Multiple APC Back-UPS Failures

[longgone]

I have had a number of APC Back-UPS XS-1000 / XS-1200 UPS power
supplies failures over the last year or two. Without exception, the
problem occurs when I turn on the power switch at the back of the
computer. It seems the current inrush overwhelms the UPS and fries it.
So far about 5 or 6 of them have been toasted. The all-up wattage of
the computer (the only device attached to the UPS) is about 600 watts.
I do try to minimize using the main power switch.

To APC's credit, they have replaced them all so far, though the return
postage is starting to add-up.

I would appreciate any advice, perhaps in terms of something that
could mitigate the current inrush?

 

[Bennett Price]

Does monitor get turned on simultaneously w. computer?
A surge suppressor between UPS and PC might help.

 

[longgone]

Just the computer is on the UPS (about 600 watts). I use a second UPS
for the monitor and accessories.

Any suggestions regarding a surge suppressor?

 

[w_tom]

Just the computer is on the UPS (about 600 watts). I use a second UPS
for the monitor and accessories.

Any suggestions regarding a surge suppressor?

Numbers are even printed on the surge protector box. Its let-
through voltage is maybe 330 volts. Does the computer or UPS rise to
330 volts because computer is powered on? That is what Bennett Price
has just claimed. Of course not. If a power surge occurs, a sudden
inrush of current causes voltage to decrease - not increase.

So what does a surge protector do? Ignores any and all electricity
until voltage exceeds 330 volts. Protectors operate like switches.
They remain open circuit until voltage gets too high. A surge
protector will do nothing for startup current - normal or excessive.
Look inside the protector. There is nothing but conductors between
protector plug and protector receptacles. Where is this rumored
protection?

Meanwhile, power supplies containing essential functions (as 30
years ago) also contain an inrush current limiter. Was that computer
power supply one of those discounted type promoted by 'computer
assemblers' only on price and watts? Then it may be missing an inrush
current limiter - causing a large current surge via UPS.

Meanwhile, even that current surge should not harm the UPS. The UPS
is apparently defective - not sufficient for its load.

Far more useful is what exactly is repaired. Information that would
elicit useful replies. But the surge protector recommendation was
from one without even a vague idea what a protector does.

 

[Rod Speed]

(e-mail address removed) wrote

Numbers are even printed on the surge protector box. Its let-
through voltage is maybe 330 volts. Does the computer or UPS rise to
330 volts because computer is powered on? That is what Bennett Price
has just claimed. Of course not. If a power surge occurs, a sudden
inrush of current causes voltage to decrease - not increase.

So what does a surge protector do? Ignores any and all electricity
until voltage exceeds 330 volts. Protectors operate like switches.
They remain open circuit until voltage gets too high. A surge
protector will do nothing for startup current - normal or excessive.

Varys with the technology. The ferroresonants do just that.

Look inside the protector. There is nothing but conductors
between protector plug and protector receptacles.

That is just ONE type of surge protector, stupid.

Where is this rumored protection?

Meanwhile, power supplies containing essential functions (as 30
years ago) also contain an inrush current limiter. Was that computer
power supply one of those discounted type promoted by 'computer
assemblers' only on price and watts? Then it may be missing an
inrush current limiter - causing a large current surge via UPS.

Meanwhile, even that current surge should not harm the UPS.
The UPS is apparently defective - not sufficient for its load.

Far more useful is what exactly is repaired. Information that would
elicit useful replies. But the surge protector recommendation was
from one without even a vague idea what a protector does.

You in spades.

 

[kony]

I have had a number of APC Back-UPS XS-1000 / XS-1200 UPS power
supplies failures over the last year or two. Without exception, the
problem occurs when I turn on the power switch at the back of the
computer. It seems the current inrush overwhelms the UPS and fries it.
So far about 5 or 6 of them have been toasted. The all-up wattage of
the computer (the only device attached to the UPS) is about 600 watts.
I do try to minimize using the main power switch.

