On Wed, 07 Mar 2007 15:44:47 +0000, Joe wrote:

I don't think 12000 hours is bad for an electrolytic. The old rule
of thumb was seven years, though of course many go far beyond that.
A bit like hard drives.

I've got an AR88D with some electrolytics coming up for their 70th
rather than 7th birthday :-).

--
Peter Parry.
http://www.wpp.ltd.uk/

On 7 Mar, 15:44, Joe wrote:

I don't think 12000 hours is bad for an electrolytic. The old rule
of thumb was seven years, though of course many go far beyond that.
A bit like hard drives.

Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.


NT

In message .com,
writes
Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.

The failure of electrolytics over time is due to the drying out of the
electrolyte. The capacitor will generally read the correct value on a
capacitance meter but will develop a high ESR (Equivalent series
Resistance) which causes ripple on a supposedly smooth DC supply. (And
timing problems in old circuitry.)

When your old TV starts to hum and there is a dark ripple across the
screen or the picture tears away at the corner.... That's a dying
electrolytic.


I missed the start of this thread. Is the subwoofer humming even when
the audio source is disconnected?

--
Clive Mitchell
http://www.bigclive.com

wrote:
On 7 Mar, 15:44, Joe wrote:

I don't think 12000 hours is bad for an electrolytic. The old rule
of thumb was seven years, though of course many go far beyond that.
A bit like hard drives.

Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.


Well I had to fix an old valve radio that was late 50's vintage..that
had packed in its electrolytics after about 40 years.


Lost of valve stuff has that as the single and only real fault.


NT

Clive Mitchell wrote:
In message .com,
writes
Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.

The failure of electrolytics over time is due to the drying out of the
electrolyte. The capacitor will generally read the correct value on a
capacitance meter but will develop a high ESR (Equivalent series
Resistance) which causes ripple on a supposedly smooth DC supply. (And
timing problems in old circuitry.)

When your old TV starts to hum and there is a dark ripple across the
screen or the picture tears away at the corner.... That's a dying
electrolytic.


I missed the start of this thread. Is the subwoofer humming even when
the audio source is disconnected?


well it hums only when the *mains* IS connected..and on..

In message , The Natural
Philosopher writes
well it hums only when the *mains* IS connected..and on..

If it's just the mains and no signal then it sounds like either the PSU
capacitor as suggested or maybe just sloppy circuitry picking up hum.
Maybe even bad placement of cables inside if it's an active unit.

--
Clive Mitchell
http://www.bigclive.com

On Thu, 08 Mar 2007 00:54:51 +0000, The Natural Philosopher
wrote:

wrote:
On 7 Mar, 15:44, Joe wrote:

I don't think 12000 hours is bad for an electrolytic. The old rule
of thumb was seven years, though of course many go far beyond that.
A bit like hard drives.

Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.


Well I had to fix an old valve radio that was late 50's vintage..that
had packed in its electrolytics after about 40 years.


Lost of valve stuff has that as the single and only real fault.

Exactly. When I was a teenager (can't remember when that was, but a
while ago), I'd fix old valve radios for relatives, friends,
neighbours and assorted hangers on.

The most common fault by far was a failed smoothing capacitor leading
to a hum. I had a box of assorted diameter ones that would fit the
clip on the chassis. It didn't particularly matter if they were
larger capacitance than the original.

The second was that they had left the radio tuned to one station for
donkeys years (normally the Light Programme or Home Service) and now
wanted to tune to something else. Of course, the tuning capacitor
would be crudded up and would cause crackles or have intermittent
contact requiring "mechanical treatment" of the set.
A quick go with the vacuum cleaner and a squirt of contact cleaner
usually fixed these.

The third, but fortunately extremely rare one was a failed selenium
rectifier. When these are on their way out, they emit the most evil
of smells known to man and having a strong emetic effect.

Any of these repairs would bring forth a nice afternoon tea plus an
amount of folding that was way in excess of the time and materials
cost of the repair.

Regarding electrolytics in these old pieces of equipment, the failure
mode was pretty obvious - I took several apart for curiosity. Inside
they would be packed with some kind of gunk soaking the dielectric. On
working ones this would be quite gooey. On failed ones it had
hardened and become cracked. The hum wouldn't suddenly appear but
would get worse over the years until it was noticably bad in
comparison with something else like the TV.


