Lightning - funny how we're not seeing him any more

P

PA20Pilot

........Leythos is posting anecdotal facts.

Facts? I think I'd need to see affidavits from those on the floors above
and below his that had problems when his equipment didn't. Hell, most
everyone has seen lightning blow stuff out of houses, not everything,
and that can't be easily explained either. Why did the garage door quit
and the TV didn't but the message machine did and the VCR didn't but the
microwave did but the electric blanket didn't but the alarm system did
but the etc.....

.......Keep it up! I love good entertainment!

Me too!

---==X={}=X==---


Jim Self
AVIATION ANIMATION, the internet's largest depository.
http://avanimation.avsupport.com

Your only internet source for spiral staircase plans.
http://jself.com/stair/Stair.htm

Experimental Aircraft Association (EAA)
Technical Counselor
 
W

w_tom

Often what appears to be strange or capricious events are
explained when hidden or overlooked electrical paths are
discovered.

To have damage, a complete circuit was exist through
electronics. Once a transient passes through everything in an
electrical path, only then does something in that path fail.
The transient does not just crash on the beach like an ocean
wave. First the transient passes through everything in a
circuit that starts at the cloud and ends up at earthborne
charges miles distant. The building simply becomes a part of
that long electrical circuit from cloud, through building,
into earth, and then over to those earthborne charges. To
suffer damage, a transistor must be between incoming and
outgoing paths.

Things often considered non-electrical conductors can become
electrical conductor during transients. EVen concrete is an
excellent conductor. If a stereo speaker cable contacts the
baseboard heat, that can become an outgoing and destructive
path through a stereo. Incoming on AC electric. Outgoing
through that stereo wire. Yes, wire insulation connects
stereo wire to baseboard heat.

A building is chock full of conductive paths; some that we
initially don't consider conductive. So many paths that some
people instead assume lightning is capricious.

Cited previously was a TV and VCR sharing same electrical
receptacle. TV was not damaged. VCR was. Contributing to
this damage was that the VCR provided a better outgoing path
to earth. VCR conducted a destructive transient to earth;
thereby acting as a very expensive surge protector to the TV.
To better appreciate why some things are damaged and other are
not, one must first learn of every conductive path inside the
building. Again, lightning is not so capricious once we
analyze damage at the electronic component level; learn of
'sneaky' external connections.

A radio station was constructed to eliminate many of those
conductive 'sneak' paths by making the building's floor
equipotential. The floor was made into one big single point
ground beneath equipment so that interior electronics remained
at a constant voltage. No voltage difference (therefore no
separate incoming and outgoing paths) means no destructive
transients:
http://scott-inc.com/html/ufer.htm

Equipotential being one way to eliminate transients through
appliances. But this author also provided his radio station
with best earthing - an Ufer ground. IOW he also made the
entire building into a most conductive earth ground. He
provided protection by two methods: using good conductivity
and making the building equipotential.

We can never make a connection to earth ground sufficient.
So we also make the building equipotential using a single
point earth ground technique. But a truly equipotential
building is not possible. So we make the connection to earth
the most conductive as possible - ie Ufer ground.

The best protection means installing an earthing system when
earth is first dug and footings are poured. If Ufer grounds
cannot be installed, another alternative is the halo ground -
a buried wire surrounding the building. However if neither is
feasible (because the plans were not done at the architects
level), we still bring all utilities into the building as a
same location AND provide utilities with good earthing. Even
one 10 foot earthing rod will be a major earthing
improvement. An electric utility demonstrates the principles
with their 'bad, ugly, and good' examples:
http://www.cinergy.com/surge/ttip08.htm

Another manufacturer demonstrates the principles in a
communication facility on Adobe page 14 at:

http://leminstruments.com/grounding_tutorial/html/index.shtml

Protection involves two basic objectives.

First is to earth a destructive transient before it can
enter the building using a most conductive earth ground.
Earth before a transient can find the so many 'sneak' paths
through appliances. This is accomplished with 'whole house'
protectors (AC electric and phone) or direct hardwire
connections (cable TV and satellite dish), made as short as
possible, to earth ground that is as large or conductive as is
reasonable.

