Author Topic: Can components die honorably?  (Read 619 times)

0 Members and 1 Guest are viewing this topic.

Offline asymcon

  • Contributor
  • Posts: 7
  • Country: pl
Can components die honorably?
« on: December 20, 2018, 07:38:51 pm »
Two AOD4126s just commited suicide rather than to mate with my HP 65W adapter. Coincidentally XL4016E1 gave in. And to top it all off, XL6009E1 died of unknown causes. All in one day, and all in devices I have no backup of. Few days ago, IRFZ44 was proclaimed dead in my Imax B6.
Can't electronics just slowly, gradually pass away with age?  |O :-DD
Now I have to wait 30 days for a shipment from China...

BTW, to be fair, I killed the 4016E1, but the rest is for Sherlock Holmes
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 3797
  • Country: au
  • Question Everything... Except This Statement
Re: Can components die honorably?
« Reply #1 on: December 20, 2018, 07:56:41 pm »
Had enough devices that slowly degraded until I got stuck diagnosing there marginal intermittent faults and where promptly removed for refurbishment.

Many power transistors running hot suffereing from beta loss,
Many bridge rectifier diodes with increased leakage,
H bridges with partial shoot-through thanks to parasitics from dust buildup,
Logic gates that degraded into linear modes.
Canbus Transceivers that fail with can low shorted to ground.

As for devices that have failed gracefully,
Generally hard shorted tantalums on current limited rail, nice clear fault, generally noting damaged, and easy to diagnose.
Open Transistors, C-E open with no damage to the base driving circuit, not as common, but a pleasure to diagnose.
Resistors that creep up in value (generally old film based power resistors), they get hotter and hotter leaving a nice tell, and generally only brownouts.
 

Offline Wan Huang Luo

  • Regular Contributor
  • *
  • Posts: 137
  • Country: us
  • 顽谎骆 from the Shenzhen Market
Re: Can components die honorably?
« Reply #2 on: December 21, 2018, 01:48:46 am »
Sounds like a question for the Klingon forum. Q'pla!
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 2947
  • Country: us
Re: Can components die honorably?
« Reply #3 on: December 21, 2018, 07:48:40 am »
In my experience most do fail gracefully.  But those of us employing them blissfully go on using them long after they don't meet their original performance and then cursing them when they give their last gasp.

The prime exception I have observed to this pattern is when we torture them by operating the welder on the same mains circuit as all of the little silicon die, or when we toss them in the closet of slow rot for months or years and then expect them to hit the ground running.
 
The following users thanked this post: Inverted18650

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 4869
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Can components die honorably?
« Reply #4 on: December 21, 2018, 11:38:04 am »
Most common cause of gradual failure of electronics in my experience seems to be aged capacitors. I rarely find silicon devices degrade gradually as opposed to fail suddenly, a notable exception being cheap smartphones used for cryptocurrency mining that tend to go more and more flaky after a year or so until they won't work anymore. I suppose it's a bit much to expect a $10 smartphone to last long under a 24/7 heavy load...
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline TerraHertz

  • Super Contributor
  • ***
  • Posts: 3498
  • Country: au
  • Why shouldn't we question everything?
    • It's not really a Blog
Re: Can components die honorably?
« Reply #5 on: December 21, 2018, 12:17:28 pm »
What do you mean by 'honorable'?
I suppose that excludes RIFA mains filter caps that die revoltingly, pooping their stinky burned guts over everything nearby, and smelling up the whole house for days.

Old carbon composition resistors, that just quietly and gradually drift out of tolerance on the high side, that seems fairly honorable to me. A respectable kind of enfeeblement in old age.
Collecting old scopes, logic analyzers, and unfinished projects. http://everist.org
 

Offline Neomys Sapiens

  • Frequent Contributor
  • **
  • Posts: 827
  • Country: de
Re: Can components die honorably?
« Reply #6 on: December 21, 2018, 01:07:02 pm »
A tad cryptic, your post. I had to look up datasheets first, because the IRFZ44 was the only thing readily identifiable for me. And HP 65W adapter means a charger/power brick (ab)used for experimental purpose?
What is the position/function of the MosFETs? The same in all cases?
 

Offline xrunner

  • Super Contributor
  • ***
  • Posts: 3593
  • Country: us
  • hp>Agilent>Keysight>?
Re: Can components die honorably?
« Reply #7 on: December 21, 2018, 01:51:43 pm »
What do you mean by 'honorable'?
I suppose that excludes RIFA mains filter caps that die revoltingly, pooping their stinky burned guts over everything nearby, and smelling up the whole house for days.

Well, Schaffner filters can die an honorable death, this could be applied to RIFA caps I suppose.
I am a Test Equipment Addict (TEA) - by virtue of this forum signature, I have now faced my addiction
 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 2379
  • Country: us
  • Black Magic Design
Re: Can components die honorably?
« Reply #8 on: December 21, 2018, 01:53:18 pm »
how components die
youtube.com/watch?v=UZmP0UJ_YgQ&t=0m45s
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 11869
  • Country: gb
  • Hero999
Re: Can components die honorably?
« Reply #9 on: December 21, 2018, 09:48:35 pm »
Yes, for the most part, components die peacefully, rather than in a cloud of acrid smoke. As mentioned above, quite often the component still works, it's just no longer within the specification. It's much more difficult to diagnose and can cause intermittent faults.

A dying op-amp's bias current may be much higher than specified, due to the beta of the input transistors declining with age. It might not materialise when the circuit is bench tested, yet it behaves erratically in the field, because the beta is temperature dependant, with a negative coefficient, so the output offset could be far too high, when it's cold, yet be perfectly fine at room temperature.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 3924
  • Country: gb
Re: Can components die honorably?
« Reply #10 on: December 21, 2018, 09:58:21 pm »
Many semiconductor devices have selflessly laid down their lives so that their protective fuses might survive, at least for a few milliseconds longer.  I suppose you could call that an honorable, if futile, death.
Chris

"Victor Meldrew, the Crimson Avenger!"
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf