Is it okay to plug a power-strip into a UPS?

[Bill Sconce]

On Wed, 15 Jun 2005 09:39:15 -0400
"Brian" <gnhlug at karas.net> wrote:

>  
> Interesting, I've never heard of that problem before.  Most surge protectors
> are really only a couple of MOV's anyway.  Do you have a link or something
> handy that has more detail?


There are actually two discussions going on.

1.  "Octopussing" - plugging more than one device into "the one remaining
outlet".  This should be no problem, with the caveat, correctly pointed out
in a previous post (to pay attention to the total load, VA/watts).  It
would be prudent to leave yourself some margin, too - UPS manufacturers
have the usual marketing reasons to be "optimistic" about how much load
their boxes can handle, especially the UPSs sold at the consumer level.

This first sub-discussion has to do with breaking out a connection using
purely passive hardware (like a cube tap - whence my homely term
"octopussing".)


2.  The other sub-discussion has to do with MOVs.  Many power strips
have them, and therefore are not the passive break-out devices in
sub-discussion 1.

There are many cautions to consider about MOVs.  Among them: I would NOT
put any device containing MOVs downstream from any UPS.

The way MOVs work makes it possible for manufacturers to call a device
such as a power strip a "surge suppressor" without actually lying.  But
you would do yourself a favor to remain skeptical.  

An MOV will indeed snub certain kind of voltage spikes.  (Although with
side effects, among them dumping the surge current into the ground lines,
which bears some thoughtful analysis).

An MOV has the tremendous advantage, from the manufacturers' standpoint,
of being extremely cheap (pennies).

The larger disadvantage, particularly with putting them downstream of
a significantly expensive device such as a UPS, is their failure mode.

MOVs wear out.

When an MOV absorbs a surge it suffers damage; eventually it fails.
The failure mode can include physical destruction or thermal runaway.
I'd simply rather not have such devices hanging off my expensive UPS.
(I don't use MOVs at all, actually.)

So:

  o Cube taps, yes, OK
  o "Power strips", probably having MOVs, NO
  o IMO

-Bill


P.S.  It's true that a power strip downstream from a UPS shouldn't
be subject to surges, and that any MOVs in it therefore shouldn't
suffer damage, in which case you break even.  Provided that's the
only place that power strip has ever been used.

P.P.S.  Links?  Sure.  Let's see what Google gives us.  "MOVs failure"
produced the following.  (Worse than I remembered!  MOVs as used in
consumer-grade products are a BAD idea, bordering on fraud.)  Here
are excerpts from the first hits:

http://powerelectronics.com/mag/power_thermally_protected_movs_3/

    Metal-oxide varistors (MOVs) are the most widely used device 
    for protecting sensitive electronics from damaging overvoltage
    transients.  To date, most consider MOVs to be sacrificial 
    devices, expected to disconnect (open) in the event of a 
    failure. Unfortunately, this is not the case. MOVs used in
    surge protection devices (SPDs) can fail explosively when 
    subjected to sustained steady-state power frequency
    overvoltages.
    
http://www.zerosurge.com/HTML/movs.html 

    The heart of many surge suppressors available today is a small 
    electronic component called an MOV (Metal Oxide Varistor). The 
    role of the MOV is to divert surge current, however, MOVs wear 
    out with use. As more surges are diverted, the MOVs life span 
    shortens, and failure becomes imminent. How soon? How many surges
    can they handle? No one knows until failure occurs! There is no
    forewarning or visual indications given - just failure. And while
    failing, they can reach very high temperatures, and actually 
    start fires. See examples of catastrophic MOV failures in the
    photos below.
    
http://www.apttvss.com/white/white-desige-for-safety.asp
    
    When a MOV experiences a transient event larger than its capacity
    or a sustained over-voltage (more than ½ cycle) it will short 
    and begin the process of thermal runaway. During thermal runaway, 
    the MOV will rupture from excessive heat and may expel hot 
    conductive materials. If the MOV is in a symmetrical array with 
    multiple MOVs, the cycle will repeat as adjacent MOVs fail in a
    cascadeing fashion.  This can continue for several seconds or 
    several minutes, depending on the number of MOVs. This can
    potentially cause damage to the SPD unit, connected gear, and
    may cause additional damage if the hot conductive material 
    causes a phase flash-over.

http://www.iaei.org/magazine/04_b/04_b_brown.htm

    If MOVs are used within their well-defined specifications, 
    degradation due to the environment is not likely. However, the
    environment that MOVs are used in is not well-defined. Low 
    voltage ac mains are subject to lightning strikes, switching 
    transients, voltage swells/sags, temporary overvoltages (TOVs)
    and other similar disturbances. Due to the variety of 
    disturbances that MOVs are exposed to, degradation or failure
    are possible in many applications.
    
http://www.nsoem.com/Rayvoss/tech_info.htm

    The page cannot be found
    
    Go to Microsoft Product Support Services and perform a title
    search for the words HTTP and 404.
    
(Heh.  Even curmudgeons need a chuckle now and then.   :)   HTH...