Second is to make the voltage differences between appliances
or between an appliance's 'incoming verses outgoing' wires to
be equipotential. This is accomplished by making that earth
ground a single point ground, addressing the protection in
terms of a building wide and geological evaluation, and again,
bringing everything that could carry a transient into a
building at the common service entrance.

We learn from damage by finding paths into and out of the
electronics that found earth ground using circuits initially
not known to be electrical conductors.

Discover why damage occurs by first learning incoming and
outgoing electrical paths. A most common path that damages
computer modems is incoming on AC electric and outgoing to
earth ground via the telco installed 'whole house' protector.
Notice a transient did not come down the phone line, damage
the modem, then stop. Destructive transients don't crash on
the beach like ocean waves. First a transient establishes a
complete electrical path to earth ground. Then something
fails in that path. Often damaged is a PNP transistor that
drives the modem's off-hook relay. The complete path includes
a direct electrical connection from relay's coil to relay's
wiper. Another example of a path that we normally consider
non-conductive. From relay's data sheets, the breakdown
voltage between that coil and wiper defines another part of
the conductive electrical path.

Lightning seeks earth ground. To discover why some
transistors are damaged and others are not, first find
surprise (sneak) paths to earth ground. To avoid future
damage, modify the incoming path with non-destructive and more
conductive paths to earth (ie. use a well earthed 'whole
house' protector). Never think of transients as capricious.
Transistors are damaged for specific reasons. Learn from
'dead bodies' (the best evidence) why damage occurred; what
was the destructive earthing path. Two principles to superior
protection being a most conductive path to earth and making
the structure equipotential. Protection is not provided by
stopping, blocking, absorbing, or filtering destructive
transients. And yet that is what a plug-in (power strip or
UPS) protector manufacturer hopes you will assume.

Damage is about destructive paths to earth ground via
transistors. Protection has always been about earthing
transients.
 
L

Leythos

This is a very enjoyable argument, but as a disinterested third party, I
have to say that w_tom is definitely blowing away Leythos. He is
posting sound facts, while Leythos is posting anectdotal facts.

I didn't really see it as an argument, I asked him to explain how the UPS
devices remained undamaged while the non-UPS devices were damaged when
both where connected to the same outlet.

It's not about being blow-away or anecdotal, it's just about a simple
observation that I've seen first hand. Since there is no logical
explanation for "something else" protecting the devices it seemed
reasonable to determine that the UPS did indeed protect the devices.

If one is to suggest, as tom does, that a UPS provides no surge
protection, then the devices should not have been undamaged.

I've still not seen an explanation of why devices connected to a UPS
remained undamaged and those not connected (to the same electrical outlet
as the UPS supply) were damaged - and from the looks of it, Tom's not
going to answer it, just keep ignoring it.

I've always come out and admitted when I was wrong, as I know I can be
wrong, but, unless he can clearly explain what I've seen several times
with my own eyes, I'm going to keep believing in the ability of a properly
connected UPS to provide at least some protection against surges.

Now, for any lurkers benefit, I do not disagree that whole-house
protection is good, in fact, I think it's a great thing. I completely
disagree with the assumption that UPS's don't protect devices from surges.
 
M

Michael W. Ryder

w_tom said:
I referred to lurkers: people who read and don't
participate. What was posted in reply to Leythos is only for
their benefit.

Now for a typical plug-in UPS. It claims protection from
two of five types of power problems - blackouts and extreme
brownouts. Neither will damage properly constructed
electronic hardware. As noted, other electrical problems are
better solved elsewhere and by other devices. For example, a
modest brownout where incandescent bulbs dim to 50% intensity:
a problem made completely irrelevant by 'protection' already
inside a minimally acceptable computer power supply. Even
Intel specs make this obvious. A computer that does not power
up everything even when lights are only at 50% intensity
violates even Intel power supply requirements.

Just because Intel (or Microsoft) makes some proclamation that from this
day forth all things will be done this way, does not mean that they are.
That is why additional add on protection is needed.

