• ailing power supply or ?

    From Nick Sebul@3:633/10 to All on Wed Apr 15 11:33:17 2026
    Hi,

    I recently replaced my desktop's PSU with an older Ultra ult-vx700 I had
    in storage. At first everything seemed ok, but then a couple of days
    ago, I noticed that the fan would intermittently stop and then restart.
    I checked what details I could find on the PSU and there is no "eco"
    mode, nor is there any setting for such in my BIOS. Since I primarily
    use the latest version of Ubuntu for my OS, I thought perhaps an update
    might have triggered some sort of power saving mode, but doesn't seem
    likely since the fan still does this while starting up before entering
    the OS.

    Today, I replaced the fan with a known good one I had as a spare. Same behavior, so beginning to suspect something with the PSU. Thoughts and comments welcome and thank you in advance.

    Nick

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Jim the Geordie@3:633/10 to All on Wed Apr 15 17:06:31 2026
    On 15/04/2026 16:33, Nick Sebul wrote:
    Hi,

    I recently replaced my desktop's PSU with an older Ultra ult-vx700 I had
    in storage.ÿ At first everything seemed ok, but then a couple of days
    ago, I noticed that the fan would intermittently stop and then restart.
    I checked what details I could find on the PSU and there is no "eco"
    mode, nor is there any setting for such in my BIOS.ÿ Since I primarily
    use the latest version of Ubuntu for my OS, I thought perhaps an update might have triggered some sort of power saving mode, but doesn't seem
    likely since the fan still does this while starting up before entering
    the OS.

    Today, I replaced the fan with a known good one I had as a spare.ÿ Same behavior, so beginning to suspect something with the PSU.ÿ Thoughts and comments welcome and thank you in advance.

    Nick

    I can hear my fan start up when I use a lot of CPU, but it stops when I
    stop using that particular program. Probably does the same if I open too
    may tabs. I just thought that was normal.

    --
    Jim the Geordie

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Nick Sebul@3:633/10 to All on Wed Apr 15 12:24:48 2026
    On 4/15/26 12:06 PM, Jim the Geordie wrote:
    On 15/04/2026 16:33, Nick Sebul wrote:
    Hi,

    I recently replaced my desktop's PSU with an older Ultra ult-vx700 I
    had in storage.ÿ At first everything seemed ok, but then a couple of
    days ago, I noticed that the fan would intermittently stop and then
    restart. I checked what details I could find on the PSU and there is
    no "eco" mode, nor is there any setting for such in my BIOS.ÿ Since I
    primarily use the latest version of Ubuntu for my OS, I thought
    perhaps an update might have triggered some sort of power saving mode,
    but doesn't seem likely since the fan still does this while starting
    up before entering the OS.

    Today, I replaced the fan with a known good one I had as a spare.
    Same behavior, so beginning to suspect something with the PSU.
    Thoughts and comments welcome and thank you in advance.

    Nick

    I can hear my fan start up when I use a lot of CPU, but it stops when I
    stop using that particular program. Probably does the same if I open too
    may tabs. I just thought that was normal.

    Not normal for mine until the last couple of days. Yes, it would speed
    up slightly or slow down, depending upon the load, but not stall
    entirely like now.


    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Wed Apr 15 12:37:09 2026
    On Wed, 4/15/2026 12:06 PM, Jim the Geordie wrote:
    On 15/04/2026 16:33, Nick Sebul wrote:
    Hi,

    I recently replaced my desktop's PSU with an older Ultra ult-vx700 I had in storage.ÿ At first everything seemed ok, but then a couple of days ago, I noticed that the fan would intermittently stop and then restart. I checked what details I could find on the PSU and there is no "eco" mode, nor is there any setting for such in my BIOS.ÿ Since I primarily use the latest version of Ubuntu for my OS, I thought perhaps an update might have triggered some sort of power saving mode, but doesn't seem likely since the fan still does this while starting up before entering the OS.

    Today, I replaced the fan with a known good one I had as a spare.ÿ Same behavior, so beginning to suspect something with the PSU.ÿ Thoughts and comments welcome and thank you in advance.

    Nick

    I can hear my fan start up when I use a lot of CPU, but it stops when I stop using that particular program. Probably does the same if I open too may tabs. I just thought that was normal.


    When Intel "invented" PWM fans, by writing a spec for it,
    the intention was, at zero control input, the fan was
    supposed to spin at MIN, not at ZERO. Having fans that
    stop spinning (one way or another), confuses and concerns
    users, as they cannot be sure what is going on.

