Maker Pro
Maker Pro

Tantalum cap failures

Hello John

Last year, we conducted testing on a virgin batch of these caps,
charging them to 60V through a 10k resistor. Around 2% of them went
spuriously short circuit, then recovered and continued charging, some
repeating this cycle several times before charging fully to 60V, after
which they would not repeat the failure during numerous re-tests.

We spoke to Arcotronic and they now carry out specific screening along
similar lines, but the problem has not gone away, nor has it noticeably
reduced.

The failures frequently occur at or near the end of extensive testing,
including temperature cycling and vibration, and my concern has been
that some aspect of our application causes the failures - hence my
original post.

However, there is clearly a reliability issue with this component, and
I am coming round to the view that it is the component and not what we
do with it that is the problem. Unfortunately however, our choices are
rather restricted when it comes to finding a suitable alternative, but
I shall take a close look at the various suggestions for a replacement.

Ted
 
Hello Mark

There may be some merit in what you suggest, but this would require
substantial redesign and, even assuming we could fit this in the
available space, would have profound implications for project time
scales and costs. (This is a military product and as such has to
undergo extensive Qualification testing that can take many months to
complete succesfully, with associated costs that can run into six
figures).

The optimum solution would be for us to find a replacement component
that we can fit in place of the SFEs, in which case we can put forward
an argument for Qualification by analogy with the existing design.
Only if all else failed and we could justify the additional costs could
we pursue the redesign route.

Ted
 
S

Spehro Pefhany

Jan 1, 1970
0
On 11 May 2005 02:28:30 -0700, the renowned
Hello Mike

I've had a look at solid polimers and, at first glance, they look very
promising. However, their expected life is only 10 years or so, and we
have a contractual requirement for not less than 25 years. (How
realistic that really is is another matter, but that is what we are
obliged to aim for).

Ted

If the typical operating temperature is not too extreme, highly
derated high-temperature aluminum caps may meet your life
specification.


Best regards,
Spehro Pefhany
 
P

Pooh Bear

Jan 1, 1970
0
Hello John

Last year, we conducted testing on a virgin batch of these caps,
charging them to 60V through a 10k resistor. Around 2% of them went
spuriously short circuit, then recovered and continued charging, some
repeating this cycle several times before charging fully to 60V, after
which they would not repeat the failure during numerous re-tests.

That sounds truly awful.

We spoke to Arcotronic and they now carry out specific screening along
similar lines, but the problem has not gone away, nor has it noticeably
reduced.

The failures frequently occur at or near the end of extensive testing,
including temperature cycling and vibration, and my concern has been
that some aspect of our application causes the failures - hence my
original post.

However, there is clearly a reliability issue with this component, and
I am coming round to the view that it is the component and not what we
do with it that is the problem. Unfortunately however, our choices are
rather restricted when it comes to finding a suitable alternative, but
I shall take a close look at the various suggestions for a replacement.

Sounds like you need to dump arcotronic for starters !

Graham
 
That's pretty much the conclusion I'm coming to.

At the time we specified them, (around five years ago now), the SFE
range was the best we could find - hence we ended up using them.
Unfortunately, the manufacturer didn't say: "Oh, by the way, our caps
will develop spurious internal shorts during normal use", so it's not
until we were using them in anger that these problems came to light.

As I've said elsewhere in the thread, we can't use electrolytics in our
designs for aircraft use, so in instances where we need relatively
large amounts of capacitance, this has invariably meant using tantalums
in our applications.

In the light of what's been said on here however, I shall check this -
it may well be that the technology has moved on sufficiently for us to
be able to use them.

Ted
 
M

Mike Monett

Jan 1, 1970
0
Hello Mike

I've had a look at solid polimers and, at first glance, they look very
promising. However, their expected life is only 10 years or so, and we
have a contractual requirement for not less than 25 years. (How
realistic that really is is another matter, but that is what we are
obliged to aim for).

Ted

The article seemed to indicate a 10 year life under full load, and very
long life in storage with low humidity:

http://www.cornell-dubilier.com/tech/spa2.pdf

That was in 2001 - things should have improved since then. Since the
solid polymer caps have lower ESR, less capacitance may be needed to meet
your requirements. This means you could derate heavily and improve the
lifetime.

How do other companies meet a 25 year life? Even the best computer grade
electrolytics have much shorter life under load. And as you have shown,
tantalums are not suited for this application.

Electrolytics may have the shortest life in a system, so a solution to
your problem could help us all.

Mike Monett
 
J

John Larkin

Jan 1, 1970
0
That's pretty much the conclusion I'm coming to.

