
tie point. And a notation of the mod inside.
Who knows, the radio may fall into the hands of someone else in the
course of 40 years.
The mod would actually be historical to the radio's life, reflecting
the evolution of technology and its adaptation to improving an
original product. Bill(oc)
Unrevealed Source
serviced. I guess I'm convinced.
I wouldn't leave the original in place "for authenticity". There are going
to be lots of non-authentic orange capacitors all over the place, and the
electrolytics mounted under the chassis (with THAT old can left in place,
obviously). But the mounting hole for the old rectifier is usually where I
mount the terminal strip.
"jughead" <oldco...@gmail.com> wrote in message
news:a49abe3d-24c7-48ef...@r24g2000prf.googlegroups.com...
jughead
which adds to the heating caused by increased forward resistance.
Unrevealed Source
> Old seleniums are also prone to developing excessive reverse leakage,
> which adds to the heating caused by increased forward resistance.
>
I'm aware of the breakdown of old selenium rectifiers, but even though I may
have decided not to use NOS ones, I still don't think we know the answer as
to whether they fail due to actual use (current flowing through them causing
heat, etc.) or whether it occurs with age, even with non-use. Is a
forty-year old NOS selenium rectifier, never ever used, just as good as the
day it was made, and will it last just as long? Or has it "aged"?
Bill M
IMHO it has 'aged'.
-Bill
D. Peter Maus
From 'Reference Data for Radio Engineers.'
"While selenium rectifiers were lighter in weight and used less
power than comparable vacuum tube rectifiers, they had the
disadvantage of finite life expectancy, increasing resistance with
age, and were only suitable to use at low frequencies. Both selenium
and copper oxide rectifiers have somewhat better tolerance of
momentary voltage transients than silicon rectifiers.
"Typically these rectifiers were made up of stacks of metal plates
or washers, held together by a central bolt, with the number of
stacks determined by voltage; each cell was rated for about 20
volts. An automotive battery charger rectifier might have only one
cell: the high-voltage power supply for a vacuum tube might have
dozens of stacked plates. Current density in an air-cooled selenium
stack was about 600 mA per square inch of active area (about 90 mA
per square centimeter)."
Suggesting that age at rest plays a factor in the finite life of
a selenium rectifier. As does use and stress in service.
Selenium is a non metal, and chemically, quite active, even in
oxide forms. In the environment, especially where there is moisture,
and considering that the discs themselves are not sealed, but
exposed to the environment, age alone would contribute to the
failure rate.
Kenneth Scharf
rectifiers were still being sold for use in new equipment. I remember
building a small 1 tube radio as a kid, and it used a small selenium
rectifier (why didn't they call them diodes?) even though a silicon
diode might have actually been cheaper and easier to use. I also
remember that there were large selenium rectifiers made for TV use, that
had a voltage rating of 300 volts. They were used in a voltage doubling
power supply for transfomerless tv sets.
Unrevealed Source
"D. Peter Maus" <dpete...@worldnet.att.net> wrote in message
news:hg4mma$3ot$1...@news.eternal-september.org...
>
> From 'Reference Data for Radio Engineers.'
>
> "While selenium rectifiers were lighter in weight and used less power
> than comparable vacuum tube rectifiers, they had the disadvantage of
> finite life expectancy, increasing resistance with age, and were only
> suitable to use at low frequencies. Both selenium and copper oxide
> rectifiers have somewhat better tolerance of momentary voltage transients
> than silicon rectifiers.
>
> "Typically these rectifiers were made up of stacks of metal plates or
> washers, held together by a central bolt, with the number of stacks
> determined by voltage; each cell was rated for about 20 volts. An
> automotive battery charger rectifier might have only one cell: the
> high-voltage power supply for a vacuum tube might have dozens of stacked
> plates. Current density in an air-cooled selenium stack was about 600 mA
> per square inch of active area (about 90 mA per square centimeter)."
>
> Suggesting that age at rest plays a factor in the finite life of a
> selenium rectifier. As does use and stress in service.
>
> Selenium is a non metal, and chemically, quite active, even in oxide
> forms. In the environment, especially where there is moisture, and
> considering that the discs themselves are not sealed, but exposed to the
> environment, age alone would contribute to the failure rate.
>
And that does seem to address the question quite adequately. Thanks!
Jeff
Phil Nelson
rectifier in my 1961 color TV. The whole mess fell into pieces when I
unsoldered the legs.
I have a tester in which someone had installed a newer selenium rectifier.
Rattling around the bottom of the case were the pieces of the old one. I
installed a silicon diode and threw all of the selenium parts into the
garbage where they belong :-)
Phil
philo
Though I'm sure it may have aged a bit
chances are it's still good..
It's up to you of course
but for authenticity I think I'd use it (as long as it's not peeling)