Page 2 of 2 FirstFirst 12
Results 11 to 16 of 16

Thread: Interesting comment about SMPS component life

  1. #11
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    1,587
    Thanks Thanks Given 
    61
    Thanks Thanks Received 
    5
    Thanked in
    5 Posts

    Default

    Quote Originally Posted by wwenze View Post
    A PSU that uses fans and capacitors as a timebomb is shoddy design - the rest of the product is overspecced. All of them should be part of the timebomb.

    Or use quality capacitors and fans.

    I need my computers to be happy at 110F ambient as I can't afford air conditioning.

  2. #12
    Join Date
    Jul 2018
    Location
    Manitoba, Canada
    Posts
    16
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by jonnyGURU View Post
    Heat and heat cycles are what cause them to degrade and die. They don't die "by design", but to assume that FETs, diodes, etc. in an SMPS are "just as good as day one" is naive.

    Keep in mind that one of my first jobs in this industry was in the RMA department of a big wholesaler. I've seen 2" tall blue and white flames shoot out of CPUs, GPUs, RAM, chipsets, etc. for inexplicable reasons.

    And now that I'm at Corsair, I can say I still see that trend. Based on some of what's in that article, and your own input about fans, that every failed PSU that comes back to Corsair under warranty failed because of failed caps or fans.. That's farthest from the truth.
    I see PSU sections die fairly often due to MLCC caps failed short, usually in laptops. As for failure at EOL, even coils and transformers have an end to their bathtub curve, although it's reasonably possible that they're designed to have effectively indefinite life compared to the longevity of the technology their used in.

    Also, heat isn't the only cause of failure. Beyond design basis voltage transients are an extremely common problem for components, although less so for consumer PSUs.

  3. The Following User Says Thank You to gdjacobs For This Useful Post:

    Orion (07-12-2018)

  4. #13
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    1,587
    Thanks Thanks Given 
    61
    Thanks Thanks Received 
    5
    Thanked in
    5 Posts

    Default

    Yes, that was my point, coils have an almost indefinite life, although I keep changing the pump on my washing machine every 5 years or so, and it is the coils that go. It is a 220V pump run in a 110V machine (by design), so I assume it is fed a square wave, and the higher harmonics may be causing more heating than it can handle.

    So maybe the Guru is right ...

  5. #14
    Join Date
    Jul 2009
    Location
    Germany
    Posts
    3,409
    Thanks Thanks Given 
    6
    Thanks Thanks Received 
    12
    Thanked in
    10 Posts

    Default

    Quote Originally Posted by ashiekh View Post
    http://www.repairfaq.org/sam/smpsfaq.htm#smpsror

    "Nothing really degrades in a switchmode power supply except possibly the electrolytic capacitors..."

    I would add fan to this comment, but I suspect that even then some people here would not agree.
    That's just bullshit.
    EVERYTHING degrades over time in a PSU, not just the electrolytics.

    Even resistors can go bad, the ICs can electromigrate, though it takes a bit of time, so does every other Silicone based thing in a PSU...

    So no, everything degrades. Even the copper on the PCB corrodes...

  6. The Following User Says Thank You to Stefan Payne For This Useful Post:

    Philipus II (07-12-2018)

  7. #15
    Join Date
    Nov 2010
    Location
    Serbia
    Posts
    2,236
    Thanks Thanks Given 
    125
    Thanks Thanks Received 
    2
    Thanked in
    2 Posts

    Default

    Sure, in an absolute sense, everything does degrade.

    The question here is: What is it that noticeably and measurably degrades over the expected service life of the product (PSU)?

    Given that context, you have to agree that electrolytic capacitors and the cooling fan are usually the first to breach the threshold of expected and acceptable performance/characteristics. This, of course, after you discount the 13007s mounted on a paper-thin sink in a crappy pre-Y2K design exploding when asked for > 250W (i.e. design flaws leading to over-stressing certain components, causing them to fail early).

    Do we really need to dissect every word and get caught up in semantics? I mean, I myself deeply value precision and clarity; but it was obvious that the timeframe considered was 5-10 years of 8-10h of daily usage, with misuse and design flaws not taken into account.
    Careful what you wish for... You just might get it.

  8. The Following User Says Thank You to McSteel For This Useful Post:

    ashiekh (07-12-2018)

  9. #16
    Join Date
    Jul 2009
    Location
    Germany
    Posts
    3,409
    Thanks Thanks Given 
    6
    Thanks Thanks Received 
    12
    Thanked in
    10 Posts

    Default

    Quote Originally Posted by McSteel View Post
    Given that context, you have to agree that electrolytic capacitors and the cooling fan are usually the first to breach the threshold of expected and acceptable performance/characteristics.
    No, it sees that the +5VSB Controller also tends to die.
    And in some designs its more likely than caps and fan.

    With cheapish units, yes, the fan might be the first to go but on mid range units not soo much.

    Quote Originally Posted by McSteel View Post
    Do we really need to dissect every word and get caught up in semantics? I mean, I myself deeply value precision and clarity; but it was obvious that the timeframe considered was 5-10 years of 8-10h of daily usage, with misuse and design flaws not taken into account.
    Yes and even then you can't say, it depends on the Design/ayout.
    How much do the caps get cooked, how many in parallel and so on...

    And in recent weeks/months, I rarely see anything that might be related to defective caps or even defective fans...

    And according to Jon, MOSFETs failure does also happen.

    So we are talking about +5VSB Chips exploding, (PFC) MOSFETs dying and wet lytics and of course dying fans...

Similar Threads

  1. SMPS car battery chargers
    By ashiekh in forum PC Power Supply Discussion
    Replies: 4
    Last Post: 04-19-2018, 09:53 AM
  2. Textbooks on SMPS theory/design?
    By Travis in forum Electronic Theory and Principle
    Replies: 5
    Last Post: 12-22-2017, 06:33 PM
  3. Building a SMPS looking for learning help..
    By Aalelan in forum PC Power Supply Discussion
    Replies: 0
    Last Post: 02-09-2011, 12:16 PM
  4. Need your comment about this PSU
    By apvm in forum PC Power Supply Discussion
    Replies: 6
    Last Post: 12-14-2008, 11:36 AM
  5. hii..comment on this PSU??
    By forum67 in forum PC Power Supply Discussion
    Replies: 8
    Last Post: 06-14-2008, 11:38 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •