1. Home
  2. Electronics Design
  3. 'Slightly' reversed biased tantalum.

'Slightly' reversed biased tantalum.

I got an email the other day from a professor who was getting flaky currents from one of =91my=92 instruments. Fortunately he is a competent tech and was able to chase the problem down to a tantalum cap. (4.7uF

35 V) The cap was getting hot, he pulled it out and replaced it. All fixed. The cap was on the output of a 5V reference and I assume it was put in backwards. (Though he didn=92t notice the polarity before he pulled it.) This unit has been in =91the field=92 for about eight years. But it only gets intermittent use. It=92s used in student labs so it=92s hard to guess how much time. Maybe a few hours a week on average. So any idea of the lifetime of a tant reversed biased at 1/7th it=92s rated voltage? (I can hear Joerg chuckling from here.) I=92m wondering how many other low voltage time bombs we have out there. I paralleled four of them and they=92ve been sitting on my bench for a few days now (at 5V reverse bias). No magic smoke yet.. the leakage current is a bit more than 1mA for all four... and slowly creeping up. I also hit one with a 5 volt pulse through a 1 k ohm resistor. The same time constant for either polarity, so when slightly reversed it still looks like a 4.7uF capacitor.

George H.

Reply to
George Herold
Loading thread data ...

Careful. I had a reversed one blow up in my face. Customer sent me their breadboard. It powered up as shorted rail. While measuring trace voltage drops it blew. Fortunately I was wearing my OptoVisor's. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.
Reply to
Jim Thompson

ent

ll

d

But it

d to

ow

r

=A0 =A0 ...Jim Thompson

=A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 | =A0 =A0 et =A0 =A0 =A0|

=A0|

=A0 =A0 =A0 |

ide quoted text -

For my test jig I've got a 100 ohm resistor in series, so not much power is availible. I'm guessing these last 'almost forever'... months, at these low voltages. The leakage current is temperature sensitive too. (A heat gun makes 'em leak more.) So it could be one hot summer day that is the last straw.

George H.

Reply to
George Herold

quoted text -

Tants don't need much external power to explode. I have seen some go exotherm early in my career. After that it depends on what else in their vicinity they can ignite. I can already picture it ... black plume of smoke in the distance, sirens wailing, flames licking up into the sky :-)

Why don't you use a 4.7uF ceramic? It's much better.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

quoted text -

You forgot your usual sound effects ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.
Reply to
Jim Thompson

quoted text -

Tants used to go like this:

*PHSSSSSOOOOOSH* 
--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

quoted text -

;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.
Reply to
Jim Thompson

How about the new polymer electrode ones? They are supposed to nullify this failure mode. I just spec-ed some onto my present board. But the pads will fit ceramic too. So I can waffle in my head for a few more weeks of work before sending them to be built.

--
_____________________
Mr.CRC
crobcBOGUS@REMOVETHISsbcglobal.net
SuSE 10.3 Linux 2.6.22.17
Reply to
Mr.CRC

quoted text -

Well, maybe he does not like the sounds being emitted from the Sir Ramic surface..

Reply to
Robert Baer

ky

petent

=A0All

ut

lled

=A0But it

hard to

dea

g how

for

age

me

=A0 =A0 =A0 ...Jim Thompson

=A0 =A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 | =A0 =A0 et =A0 =A0 =A0|

=A0 =A0|

=A0 =A0 =A0 =A0 |

=A0|

|

- Hide quoted text -

ne

Thanks Joerg. I think 'production' would have a hissy fit if I suggested replacing all the tant caps. (a dozen plus instruments with hundreds of tants.) I'm having a hard enough time getting them to double check the polarity on the present boards. And I wonder if changing to all ceramics would cause some of the present circuits to get a bit flaky. But I can certainly not use them in new designs. Do you stick with just the X7R/ X5R type of ceramics? Or delve down into the Y's and Z's?

George H.

Reply to
George Herold

laky

ompetent

uF

=A0All

put

pulled

. =A0But it

s hard to

idea

ing how

h for

akage

r.

ed

t me

ng

=A0 =A0 =A0 ...Jim Thompson

=A0 =A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 =A0 | =A0 =A0 et =A0 =A0 =A0|

=A0 =A0|

=A0 =A0 =A0 =A0 |

=A0|

=A0 |

d.- Hide quoted text -

one

ir

:-)

The ones 'cooking' on my bench at -5 volts are leaking less today. At some level everythings a temperature sensor.

George H.

Reply to
George Herold

r

-)

he

I'm not sure what polymer electrode ones are? Are they still tantalum or something else? (Where do I find them on the Digikey website?)

Thanks

George H.

Reply to
George Herold

In an RF amp I did manage to explode one of those. A big one. When I got busy with broom and shovel to clean up the mess I saw that part of the ceramic had turned into bubbly green glass.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

quoted text -

Occasionally I use Y and Z ceramic for bypassing or AC-coupling. They can also be handy to build poor-man's VCOs at low frequencies. But usually I use X7R, the good stuff. It's so cheap these days. I really do not see any reason for tantalums. Relying on their ESR to placate some ill-designed LDO is like Russian roulette. Nothing is guaranteed, playing with fire, and it would (or at least should) get shot down in a good design review. Plus I do not use LDOs in my designs.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

[...]

I have found X7R to be a bit microphonic, so sometimes I use tantalum for filtering low-noise biasing voltages (since NPO gets big and expensive for this).

--

John Devereux
Reply to
John Devereux

quoted text -

Careful, maybe it's just like the eye of a hurricane :-)

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

quoted text -

Exactly what are you going to do with the results of this experiment?

Using a part "off-label" can be innovative...risky, but innovative...sometimes...if you have control of the source. It's not unusual to get a call from production when the part vendor changes and the application quits working.

Using a part beyond it's specified limits in a production device is grounds for redesign.

Anybody who deliberately uses a device in a mode that, by virtue of the physics and chemistry of the device, guarantees its early demise, possibly in a ball of flame, should be SHOT!! OK, OK, maybe you just cut off their favorite appendage for the first offense.

Your time-wasting experiment provides no justification or redemption for fielding such a design.

Back in the day, Tektronix had a company-wide design rule that all tantalum cap designs must have some minimum equivalent series resistance. Field reliability went way up after implementing that directive.

Reply to
mike

laky

ompetent

uF

=A0All

put

pulled

. =A0But it

s hard to

idea

ing how

h for

akage

r.

ed

t me

ng

=A0 =A0 =A0 ...Jim Thompson

=A0 =A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 =A0 | =A0 =A0 et =A0 =A0 =A0|

=A0 =A0|

=A0 =A0 =A0 =A0 |

=A0|

d.- Hide quoted text -

one

ir

:-)

h

Thanks Joerg. I still have to try the Z type ceramic VCO.. someday.

George H.

Reply to
George Herold

ing flaky

competent

.7uF

t. =A0All

t was put

e pulled

rs. =A0But it

=92s hard to

ny idea

ering how

bench for

leakage

g

tor.

rsed

ent me

ring

=A0 =A0 =A0 =A0 =A0...Jim Thompson

=A0 =A0 =A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 =A0 | =A0 =A0 et =A0 =A0 =A0|

s =A0 =A0|

=A0 =A0 =A0 =A0 =A0 |

t =A0|

|

ood.- Hide quoted text -

re

be one

heir

f

y :-)

t

Geesh, lighten up Mike. The caps are sitting on my bench, it took a few minutes to solder some together and add a current limiting resistor. Once or twice a day I measure the current. The question I'm trying to answer is: Will a few day burn in casue the circuit (with backwards tantalums) to fail? If it does then we can hope to find the problem during testing. If not then it's a potential time bomb. And another solution needs to be found. My boss thought the caps would fail in a few hours. I think I can stop the test after a week.

George H.

I must admit it's kinda interesting that for small reverse voltages they are still OK capacitors.

um

oted text -

Reply to
George Herold
[...]

They really aren't. Be careful with conclusions from such a test, it remains unpredictable:

formatting link
formatting link
$file/2002%20CARTS%20Reverse%20Voltage%20Failures.pdf

[...] 
--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.