| Curious George 2006-03-06, 5:47 pm |
| On 5 Mar 2006 15:50:35 -0800, "w_tom" <w_tom1@usa.net> wrote:
> Look at numerical specs for that UPS. Where do they even claim
>protection from the typically destructive transient? They do not.
It's very common to offer surge filtering and protection. Some even
offer $25,000 Lifetime equipment warranties or similar against
transients &/or defect.
>Once UPSes would claim protection only from normal mode transients.
>Now those same manufacturers don't even mention the various types of
>transients so that consumers will assume protection from ALL types of
>transients. That is what happens when someone declares a plug-in UPS
>as surge protection. They are assuming.
Certainly not all are equal. But the point being discussed here is
whether HDD's can get fried regardless of _any_ attempts of
protection.
> Transient that typically damages computer hardware; plug-in UPSes
>don't even claim to protect from them. Then when this transient (that
>occurs maybe once every eight years) damages hardware, others assume
>nothing is 100%. In reality, the protection in a plug-in UPS is the
>same protection found in power strip protectors. And during the only
>destructive transient, it accomplished nothing.
If a lightening strike is reaching the UPS something is wrong IMHO.
Manufacturer's regularly make claims about plug-in products making
such protection.
> If a disk drive and nothing else is damaged by a surge, then there is
>something seriously wrong with how that computer was assembled. To be
It would be either be made of fairy dust, or the user had it open and
was tinkering around, but is wrongfully blaming it on the weather.
>damaged, something must have both an incoming and outgoing transient
>path. This is why surges only damage some parts and do not damage
>others. Many parts of a computer do not have both incoming and
>outgoing paths.
That makes for an interesting electrical circuit ;)
> In one location, a plug-in protector was attached to a powered off
>computer. That plug-in protector provided a surge with more paths into
>that computer.
No it didn't. The paths to a particular computer are fixed. The
surge was external to the plug-in protector. It's the fact that
computers & peripherals are connected and/or on the same circuit that
counts.
>Surge entered this computer, out via network wire, into
>another computer, then out via phone line. We traced that surge by
>replacing every damaged IC in both computers. Yes, adjacent protectors
I hope you mean "retracing." "Replacing" makes no sense here.
>(UPS or power strip since both have same protector circuit) can even
>contribute to damage of the computer. And in this case, other
>computers on same network were also damaged.
The big question is: was the phone line & Ethernet fully protected?
From what you're describing it's much more likely that the lightening
entered via the telephone line, unprotected device on the network, or
exposed wiring.
Were they cheap power strips? There are a lot of cheap products of
every kind that don't work the way they're supposed to.
> Notice a disk drive was not damaged in any computer. Why? Surge had
>an incoming path but no outgoing path through disk drive. This is why
>surges should not damage disks.
How does it not have an outgoing path when it's fixed on the same
circuits as everything else?
> Also notice that computers were powered off when damaged. Just
>another reason why better backup systems also feature properly earthed
>'whole house' protectors where all wires enter a building. The
A properly grounded building & electrical is a necessity that is
totally separate from backup.
As was being discussed before backups need to be taken offline because
they should not be at the mercy of acts of God, or crackers, dumb
users, or whatever. But the thinking should be going way beyond
surge.
>protector is only as effective as its earth ground which is why plug-in
>UPSes do not provide transient protection. Plug-in UPSes are for data
They're supposed to when grounded. There are sprinkler timers that
are sold with lightening surge protection that are not grounded. Lord
knows how either feat are accomplished.
>protection from blackouts and extreme brownouts - a completely
>different type of electrical problem.
Good UPS's claim to do it all. Although it is far more sensible to
rely on proper building grounding and protection at the input panel
for lightening-type severe surges.
> Plug-in UPSes don't even claim to protect from the type of
>transient that causes computer damage. One plans for the worst by
Sure they do. That's a main selling point of these products. Have
you ever read their marketing like "APC Currents," etc.
>installing a less expensive and superior solition - a properly earthed
>'whole house' protector. Such solutions are even sold in Home Depot
>and Lowes.
It is indeed superior. The Electrical co earths the feed externally
to the building. The house & input panel & feed are supposed to be
grounded as well. One can even get surge protection at the circuit
panel/circuit breaker. Outlets need also be grounded. By this point
a UPS's surge capabilities are largely moot.
[vbcol=seagreen]
>Howard Kaikow wrote:
Anything can happen. But that need not be necessary or likely
[vbcol=seagreen]
Nothing is 100% failsafe.
[vbcol=seagreen]
Which involves more than an ext HDD, or at least planning is not tied
specifically to an ext HDD.
|