At the time we specified them, (around five years ago now), the SFE
range was the best we could find - hence we ended up using them.
Unfortunately, the manufacturer didn't say: "Oh, by the way, our caps
will develop spurious internal shorts during normal use", so it's not
until we were using them in anger that these problems came to light.

As I've said elsewhere in the thread, we can't use electrolytics in our
designs for aircraft use, so in instances where we need relatively
large amounts of capacitance, this has invariably meant using tantalums
in our applications.

In the light of what's been said on here however, I shall check this -
it may well be that the technology has moved on sufficiently for us to
be able to use them.

Ted

Considered wet-slug tantalums, like the MIL CSR-series? They don't
have the short-circuit failure mechanism that the dry-slug parts do.

Or Maxwell supercaps?

John
 
T

Terry Given

Jan 1, 1970
0
Joerg said:
Hello John,



Thanks for the info. We have used Panasonic caps a lot as well. No
problems. I also like their ECH and ECHU series when film caps are
called for. Oversizing in capacitance is indeed the thing to do.



The only significant failures I ever saw from aluminum electrolytics was
when people placed them too close to hot stuff, like processors,
regulators or heat sinks. Even then they usually go gracefully, not 4th
of July style like some tantalums.

Regards, Joerg

backwards is never a good idea either :)

I once saw our production manager annihilate the DC bus caps in a 10kW
drive - they exploded, and sprayed boiling electrolyte everywhere. It
took hours to clean up the production test bench! He was testing a 230V
drive on a 400V test bench, and using a motor-driven variac to wind up
the AC line voltage. Problem was, the variac overshot dramatically
(inertia?), and boy the Hitachi AIC caps did NOT like it. Safety goggles
saved his eyes, but his face got burned.

Cheers
Terry
 
Hello again Mike

The short answer as to how other companies meet a 25 year life
requirement is that I don't know that they do.

The 10 year shelf-life/25 year operational requirement is not universal
across BAE, but depends on where the product is to be used - Typhoon
requirements are particularly onerous in all areas.

I only get to see what's going on in external suppliers' products when
I'm asked to sit in on a design review, (usually as a result of
problems with the product), and I would certainly ask questions if I
saw electrolytics being used on a product for Typhoon.

However, things are changing on the military electronics front and some
of the old restrictions are being reviewed, in the light of relatively
recent changes on the component front - the acceptance of the use of
plastic devices being one example of this, brought about by the rapidly
diminishing availability of military ceramic packaged parts.

Ted
 
P

Pooh Bear

Jan 1, 1970
0
Hello again Mike

The short answer as to how other companies meet a 25 year life
requirement is that I don't know that they do.

The 10 year shelf-life/25 year operational requirement is not universal
across BAE, but depends on where the product is to be used - Typhoon
requirements are particularly onerous in all areas.

I only get to see what's going on in external suppliers' products when
I'm asked to sit in on a design review, (usually as a result of
problems with the product), and I would certainly ask questions if I
saw electrolytics being used on a product for Typhoon.

However, things are changing on the military electronics front and some
of the old restrictions are being reviewed, in the light of relatively
recent changes on the component front - the acceptance of the use of
plastic devices being one example of this, brought about by the rapidly
diminishing availability of military ceramic packaged parts.

I forget the name of the process involved but I distinctly recall a US
aerospace engineer saying that commercial reliability figures were so good
that they were being accepted for new build including military.

There's a good reason for this. MIL spec parts are unique and high price
without any attendant benefit for the most part. Indeed by their nature,
they are often 'old technology' and therefore inferior to modern commercial
parts.

I suggest you give BAe a kick up the arse ! They deserve one for so many
reasons it has to be said.


Graham
 
M

Mike Monett

Jan 1, 1970
0
Hello again Mike

The short answer as to how other companies meet a 25 year life
requirement is that I don't know that they do.

The 10 year shelf-life/25 year operational requirement is not universal
across BAE, but depends on where the product is to be used - Typhoon
requirements are particularly onerous in all areas.

I only get to see what's going on in external suppliers' products when
I'm asked to sit in on a design review, (usually as a result of
problems with the product), and I would certainly ask questions if I
saw electrolytics being used on a product for Typhoon.

However, things are changing on the military electronics front and some
of the old restrictions are being reviewed, in the light of relatively
recent changes on the component front - the acceptance of the use of
plastic devices being one example of this, brought about by the rapidly
diminishing availability of military ceramic packaged parts.

