Film capacitor as power-supply filter

On Friday, 11 October 2019 01:40:22 UTC+1, John Larkin  wrote:
> On Thu, 10 Oct 2019 17:00:16 -0700 (PDT), Michael Terrell
> <terrell.michael.a@gmail.com> wrote:
> >On Thursday, October 10, 2019 at 7:40:27 PM UTC-4, John Larkin wrote:
> >> On Thu, 10 Oct 2019 16:36:35 -0700 (PDT), whit3rd wrote:
> >> >On Thursday, October 10, 2019 at 7:50:51 AM UTC-7, John Larkin wrote:

> >I had a Sprague TO6 capacitor analyzer that could reform to 500VDC, and it metered the leakage current. If the oxide layer was bad, you could slowly ramp it up until it had acceptable leakage at the rated voltage. The Military had a set schedule to test and reform NOS electrolytics that were in warehouses. I used to see a lot of surplus caps with a label showing the dates.
> 
> Are modern electrolytics any better than the old ones, from the aspect
> of un-forming themselves? 
> 
> Or maybe the old ones are, well, just old.

The oldest still working lytics I have are 1930s. I doubt any modern lytic will last that long. OTOH olduns were quite leaky.


NT

On Thursday, October 10, 2019 at 8:40:22 PM UTC-4, John Larkin wrote:
> On Thu, 10 Oct 2019 17:00:16 -0700 (PDT), Michael Terrell wrote:
> 
> >On Thursday, October 10, 2019 at 7:40:27 PM UTC-4, John Larkin wrote:
> >> 
> >> I meant an OLD instrument.
> >
> >Back in the '60 and '70s we would apply 12 to 24 volts to old equipment to warm the power transformer to drive out moisture. The tubes were pulled, first and it was left on this setup for 24 hours. It seemed like every old piece of test equipment was store in a dank basement.
> >
> >I had a Sprague TO6 capacitor analyzer that could reform to 500VDC, and it metered the leakage current. If the oxide layer was bad, you could slowly ramp it up until it had acceptable leakage at the rated voltage. The Military had a set schedule to test and reform NOS electrolytics that were in warehouses. I used to see a lot of surplus caps with a label showing the dates.
> 
> Are modern electrolytics any better than the old ones, from the 
> aspect of un-forming themselves? 
> 
> Or maybe the old ones are, well, just old.

I have some NOS Sprague that I will try to dig out. They are Axial, and over 20 years old. They are still in sealed plastic bags of 100. I will do some tests on them, when I can dig them out. I have a Sencore LC53 Digital LC Meter to use for the testing.

As far as time differences in how long the formed surface lasts, most of the newer types haven't been around long enough for a real comparison. How can you compare recent parts to ones that are approaching 100 years old?

The older Sprague Capacitors had better seals. Their aluminum cans were sealed with hot tar. They held so well that some I threw into the trash were incinerated. The formed can was almost spherical, before they blew out the seal. This was 50+ years ago, and some were 20 years old at that time..

Sprague was the best, with Cornell Dubilier second and Mallory third, as far as being well sealed. Other brands failed more often, especially unbranded ones used by low end OEMs of consumer electronics. 

Sprague was always our first choice for replacements. The suppliers were always pushing lower grades, because they made more money on them.

The highest failure rates for Electrolytics in TVs in the '60s and '70s were the 160uF@250VDC High Ripple Current rated parts desiged for the input capacitors used in the voltage doubler in transformerless chassis. People bitch about the doubler concept, yet it is still used in computer power supplies.

The Electrolytes have improved, which allow the same ratings in smaller packages but those can't dissipate as much heat as the larger cans did. 

Most small Electrolytics these days appear to only have a rubber plug that is crimped into the can. That's why they end up crooked when they overhead and push part or all of the seal out of the can. I had one HP desktop on the bench when one failed in the power supply. The can shot out through the running fan, followed by foot long flames.

A company I worked for near Orlando 30 years ago was hired to do asset recovery when Sprague closed their Orlando facility. Their manager told us that very few customers were buying quality Electrolytics anymore. All they cared about was a really price, and just enough quality to make it through their warranty to their end customers.