To APC's credit, they have replaced them all so far, though the return
postage is starting to add-up.

I would appreciate any advice, perhaps in terms of something that
could mitigate the current inrush?

What makes and models are these PSU in the system(s)?
It appears they don't have sufficient inrush current
limiting, either that or the UPS you have has a design
defect, or less likely a component QC problem (difficult to
presume without any evidence of which part(s) in the UPS
have failed.

You might unplug your PSU from AC for a few minutes, open
them and look for inrush current limiting, or if unable to
identify this feature, linking a good high-res, top-down
picture of one should aid others to determine this.

If it lacks the feature and you're adept at ordering parts
and soldering, you might be able to add it yourself.
Otherwise the PSU or the UPS needs replaced depending on
whether the problem is the former or latter mentioned above,
as we can't really be certain based on the info provided. I
would tend to suspect the UPS is to blame even without
knowing about the PSU, as the UPS itself should also have
sufficient current limiting to prevent damage to itself.

 

[Bennett Price]

Just the computer is on the UPS (about 600 watts). I use a second UPS
for the monitor and accessories.

Any suggestions regarding a surge suppressor?

You want a suppressor that has choke coils; e.g., >
http://www.kalglo.com/s-spiker.htm
since you want to limit current, (not voltage).

As others have suggested, the choke coils could/should be in the PSU
but if the PSU doesn't limit inrush current, an external choke will.

And, like others, I'd suggest asking APC for advice.

 

[longgone]

I have to admitt that I am assuming the cause of the failure is
current inrush. The PSU is an Antec Phantom 500 watt, so I guess that
wattage would be the normal sustained load for which it is designed. I
believe my earlier comment regarding normal computer internal (CPU,
hard drives etc) wattage is high.

The UPS units are APC Back-UPS XS-1000 and XS-1200 models which I
believe are rated in-excess of the 500 watts of the PSU.

 

[kony]

I have to admitt that I am assuming the cause of the failure is
current inrush. The PSU is an Antec Phantom 500 watt, so I guess that
wattage would be the normal sustained load for which it is designed. I
believe my earlier comment regarding normal computer internal (CPU,
hard drives etc) wattage is high.

The UPS units are APC Back-UPS XS-1000 and XS-1200 models which I
believe are rated in-excess of the 500 watts of the PSU.

The PSU wattage does have an indirect correlation to the
inrush current "possible" (depending on existence of and
spec of component(s) used to provide the current limiting),
but does not directly determine the peak possible. That
peak is merely a result of what an unrestricted current flow
into the electrolytic capacitors would cause.

In other words, it doesn't matter at all how much current
the computer itself uses in this case, the problem could
occur even if there wasn't any computer connected to the UPS
at the time (it was turned *on* by the AC input switch).

 

[w_tom]

The UPS units are APC Back-UPS XS-1000 and XS-1200 models which I
believe are rated in-excess of the 500 watts of the PSU.

Your 500 watt supply, like most desktop computers, probably never
draws anywhere near to those 500 watts - with or without inrush
current limiting. Those UPSes should supply something like 500 and
700 watts continuous - even higher during short and sudden current
demands (also called a power surge). Numbers suggest damage is not
due to excessive load.

Numerous AC problems can exist including brownouts, noise, and
harmonics. Excessive harmonics can 'stress' some equipment. Is your
problem based in excessive harmonics? (Harmonics can be created by
other devices on AC electric such as 'ceilings' of fluorescant
lights). Harmonics can cause some equipment to consume more power;
create more heat. IOW harmonics can damage electronics even if UPS
was not providing a load. UPS should not be damaged by excessive
harmonics. Should ...

Were UPSes fixed by replacing failed batteries? Just another
speculation.

Problem: no provided facts explain your damage. At best, we can
only speculate. Known numbers imply that damage did not happen. Your
situation is a failure where facts to explain that failure are
missing.