    This is an Ultra brand ATX, a company consisting of four guys in
    Cali, who contracted the manufacturing to others. In their
    support forum, they used to "berate" users who reported
    valid concerns with the products. I don't know if they're
    still in business or not. I also cannot tell you who makes
    their PSUs (since they didn't have their own factory in Cali).
    Lots of other companies have used this model (PCPC),
    and they had a much better reputation for knowing their
    stuff, even though they were not personally making the items.

    *******

    For that particular model, someone had a situation where
    the fan speed went up for no reason.

    https://www.overclock.net/threads/psu-fan-speeds.410102/

    There was one build with a failure to start, which isn't
    conclusive of anything.

    The era might be the year 2007 ??? The fan should be running
    off the +12V rail, which is 12V @ 40A or so. Via current
    limiters, that would be split, so 20A or so would go to
    the CPU, and 20A would go to the 24 pin connector. At a guess.

    https://en.wikipedia.org/wiki/Capacitor_plague

    It's hard to say what kind of circuit feeds the fan speed.
    There's no processor inside most ATX power supplies, so
    an analog circuit controls the fan speed. On the PC itself,
    the SuperI/O has some automation for adjusting the fan speed,
    but that's not always "clever enough" for the job, so the
    BIOS calls via SMI, could include an opportunity for adjustment.
    The PSU is a lot dumber, and should react to a temperature
    measured somewhere.

    *******

    OK, I have a theory. The circuit uses an electrolytic cap,
    the cap is leaking, and the cap is behaving as much as a
    resistor as it is a capacitor.

    You can undo the four screws that hold the ATX top sheet metal
    U-plate cover. You don't need to touch any wiring or
    components inside. Just look at the tops of the caps for
    brown deposits (leaking at the vent on top). The Capacitor Plague
    article has pictures of the brown stuff. The vent
    is there for emergency pressure release (to avoid an
    explosion as the release mechanism).

    I had an Antec, a ChannelWell OEM, where the four +5V output
    caps were leaking. And the PSU, when the machine was cold,
    a puff of gray smoke used to come out of the PSU vent.
    That tells you that failure due to leaking caps is imminent.
    A second Antec, it had some sort of defect, but there
    were no visual indicators at all of what the problem was
    with it. The problem would be somewhere around the
    dual transistors that switch the primary side of the transformer.
    But nothing was burned or discolored. And that was a long
    time ago now.

    Paul

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Nick Sebul@3:633/10 to All on Thu Apr 16 11:56:57 2026
    On 4/15/26 12:37 PM, Paul wrote:
    On Wed, 4/15/2026 12:06 PM, Jim the Geordie wrote:
    On 15/04/2026 16:33, Nick Sebul wrote:
    Hi,

    I recently replaced my desktop's PSU with an older Ultra ult-vx700 I had in storage.ÿ At first everything seemed ok, but then a couple of days ago, I noticed that the fan would intermittently stop and then restart. I checked what details I could find on the PSU and there is no "eco" mode, nor is there any setting for such in my BIOS.ÿ Since I primarily use the latest version of Ubuntu for my OS, I thought perhaps an update might have triggered some sort of power saving mode, but doesn't seem likely since the fan still does this while starting up before entering the OS.

    Today, I replaced the fan with a known good one I had as a spare.ÿ Same behavior, so beginning to suspect something with the PSU.ÿ Thoughts and comments welcome and thank you in advance.

    Nick

    I can hear my fan start up when I use a lot of CPU, but it stops when I stop using that particular program. Probably does the same if I open too may tabs. I just thought that was normal.


    When Intel "invented" PWM fans, by writing a spec for it,
    the intention was, at zero control input, the fan was
    supposed to spin at MIN, not at ZERO. Having fans that
    stop spinning (one way or another), confuses and concerns
    users, as they cannot be sure what is going on.

    This is an Ultra brand ATX, a company consisting of four guys in
    Cali, who contracted the manufacturing to others. In their
    support forum, they used to "berate" users who reported
    valid concerns with the products. I don't know if they're
    still in business or not. I also cannot tell you who makes
    their PSUs (since they didn't have their own factory in Cali).
    Lots of other companies have used this model (PCPC),
    and they had a much better reputation for knowing their
    stuff, even though they were not personally making the items.

    *******

    For that particular model, someone had a situation where
    the fan speed went up for no reason.

    https://www.overclock.net/threads/psu-fan-speeds.410102/

    There was one build with a failure to start, which isn't
    conclusive of anything.

    The era might be the year 2007 ??? The fan should be running
    off the +12V rail, which is 12V @ 40A or so. Via current
    limiters, that would be split, so 20A or so would go to
    the CPU, and 20A would go to the 24 pin connector. At a guess.

    https://en.wikipedia.org/wiki/Capacitor_plague

    It's hard to say what kind of circuit feeds the fan speed.
    There's no processor inside most ATX power supplies, so
    an analog circuit controls the fan speed. On the PC itself,
    the SuperI/O has some automation for adjusting the fan speed,
    but that's not always "clever enough" for the job, so the
    BIOS calls via SMI, could include an opportunity for adjustment.
    The PSU is a lot dumber, and should react to a temperature
    measured somewhere.

    *******

    OK, I have a theory. The circuit uses an electrolytic cap,
    the cap is leaking, and the cap is behaving as much as a
    resistor as it is a capacitor.

    You can undo the four screws that hold the ATX top sheet metal
    U-plate cover. You don't need to touch any wiring or
    components inside. Just look at the tops of the caps for
    brown deposits (leaking at the vent on top). The Capacitor Plague
    article has pictures of the brown stuff. The vent
    is there for emergency pressure release (to avoid an
    explosion as the release mechanism).

    I had an Antec, a ChannelWell OEM, where the four +5V output
    caps were leaking. And the PSU, when the machine was cold,
    a puff of gray smoke used to come out of the PSU vent.
    That tells you that failure due to leaking caps is imminent.
    A second Antec, it had some sort of defect, but there
    were no visual indicators at all of what the problem was
    with it. The problem would be somewhere around the
    dual transistors that switch the primary side of the transformer.
    But nothing was burned or discolored. And that was a long
    time ago now.

    Paul

    I haven't checked the caps yet, but I am going to at some point. I am beginning to wonder though if this isn't some sort of normal behavior
    for this PSU. When the room temps were cooler, both of its fans ran continuously at a slow speed. It has only been since the temps have approached or went above 70 F in here that this behavior started.

    I also checked the fan at the rear of the PSU as this is a dual fan
    unit. Both stop. Then, once a few seconds go by, both start up again.

    The only difference with the PSU behavior is that now, when the fans
    start, they are spinning at a faster rate but not continuously and then
    go off. The behavior when the room was cooler was continuous fans
    running at a slow speed (but, to be honest, maybe they weren't running
    at all and I thought they were running slowly).

    My understanding was that this PSU was intended for lower end gaming
    systems. As such, I never owned anything like it, only standard
    supplies (basically whatever came with the PC). So, this behavior I'm thinking is odd might be normal.

    I guess I'll know for sure once the cool snap arrives and the room drops
    back into the low 60's/ upper 50's. If the behavior ceases, then I'll
    pretty much know its how its designed.


    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Thu Apr 16 14:55:17 2026
    On Thu, 4/16/2026 11:56 AM, Nick Sebul wrote:


    I haven't checked the caps yet, but I am going to at some point.ÿ I am beginning to wonder though if this isn't some sort of normal behavior for this PSU.ÿ When the room temps were cooler, both of its fans ran continuously at a slow speed.ÿ It has only been since the temps have approached or went above 70 F in here that this behavior started.

    I also checked the fan at the rear of the PSU as this is a dual fan unit.ÿ Both stop.ÿ Then, once a few seconds go by, both start up again.

    The only difference with the PSU behavior is that now, when the fans start, they are spinning at a faster rate but not continuously and then go off.ÿ The behavior when the room was cooler was continuous fans running at a slow speed (but, to be honest, maybe they weren't running at all and I thought they were running slowly).

    My understanding was that this PSU was intended for lower end gaming systems.ÿ As such, I never owned anything like it, only standard supplies (basically whatever came with the PC).ÿ So, this behavior I'm thinking is odd might be normal.

    I guess I'll know for sure once the cool snap arrives and the room drops back into the low 60's/ upper 50's.ÿ If the behavior ceases, then I'll pretty much know its how its designed.

    Does your case have enough intake-fans, to pressurize the case
    and cause some air to blow through the PSU anyway ?

    That's one way to provide air to a PSU, if it won't
    do the job itself. Not a particularly good way,
    but it's better than nothing.

    While in theory, you can fit your own fan
    to the exterior of the PSU, there isn't always room
    or mount points, to do it that way.

    Paul

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Nick Sebul@3:633/10 to All on Thu Apr 16 15:20:17 2026
    On 4/16/26 2:55 PM, Paul wrote:
    On Thu, 4/16/2026 11:56 AM, Nick Sebul wrote:


    I haven't checked the caps yet, but I am going to at some point.ÿ I am beginning to wonder though if this isn't some sort of normal behavior for this PSU.ÿ When the room temps were cooler, both of its fans ran continuously at a slow speed.ÿ It has only been since the temps have approached or went above 70 F in here that this behavior started.

    I also checked the fan at the rear of the PSU as this is a dual fan unit.ÿ Both stop.ÿ Then, once a few seconds go by, both start up again.

    The only difference with the PSU behavior is that now, when the fans start, they are spinning at a faster rate but not continuously and then go off.ÿ The behavior when the room was cooler was continuous fans running at a slow speed (but, to be honest, maybe they weren't running at all and I thought they were running slowly).

    My understanding was that this PSU was intended for lower end gaming systems.ÿ As such, I never owned anything like it, only standard supplies (basically whatever came with the PC).ÿ So, this behavior I'm thinking is odd might be normal.

    I guess I'll know for sure once the cool snap arrives and the room drops back into the low 60's/ upper 50's.ÿ If the behavior ceases, then I'll pretty much know its how its designed.

    Does your case have enough intake-fans, to pressurize the case
    and cause some air to blow through the PSU anyway ?

    It's whatever the case for the Dell XPS420 provides. Other than the
    power supply having the two fans, there is of course another fan for the
    CPU. Otherwise, nothing else.


    That's one way to provide air to a PSU, if it won't
    do the job itself. Not a particularly good way,
    but it's better than nothing.

    This PSU is actually overkill for the Dell as it's highest rated PSU was
    450W I believe.

    I may try running it out of the case for a while to see what happens.


    While in theory, you can fit your own fan
    to the exterior of the PSU, there isn't always room
    or mount points, to do it that way.

    Are you saying to mount another fan on top of one of the existing ones?

    One thing's for sure though. I do audio work and have a decent set of
    monitor speakers. When the fans kick on at the speed they currently
    are, it's already interfering with my hearing. I can't imagine when
    room temps get into the low 80's in another couple of months what the
    sound will be like. I could go back to headphones, but that defeats the purpose of the monitors.

    I think it might be time for a different PSU.


    Paul



    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From J. P. Gilliver@3:633/10 to All on Thu Apr 16 22:22:35 2026
    On 2026/4/16 16:56:57, Nick Sebul wrote:
    []
    for this PSU. When the room temps were cooler, both of its fans ran continuously at a slow speed. It has only been since the temps have approached or went above 70 F in here that this behavior started.

    I also checked the fan at the rear of the PSU as this is a dual fan
    unit. Both stop. Then, once a few seconds go by, both start up again.

    The only difference with the PSU behavior is that now, when the fans
    start, they are spinning at a faster rate but not continuously and then
    go off. The behavior when the room was cooler was continuous fans
    running at a slow speed (but, to be honest, maybe they weren't running
    at all and I thought they were running slowly).

    My understanding was that this PSU was intended for lower end gaming systems. As such, I never owned anything like it, only standard
    supplies (basically whatever came with the PC). So, this behavior I'm thinking is odd might be normal.

    I guess I'll know for sure once the cool snap arrives and the room drops back into the low 60's/ upper 50's. If the behavior ceases, then I'll pretty much know its how its designed.

    Could you turn up the A/C - or, put a desk fan on the PC - to see?
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()ALIS-Ch++(p)Ar++T+H+Sh0!:`)DNAf

    The Daily Mail has led the campaign to limit pornography - "it demeans
    and belittles women," they explain, "and that's our job."
    (Sandi Toksvig [scripted], News Quiz 2013-7-26.)

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Thu Apr 16 17:22:50 2026
    On Thu, 4/16/2026 3:20 PM, Nick Sebul wrote:


    I think it might be time for a different PSU.

    There are some pretty efficient power supplies out there.
    Some will remain fanless up to 300-400W of load
    (which is enough power to run anything I own in the room here).

    There is a color chart, 25% of the way down this web page.

    https://en.wikipedia.org/wiki/80_Plus

    At my computer store here, I'd check online to see if they
    had stock. For one of my supplies, I headed over to Best Buy
    and there was a suitable one there (which the computer store
    didn't have stock). It's not that PSU are under particular
    pressure, but sometimes there aren't a lot of units in stock
    for any one model line. I buy locally, so if there is
    infant mortality on an item, I can get a quick return
    without a fuss. For example, out of slightly more than
    30 HDD, only one failed on me when I got it home,
    the motor refused to spin in it. And those are supposed
    to be 100% tested too. I don't think I've had any PSU
    fail on the first day -- I check them with a home-made
    load tester (just 100 watts worth of resistors, nothing
    fancy).

    In the leaky cap era, I bought one supply as a spare,
    it sat on the shelf here for two years. I took it off
    the shelf and the supply had already failed (leaking
    caps on the output side were gone). That's an example of
    a capacitor failing "not under bias", which is never
    supposed to happen. That's because the pH of the electrolyte
    on those caps is wrong and they eat right through their
    casing. And two years sitting quietly, is enough for
    them to eat right through. Otherwise, we rely on
    high temperatures and/or high ripple currents,
    to accelerate the demise of caps. The caps contain
    liquid, and when the seal at either end fails, they
    dry out. A puff of grey smoke at power up, is
    another sign of impending failure -- if you see
    grey smoke, the supply will die in around seven days
    or so.

    Paul

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Nick Sebul@3:633/10 to All on Thu Apr 16 19:12:27 2026
    On 4/16/26 5:22 PM, J. P. Gilliver wrote:
    On 2026/4/16 16:56:57, Nick Sebul wrote:
    []
    for this PSU. When the room temps were cooler, both of its fans ran
    continuously at a slow speed. It has only been since the temps have
    approached or went above 70 F in here that this behavior started.

    I also checked the fan at the rear of the PSU as this is a dual fan
    unit. Both stop. Then, once a few seconds go by, both start up again.

    The only difference with the PSU behavior is that now, when the fans
    start, they are spinning at a faster rate but not continuously and then
    go off. The behavior when the room was cooler was continuous fans
    running at a slow speed (but, to be honest, maybe they weren't running
    at all and I thought they were running slowly).

    My understanding was that this PSU was intended for lower end gaming
    systems. As such, I never owned anything like it, only standard
    supplies (basically whatever came with the PC). So, this behavior I'm
    thinking is odd might be normal.

    I guess I'll know for sure once the cool snap arrives and the room drops
    back into the low 60's/ upper 50's. If the behavior ceases, then I'll
    pretty much know its how its designed.

    Could you turn up the A/C - or, put a desk fan on the PC - to see?

    Actually, yes, I could. I have the desktop side cover off anyway so no
    chore to send some air in that direction. Thanks for the suggestion.

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Hank Rogers@3:633/10 to All on Thu Apr 16 19:38:41 2026
    Paul wrote on 4/16/2026 4:22 PM:
    A puff of grey smoke at power up, is
    another sign of impending failure -- if you see
    grey smoke, the supply will die in around seven days
    or so.

    Paul

    What if the smoke is green? Is there a way to calculate when the power
    supply will die?

    I think you'd get less than seven days if the smoke isn't grey.



    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Fri Apr 17 02:55:34 2026
    On Thu, 4/16/2026 8:38 PM, Hank Rogers wrote:
    Paul wrote on 4/16/2026 4:22 PM:
    A puff of grey smoke at power up, is
    another sign of impending failure -- if you see
    grey smoke, the supply will die in around seven days
    or so.

    ÿÿÿ Paul

    What if the smoke is green?ÿ Is there a way to calculate when the power supply will die?

    I think you'd get less than seven days if the smoke isn't grey.



    No matter how many companies make them, electrolytics
    are all made the same way. When they explode, matte black
    confetti comes out, and it's a bit "greasy" and makes a mess.
    That tells you that the available colours inside there are
    limited. The grey smoke on the other hand, the vent is
    open and there is an escape route for the smoke. Once
    the cap warms a bit at startup, the smoke stops.

    The reason you don't want to leave defective kit running
    for too long, is OCP might not work. A review in Tomshardware
    yesterday for a 1200W PSU, the OCP was set to around 149%, which
    is too high to be protecting properly (120-130%). If a fault develops
    in that supply, it could continue to run with the
    fault in place, and it might support an open flame
    while doing that. The design rules are, you're
    not supposed to support an open flame. One USENET
    poster had a small flame shoot out of the fan hole,
    which is one reason I can tell visitors here, to not
    set a desktop too near the curtains in a room. But that
    doesn't happen that often. The components come
    with flammability ratings, and on your BOM you would
    hope every item meets the same flammability spec.
    It means you can't make capacitors out of paraffin :-)

    Why does a lithium battery laptop not follow the rules ?
    Don't ask, don't tell.

    Paul

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)