I'm certain that the Military did enough testing to justify their scheduled reforming. It gave them more than enough of a part number to outlast whatever they were made for.

This is just a guess, but I'd guess that two things have happened to the making of electrolytic capacitors in the past 50 years, and they have opposite effects:

(1) Improvement.  Higher specs, longer lifetime at high temperature, etc.  This should make newer capacitors last longer than old ones.

(2) Cheapening.  With mass demand, there's a desire to make capacitors as cheaply as possible.  Not everybody will pay for 105 C temperature rating or name-brand reliability.  Look at the unrecognizable brand names coming out of Asia.

So... they've gotten better... except some of them... maybe?

Perhaps one economic factor is that, as the cost of other components has fallen, people expect the electrolytic capacitors to be very cheap also.  The capacitors now stand out as the expensive components.  And that's an incentive to lower standards.

On Friday, October 11, 2019 at 9:49:58 AM UTC-4, m...@uga.edu wrote:
> Perhaps one economic factor is that, as the cost of other components has fallen, people expect the electrolytic capacitors to be very cheap also.  The capacitors now stand out as the expensive components.  And that's an incentive to lower standards.

Some of them are approaching zero value. Older equipment was expected to last 10 years or more. A lot of today's junk is lucky to last a year or two. In that respect, the costs are going up, not down. Lower grade caps are a very high failure item. I had one C-band receiver on the bench a few years ago. It had about 85 electrolytics, and every one was bad Most had no brand name, and some weren't marked for temperature. It belonged to a bar, and they needed it right away, so they paid a large bill to have it working the next day. I generally use Panasonic and Rubicon for repairs and my projects, these days.

Michael Terrell <terrell.michael.a@gmail.com> wrote:
> On Friday, October 11, 2019 at 9:49:58 AM UTC-4, m...@uga.edu wrote:
>> Perhaps one economic factor is that, as the cost of other components has
>> fallen, people expect the electrolytic capacitors to be very cheap also.
>> The capacitors now stand out as the expensive components.  And that's an
>> incentive to lower standards.
>
> Some of them are approaching zero value. Older equipment was expected to last
> 10 years or more. A lot of today's junk is lucky to last a year or two. In that
> respect, the costs are going up, not down. Lower grade caps are a very high
> failure item. I had one C-band receiver on the bench a few years ago. It had
> about 85 electrolytics, and every one was bad Most had no brand name, and some
> weren't marked for temperature. It belonged to a bar, and they needed it right
> away, so they paid a large bill to have it working the next day. I generally use
> Panasonic and Rubicon for repairs and my projects, these days.

Near the turn of the century a Missouri man named Chris started 
badcaps.net Chris' "master list" enumerates the capacitors that he
trusts:

https://www.badcaps.net/index.php?pageid=master_list

Chris also hosts a lively forum at https://www.badcaps.net/forum/
Although its more apropos to repair.

Thank you, 73,

-- 
Don Kuenz KB7RPU
There was a young lady named Bright Whose speed was far faster than light;
She set out one day In a relative way And returned on the previous night.

On Monday, October 7, 2019 at 8:13:35 PM UTC-4, Phil Allison wrote:
> bloggs.fre...@gmail.com wrote:
> 
> -------------------------------
> 
> > 
> > Look at the ripple voltage and verify that your film can take the
> > ripple current.
> 
> ** Absurd advice. 
> 
> Then there is the turn-on time of the circuit. Does this Heath Kit have a lot of tubes in it, or what kind of electronic does it have. The old tube circuit DC supplies would overshoot at turn-on until the tube filaments warmed up and the tubes started conducting, drawing down the rail voltage.
> 
> ** Funny how rectifier tubes have long warm up times too. 
> 
> > Until then, the applied DC voltage rail would be way high, like 150% nominal. 
> 
> 
> ** Whata pile of fucking crap !!!
> 
> Even with silicon diodes in the PSU, the voltage excess is never more than 15% - something electros ARE rated for FFS. 
> 
> 
> And this could last for tens of seconds.
> 
> ** No it wouldn't. 
> 
> > The electrolytics are designed to handle this temporary overvoltage, usually spec'd for 30 seconds, the film definitely are not.
> 
> ** More complete bullshit .
> 
> Film caps have plenty of overvoltage capacity, most do not break down until double or triple DC voltage is applied.
> 
> Go away you alarmist IDIOT !!! 
> 
> 
> ....  Phil