Ted

Hi Ted,

I didn't know you were working on the Typhoon - avionics has got to be
one of the toughest design problems, due to rapid and extreme changes in
temperature, power supply spikes and overvoltage, severe vibration,
arcing problems at high altitude, and so on.

My hat's off to you. Let us know if you find a suitable replacement for
the tantalums!

Mike Monett
 
S

Steve Sousa

Jan 1, 1970
0
Hello again Mike

The short answer as to how other companies meet a 25 year life
requirement is that I don't know that they do.


My impression from this thread is that avionics don't use electrolitics
anywhere, is that so? What caps do you use on the power supplies?
 
It depends what the equipment is going to be used on.

In some instances, electrolytics are used, but if I were to try to
introduce them to a design for Typhoon, I would have to be able to
substantiate any claim that they were suitable for use. The general
rule is that we would not use electrolytics on Typhoon, for reasons
primarily of shelf life and operating life requirements.

However, that position may well have changed by the time we bottom this
particular problem.

Ted
 
L

legg

Jan 1, 1970
0
On 10 May 2005 02:20:40 -0700, [email protected] wrote:

.........
So far as dV/dt and dI/dt ratings are concerned, I can find no figures
for either on the data sheet, other than reference to charge/discharge
at rated voltage through 0.5 ohms for 1E6 cycles. We have however
modified our test regime, so that changes in supply voltage occur more
slowly, (~1ms time constant), than was originally the case, It's
worth noting however that some failures have occured when winding the
supply voltage up to the new value on a bench power supply, so dV/dt
was not an issue in these instances and all evidence points to the
problem being voltage related, albeit possibly brought about by some
other mechanism.

Did the bench test failures occur in units that had been previously
'autotested' with the uncorrected test program?

RL
 
Our test regime was changed around six months ago, so, since then,
units have been subjected only to the revised testing, with no
noticeable improvement. Nor have we seen any significant change since
Arcotronic introduced their addtional screening.

Ted
 
M

Mike Monett

Jan 1, 1970
0
Our test regime was changed around six months ago, so, since then,
units have been subjected only to the revised testing, with no
noticeable improvement. Nor have we seen any significant change since
Arcotronic introduced their addtional screening.

Ted

Ted,

Earlier, you posted this comment:
We have traditionally been prevented from using electrolytics in
the military environment, due partly to our 10 year shelf-life
shelf-life requirement, (not 20 years as I stated earlier in the
thread). I will have to check with the powers that be regarding
the possibility of using electrolytics. Clearly, the current level
of failures is unacceptable and we are going to have to do
something to address it.

Perhaps a 10-year cap life is not realistic in this application.
Every aircraft must go through periodic inspections to check for
corrosion, fatigue cracks, hidden damage, etc. Many items are life
or wear limited, such as tires, engines, actuators, and so on. These
must be replaced periodically.

Why not treat the avionics the same as the rest of the a/c and
simply replace the caps during a scheduled inspection? This would
allow you to use whatever performs best in your application, for
example hi-rel solid electrolytics. You could also switch to a
different cap as the product reliability improves over the years,
and save money and time.

Perhaps the avionics already has scheduled inspections. If so, pop
in some new caps and the problem is solved:)

As a pilot, I'd much rather prefer the slow degredation of solid
electrolytics over the unpredictable and catastrophic failure of
tantalums. Consider when you are in a storm approaching the airport
with the weather at minimums. That is the worst time for a system
failure, and the resulting confusion could cause an accident.

So you could argue flight safety as a good reason for the change.

Mike Monett
 
P

Pooh Bear

Jan 1, 1970
0
Our test regime was changed around six months ago, so, since then,
units have been subjected only to the revised testing, with no
noticeable improvement. Nor have we seen any significant change since
Arcotronic introduced their addtional screening.

Is the power supply a flyback or forward type btw ? Flyback in particular
causes high dI/dt.

Graham
 
D

douglas dwyer

Jan 1, 1970
0
In message said:
Our test regime was changed around six months ago, so, since then,
units have been subjected only to the revised testing, with no
noticeable improvement. Nor have we seen any significant change since
Arcotronic introduced their addtional screening.

Ted
Most new designs for power supplies increase the frequency and therefore
require lower value caps for smoothing. The high value ceramics that
can then replace electrolytics rely on very thin dilectric layers with
greater quality of material physically bigger caps could/would be less
stressed.
Interesting to consider active capacitors ie small caps multiplied by
gain blocks, could this yield a more reliable product.?
 
J

John Larkin

Jan 1, 1970
0
Interesting to consider active capacitors ie small caps multiplied by
gain blocks, could this yield a more reliable product.?


Active parts don't store energy.

John
 
Top