Results 1 to 6 of 6

Thread: Cap capacity and lifespan

  1. #1
    Join Date
    Aug 2018
    Location
    Korea
    Posts
    243
    Thanks Thanks Given 
    124
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default Cap capacity and lifespan

    Hold cap capacity decreases over time and noticeable difference occurs after long time usage..
    (It might differ by products, but I saw a guy actually checked it and it was about 10y 50%?)

    - What happends when hold cap capacity drops below 'safe' levels?
    - Should I consider cap capacity when buying PSU? Does it really matter?
    - Higher wattage has larger hold cap, how do you think about buying higher wattage for just this reason?

  2. #2
    Join Date
    Nov 2010
    Posts
    380
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default

    Sigh...

    Capacitance of the "hold cap" or primary capacitor or high voltage capacitor is supposed to be big enough to sustain the power supply at its maximum output for at least one AC cycle (1/60 s for the 60 Hz countries), around 16ms. In real world, some power supplies barely manage 10-12ms
    Still, if you have unreliable power, a UPS should be able to switch to providing power from batteries in less than 10ms

    I don't think the actual capacitance value decreases by a significant amount over time. Certainly doubt it's 50% over 10 years. Maybe if you have lousy power and over-voltages that would cause damage to the actual material inside the capacitors.

    What happends when hold cap capacity drops below 'safe' levels? I suppose if you use the power supply near its maximum output and you have unstable AC power, glitches, brownouts, sudden voltage drops, the PC would just instant shut down or reset if the primary capacitor doesn't have enough charge to keep the psu up for those milliseconds of mains power downtime.

    I don't bother analyzing the primary capacitor when making buying decisions but I'm lucky to live in an area where mains electricity is quite good and stable - my computer uptime is usually counted in months as I don't turn off my pc.

  3. #3
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    1,965
    Thanks Thanks Given 
    114
    Thanks Thanks Received 
    55
    Thanked in
    48 Posts

    Default

    Get a high quality unit which uses high quality capacitors; only a few fools (myself included) are still using supplies that are 10 years old.

    It used to be that primary caps rarely failed, but that may be changing with APFC
    Last edited by ashiekh; 06-23-2019 at 11:16 AM.

  4. #4
    Join Date
    Jul 2009
    Location
    Germany
    Posts
    3,654
    Thanks Thanks Given 
    6
    Thanks Thanks Received 
    66
    Thanked in
    60 Posts

    Default

    Look at the Datasheet again, its not saying that the Cap will decrease, its saying that within this it stays within 20% of the specified value for the time.
    What happens after that is not specified and people assume that the capacitors will lose capacity, wich can be the case but it can also be that the leakage increases or the ESR increases. Its just not necessarily inside the spec defined by the manufacturer.

    Quote Originally Posted by mariushm

    Capacitance of the "hold cap" or primary capacitor or high voltage capacitor is supposed to be big enough to sustain the power supply at its maximum output for at least one AC cycle (1/60 s for the 60 Hz countries), around 16ms. In real world, some power supplies barely manage 10-12ms
    Still, if you have unreliable power, a UPS should be able to switch to providing power from batteries in less than 10ms
    ...wich isn't really a problem either way as for 50Hz AC you need 20ms anyway, wich most PSU do not do. So for most of the World, you need a UPS that switches within half of a Sinewave anyway (~10ms)...

    The 16ms Holdup time is only useful for Countrys that use a 60Hz AC and it specifies that the PSU should be able to stay on within one entire sinewave of the voltage...
    ...wich was defined 23 years ago, in ~1996, when the SMPS Technology was still pretty shitty and nobody really reworked it since then, wich is needed more and more each year...

  5. #5
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,086
    Thanks Thanks Given 
    540
    Thanks Thanks Received 
    275
    Thanked in
    199 Posts

    Default

    Quote Originally Posted by Stefan Payne View Post
    The 16ms Holdup time is only useful for Countrys that use a 60Hz AC and it specifies that the PSU should be able to stay on within one entire sinewave of the voltage...
    ...wich was defined 23 years ago, in ~1996, when the SMPS Technology was still pretty shitty and nobody really reworked it since then, wich is needed more and more each year...
    Yes.. and no.

    1/50th of a second is longer, but over the years mains power quality has gotten better.

    And while the spec was written long ago, that doesn't mean it's required. Any PSU manufacturer can make a PSU with a 20ms hold up time if they wanted to... or if it was even necessary.

  6. #6
    Join Date
    Sep 2016
    Location
    VN
    Posts
    185
    Thanks Thanks Given 
    19
    Thanks Thanks Received 
    5
    Thanked in
    5 Posts

    Default

    temperature effect . Caps derate faster if it work in hotter temperature.
    Normal my room temp around 32oC. and I feel cold if temp below 28oC. some friends come from taiwan ask me why in here so hot when I set 25oC in Air conditioner :|
    so every test in 25oC is useless if this psu working in my country . in case it can go upto 50oC . no wonder if some manufacturer ask why your rate RMA is sky high than other

Posting Permissions

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