UPS recommendations?

Discussion in 'Dell' started by Steve, Jun 26, 2007.

  1. Steve

    w_tom Guest

    I assume you are referring to the only link in your post from
    visitinggeeks.com. That link only claims brownouts cause hardware
    damage. Not true. Brownouts also have no relationship to spikes. So
    what is the UPS doing? That link falsely claims it protects from
    hardware damage due to brownouts. Only true in bad fiction written by
    English majors.

    A computer power supply must contain various functions. For example
    all voltages inside computer must be perfectly stable even when AC
    brownout causes incandescent lamps to dim to 40% intensity. A
    function specifically stated even in Intel ATX specs.

    How do spikes get to a disk drive? Well, AC voltage from mains first
    gets filtered. Then converted to about 300 volts DC and filtered
    again. Then chopped into pulses in the kilohertz range. Then that
    hundreds of volts goes through a transformer. Then converted to DC
    again. Then filtered again. How does that spike ever get to disk
    drive? It does not.

    However many clone computer assemblers only buy power supplies on
    dollars and watts. They have little if any idea of what is in that
    previous paragraph. Therefore power problems that are supposed to be
    made irrelevant by the power supply are 'corrected' by a UPS? By
    saving $10 or $20, they solve the resulting hardware problem with a
    $100 UPS?

    Typical UPS does not even claim to remove spikes. Don't take my
    word for it. Quote its numerical specifications. Worse, some of the
    'spikiest' power is created by 'computer grade' UPSes while in battery
    backup mode. Why is that 'dirty' power irrelevant to computer and
    disk drive? Because computer power supply's job is to make such
    electrical problems irrelevant. How does a spike get through all
    those steps in paragraph 3? It does not - except where myths are
    promoted. Even visitinggeek.com do not discuss spikes.

    UPS is recommended for data protection. To protect data that was
    not saved to disk. Even numerical specifications for computer grade
    UPSes do not claim to 'clean' electricity. What does every disk drive
    demand for clean power? Functions that must be provided in a
    computer's power supply. But again, it means a certified computer
    assembler know things that are not even required by his A+
    certification test - such as how electricity works. Too many
    computers built with power supplies that are missing essential
    functions - bought only on price and watts.
     
    w_tom, Jun 30, 2007
    #41
    1. Advertisements

  2. Brownouts can cause hardware damage although it's not common. Devices
    will often compensate for a low voltage by drawing more current. In
    extreme marginal cases, the greater current can exceed the ratings of
    some components. Really, the increase shouldn't be enough to cause a
    problem, but there are a lot of really cheap, really marginal, really
    badly designed PC power supplies being sold.

    But you continue to insist that UPS' don't suppress spikes when they do;
    they virtually all have a surge/spike suppressor as part of their
    design, and it works, within limits.
     
    Barry Watzman, Jun 30, 2007
    #42
    1. Advertisements

  3. Steve

    w_tom Guest

    To have damage from a spike, first, both an incoming and outgoing
    path must exist. Take modem damage as an example. Typically the
    spike travels through motherboard, through modem, and then to earth
    via phone line. Did memory, NIC, and keyboard also see that same
    spike? Yes, but there was no outgoing path through memory, NIC, or
    keyboard. No damage if only an incoming path (no outgoing path)
    exists.

    Spikes don't crash on electronics like waves on a beach.

    Many somehow feel a spike enters on one wire, destroys electronics,
    then stops. Damage is not created that way. Spikes are electricity.
    That means the spike is first flowing through everything in a
    circuit. In the previous example, that circuit was into motherboard,
    out motherboard into modem, out modem to phone line. Long after that
    spike is flowing through everything in a circuit, only then does
    something fail.

    Two requirements for spike damage. First a complete circuit
    exists. That means both an incoming and outgoing path through device
    must exist. Second, spike flows through everything in that circuit
    before something fails. What would be the incoming and outgoing path
    through a disk drive?

    Meanwhile, any spike on AC mains is simply connected directly to
    computer power supply by a typical UPS. UPS connects computer
    directly to AC mains when not in battery backup mode. Nothing in that
    UPS would stop a spike. UPS numerical specs do not even claim to stop
    that spike.
     
    w_tom, Jun 30, 2007
    #43
  4. Steve

    w_tom Guest

    Even Intel specs were blunt about brownouts not causing damage.
    Specs even stated that computer must power up when incandescent bulbs
    are at less than 40% intensity. Damage from brownouts is possible
    when a system is built missing essential and required functions - a
    human created failure..

    You claim a UPS suppresses spikes? Fine. A 300 volt spike appears
    on the 120 VAC. Show me the manufacturer numerical spec number that
    even claims to suppress that spike. Such functions are provided in
    full featured (building wide) UPSes. Show me where any computer grade
    UPS even claims to suppress such spikes? Little hint. Does not
    happen until that UPS costs $500 and up.

    Meanwhile, power supplies for all electronics are required to make
    that 300 volt spike irrelevant. No wonder a UPS can output a spike of
    up to 270 volts when in battery backup mode - and not harm
    electronics. Where is the protection? Must exist inside
    electronics.

    Barry is asked to provide manufacturer numerical specs for his
    claim. Where are these numbers? Show me.

    Brownouts also do not cause damage. But humans do create (enable)
    that damage. We call it 'cost controls'.
     
    w_tom, Jun 30, 2007
    #44
  5. Steve

    olfart Guest

    YAWN
     
    olfart, Jun 30, 2007
    #45
  6. Steve

    RnR Guest


    Gee, I forgot to post it after my long spiel. Ok, I apologize but let
    me see if I can find it again !!!

    Sorry all...............
     
    RnR, Jun 30, 2007
    #46
  7. Steve

    RnR Guest

    I don't recall saying anyone did say this... I said that it would not
    prevent it .. that's all !!
     
    RnR, Jun 30, 2007
    #47
  8. Steve

    RnR Guest


    Craps.... I can't find the same URL I had. Of course I can find other
    URL's but I'd like to find the same one that I saw this morning.
    I'll try later tonite. My bad guys.

    Sorry again !!
     
    RnR, Jun 30, 2007
    #48
  9. Steve

    olfart Guest

    I've got the Tar
    anyone got Feathers???
     
    olfart, Jun 30, 2007
    #49
  10. Steve

    RnR Guest

    Cute <g> !!
    Ok, put me in the doghouse today :( !!
     
    RnR, Jun 30, 2007
    #50
  11. Steve

    S.Lewis Guest


    *snicker*
     
    S.Lewis, Jul 1, 2007
    #51
  12. Steve

    Sudohnim Guest

    I wouldn't be surprised if some designs failed a worst case analysis.
    Frankly, I wouldn't be surprised if some designs weren't worst case
    analyzed. I would be dumbfounded if that were the case for all
    designs, and shocked if it weren't extremely rare for a mainstream
    design to fail a moderate case analysis. Care to clarify your assertion
    and provide supporting info?
     
    Sudohnim, Jul 1, 2007
    #52
  13. Steve

    RnR Guest


    I think I found the URL that I omitted earlier today. It describes
    what is / causes hard drive failure.
    http://www.pcmech.com/show/harddrive/664
     
    RnR, Jul 1, 2007
    #53
  14. Steve

    Tony Harding Guest

    LOL
     
    Tony Harding, Jul 1, 2007
    #54
  15. Steve

    Tony Harding Guest

    w_tom wrote:

    Where would I obtain such a battery for my Honda? My real world
    experience with a variety of automobiles says 3-4 years is about the
    norm for a car battery. No guarantee, of course, that every car battery
    will last at least 3 years, nor that it won't exceed 4 years.
    The biggest myth I've seen here quotes 7-9 years for a car battery used
    every day.

    ;)
     
    Tony Harding, Jul 1, 2007
    #55
  16. Steve

    Journey Guest

    Agent 99 is now a book author. If I recall correctly, I think one of
    her books is how to be happy as a single person. I think her name's
    Barbara Feldon.

    The shows back then were very original and different -- Gilligan's
    Island, Bewitched, I Dream of Genie .....
     
    Journey, Jul 1, 2007
    #56
  17. Steve

    olfart Guest

    *surprised*
    *dumbfounded*
    *shocked*

    sounds like a review for a grade B movie

    double YAWN
     
    olfart, Jul 1, 2007
    #57
  18. Steve

    Tom Scales Guest

    The movie version of Get Smart is filming with Steve Carrell as Maxwell
    Smart and Anne Hathaway as Agent 99. Good casting.
     
    Tom Scales, Jul 1, 2007
    #58
  19. Steve

    Ron Hardin Guest

    The original Get Smart, all 5 seasons, is available from time-life at $199,
    a superb set, great production values.

    99 in my opinion held the series together, being basically a love story
    for guys (ie no soap opera) ; Max goes on a quest, screws up, 99 accepts him
    anyway. Every male's situation.

    Max had all the action, but 99 focussed everything in a single place, being
    the one who can say that it was okay with her.

    I don't think they realized that that was the draw of the series, and the new
    movie may not do it right.

    The Nude Bomb certainly didn't (no 99 in that one); Get Smart Again did, and was
    nice to see, that their marriage continued into old middle age as you'd hope.

    I don't hold out much hope for the new cast, but it might be okay.

    (I threw out my TV in 1971 in disgust; but Get Smart, my only DVD set, is probably
    the pinnacle of TV art. Up there with isolated episodes of Ernie Kovacs, and
    sometimes in the physical humor with Charlie Chaplin's films.)
     
    Ron Hardin, Jul 1, 2007
    #59
  20. Steve

    Ron Hardin Guest

    She's more of a theater type; probably why none of her relationships in fact
    worked out in the end. A hard role for the guy.

    The book is sad to read, in that she couldn't pull off the simple thing she
    did intuitively and naturally in the series itself, in real life.

    It's meant to be positive and helpful, but reads as pretty clearly a bunch
    of emergency measures to keep from getting depressed, for women living alone,
    just under the surface.

    There would be no corresponding book for guys.
     
    Ron Hardin, Jul 1, 2007
    #60
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.