You go away, bitter old fool. You're just miffed that your usual trivialization of the effect once again proves your inexperience. The effects I described have been observed in several different types of tube circuits, and are accurate. Most of the people here are old enough to remember waiting the better part of a minute for the old tube radios, TVs and whatnots to begin to come on. WTH do you think it was doing? A self-test? Not likely. And a lot of the old power tube circuits used selenium rectifiers, although I observed the same effect in vaccuum tube rectifier circuits too. Voltage and temperature (= ripple current losses) are the two primary stresses on film caps leading to failure.

On Friday, October 11, 2019 at 12:13:10 PM UTC-4, Don Kuenz wrote:
> Michael Terrell wrote:
> > On Friday, October 11, 2019 at 9:49:58 AM UTC-4, m...@uga.edu wrote:
> >> Perhaps one economic factor is that, as the cost of other components has
> >> fallen, people expect the electrolytic capacitors to be very cheap also.
> >> The capacitors now stand out as the expensive components.  And that's an
> >> incentive to lower standards.
> >
> > Some of them are approaching zero value. Older equipment was expected to last
> > 10 years or more. A lot of today's junk is lucky to last a year or two. In that
> > respect, the costs are going up, not down. Lower grade caps are a very high
> > failure item. I had one C-band receiver on the bench a few years ago. It had
> > about 85 electrolytics, and every one was bad Most had no brand name, and some
> > weren't marked for temperature. It belonged to a bar, and they needed it right
> > away, so they paid a large bill to have it working the next day. I generally use
> > Panasonic and Rubicon for repairs and my projects, these days.
> 
> Near the turn of the century a Missouri man named Chris started 
> badcaps.net Chris' "master list" enumerates the capacitors that he
> trusts:
> 
> https://www.badcaps.net/index.php?pageid=master_list
> 
> Chris also hosts a lively forum at https://www.badcaps.net/forum/
> Although its more apropos to repair.

I visited that site several times in its early days, when most people didn't believe that the industry had been duped with substandard parts. I wanted to see what brands were causing problems, and which ones were being faked. I left manufacturing in 2001.

On 2019-10-10 19:36, whit3rd wrote:
> On Thursday, October 10, 2019 at 7:50:51 AM UTC-7,
> jla...@highlandsniptechnology.com wrote:
> 
>> I've heard of people using a variac to slowly bring up an old 
>> instrument.
> 
> A switchmode regulated supply, though, takes maximum current input 
> when voltage is low-to-borderline.   So that mainly is good for 
> unregulated power (or power-wasting series or shunt regulated). It's
> also questionable if there's a DC fan, or ferroresonant circuit...
> 

Well, the variac trick is for old-old stuff--linear power supplies,
maybe tube rectifiers....

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

The rest of the story...

For those who tuned in late, I'm replacing the power supply filters in a Heathkit IT-11 capacitor checker.  In the original circuit they are two 40-uF 350-V in series (to handle 600 V).  I ordered a 20-uF 1-kV film capacitor to replace them.

Well, there just isn't room for that film capacitor under this chassis!  So I had to resort to two 47-uF 450-V in series, like Heathkit's original.

There was debate in this thread about whether to add balancing resistors, and I didn't.  John Larkin said "they take care of themselves."  Indeed they do, sharing the voltage to within a few percent of equal.  

I wonder if Mr. Larkin could explain further.  Is there some effect that makes them equalize if they have different leakages to start with?  I would expect the opposite, that the leakier one would carry a lower voltage and would form less, causing the difference in leakage to increase -- but I might be wrong.

I am going to look back at the bleeder network and see if there's a good way to split it in half.