--

..andy

On 7 Mar, 23:39, Clive Mitchell wrote:
In message .com,
writes

Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.

The failure of electrolytics over time is due to the drying out of the
electrolyte. The capacitor will generally read the correct value on a
capacitance meter but will develop a high ESR (Equivalent series
Resistance) which causes ripple on a supposedly smooth DC supply. (And
timing problems in old circuitry.)

It seems no-one can backup this 7 yr rule so far. Its far removed from
my experience with lytics, and lord knows I've worked with enough of
them. IME around 4 out of 5 1930s ones have still worked, so thats a
mttf of ballpark 70 x 4 = 280 yrs.

I'm surprised by Andy's experience, I've also fixed lots of valve kit,
and lytic failures have happpened but been fairly rare.

If there had been any sort of 7 yr rule, a lot of the new kit I've
worked on would never have got out the door, as 7 yr life would have
been unacceptably short.


NT

On 8 Mar 2007 02:00:42 -0800, wrote:

On 7 Mar, 23:39, Clive Mitchell wrote:
In message .com,
writes

Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.

I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.

The failure of electrolytics over time is due to the drying out of the
electrolyte. The capacitor will generally read the correct value on a
capacitance meter but will develop a high ESR (Equivalent series
Resistance) which causes ripple on a supposedly smooth DC supply. (And
timing problems in old circuitry.)

It seems no-one can backup this 7 yr rule so far. Its far removed from
my experience with lytics, and lord knows I've worked with enough of
them. IME around 4 out of 5 1930s ones have still worked, so thats a
mttf of ballpark 70 x 4 = 280 yrs.

I'm surprised by Andy's experience, I've also fixed lots of valve kit,
and lytic failures have happpened but been fairly rare.

If there had been any sort of 7 yr rule, a lot of the new kit I've
worked on would never have got out the door, as 7 yr life would have
been unacceptably short.


NT

Bear in mind that the lifetime of electrolytics will follow some kind
of normal distribution curve, and the quoted seven years will
represent the very start of the lower end of the tail - probably 0.1%
of the entire distribution, maybe even less. So the average life is
probably more like twenty or thirty years, with the upper tail
extending possibly to a hundred or so.

d

--
Pearce Consulting
http://www.pearce.uk.com

Don Pearce wrote:
On 8 Mar 2007 02:00:42 -0800, wrote:

On 7 Mar, 23:39, Clive Mitchell wrote:
In message .com,
writes
Where do you get the 7 year rule from, and what exactly is it? I've
never come across it, and if it existed I really ought to know.
I've still got 1930s lytics in service, and IME lytics are a long way
down the list of most likely failures in old equipment. The ida that
lytics have short lives seems to be an offshoot of the spate of bad
caps on mobos in the 90s.
The failure of electrolytics over time is due to the drying out of the
electrolyte. The capacitor will generally read the correct value on a
capacitance meter but will develop a high ESR (Equivalent series
Resistance) which causes ripple on a supposedly smooth DC supply. (And
timing problems in old circuitry.)
It seems no-one can backup this 7 yr rule so far. Its far removed from
my experience with lytics, and lord knows I've worked with enough of
them. IME around 4 out of 5 1930s ones have still worked, so thats a
mttf of ballpark 70 x 4 = 280 yrs.

I'm surprised by Andy's experience, I've also fixed lots of valve kit,
and lytic failures have happpened but been fairly rare.

If there had been any sort of 7 yr rule, a lot of the new kit I've
worked on would never have got out the door, as 7 yr life would have
been unacceptably short.


NT

Bear in mind that the lifetime of electrolytics will follow some kind
of normal distribution curve, and the quoted seven years will
represent the very start of the lower end of the tail - probably 0.1%
of the entire distribution, maybe even less. So the average life is
probably more like twenty or thirty years, with the upper tail
extending possibly to a hundred or so.

d

Certainly when called upon to fix the valve radio, I did a lot of web
research, and it was, with selenium rectifiers, the largest single problem.

Ageing coils needing tweaking and ageing valves were most of the others.

resistors rarely go, and in modern kit semiconductors are nearly as good.

Dry joints in PCB's are the final big failure mode.