For transient protection: single point earth ground. A
solution located elsewhere. Anything additional is effective
only if a transient is connected less than 10 feet to that
earth ground. Furthermore, a power cord 'isolator' does not
exist. The green safety ground wire makes such isolation
impossible.

Grounds are only as effective as the surroundings. They probably work
much better in Florida than here in Las Vegas. Even though our building
has the necessary ground stakes, etc. does not mean they work as well as
the same in a wetter climate. The isolator I was talking about is much
like a very large choke to cut down the "noise" in the power signal.

As for ripple, well, a protector is not for such trivial
voltages. On 120 volt service, the protect does zilch until
that 'ripple' increases to 300+ volts. 300+ volts is far above
'ripple' voltages. Ripple being variations of single digit or
tens of volts. But then, this 'ripple' must be eliminated in
any minimally acceptable power supply. An expression
carefully worded because many 'clone' computers don't have
minimally acceptable power supplies. A problem created by
many computer assemblers who don't even have basic electrical
knowledge.

Of course, the numbers posted above should even be provided
in specs for those products and in corresponding standards.
Above concepts are so basic as to be common knowledge among
those with basic technical experience. Those numbers, such as
let-through voltage and normal operating voltage limits, are
even printed on the devices. A messenger need not have any
credibility because those are numbers required to be printed
on the corresponding appliance or protector.

So what does a UPS 'clean'? What does it do? The plug-in
UPS connects computer directly to AC mains when not in power
supply mode. In battery backup mode, plug-in UPS exposes
computer to some of the 'dirtiest' electricity. For example,
a UPS in battery backup mode creates a "modified sine wave"
120 volt AC that is ... two 200 volt square waves with up to a
270 volt spike between those square waves. Is that a sine
wave? Yes. A modified sine wave.

So where is this 'clean' electricity? That 'dirty' battery
backup electricity is still more than clean enough for
computers.

Again, the plug-in UPS outputs a 'clean' sine wave when not
in battery backup mode. Why? It connects computer directly
to AC mains. You can see this on any oscilloscope. Again,
you are expected not to take my word for it. You are expected
to confirm this yourself. Numbers were provided so that you
can even see this yourself.

Other more expensive UPSes do additional functions. A line
interactive UPS would cost $500+. A serious UPS (that even
provides transient protection) is a building wide system
installed back at the breaker box (with a less than 10 foot
connection to earth ground). If you have $25,000 of disk
drives, then you probably has a building wide UPS that
includes many times more functions than found in a plug-in
protector. For example, that building wide UPS may even
address harmonic problems. No plug-in UPS even mentions such
solutions. Your solution would not be a $100 'computer
grade' UPS. Computer grade? It can output a modified sine
wave that may harm electric motors but is sufficient to power
computers. Why? Because computers are more resilient.

The UPS our company went to was a full time 4KVA system. Once it was
installed we never had another equipment failure. So obviously the UPS
did its job. One that grounds, etc. were not able to do. I'm not
saying that our solution was for everyone, but am pointing out that
relying on just one solution is not an answer either.
 
P

PA20Pilot

Hi again,

........First a transient establishes a complete electrical path to earth
ground. Then something fails in that path.

Interesting, hadn't thought about it that way before.

It's easy to see how the cases of equipment is grounded by the
recepticals, but how can the hot legs be grounded too? My neighbor has
one of the Pepsi sized cans installed in his breaker box that's supposed
to fail if struck, is that the answer? I've noticed the power company
runs ground wires down their poles around here just about every third
pole. They take grounding serious.

---==X={}=X==---


Jim Self
AVIATION ANIMATION, the internet's largest depository.
http://avanimation.avsupport.com

Your only internet source for spiral staircase plans.
http://jself.com/stair/Stair.htm

Experimental Aircraft Association (EAA)
Technical Counselor
 
P

PA20Pilot

Hi Leythos,

Civility, wonderful!

......I completely disagree with the assumption that UPS's don't protect
devices from surges.

Maybe they do, but according to Tom, not by design, at least they're not
advertised as such.

---==X={}=X==---


Jim Self
AVIATION ANIMATION, the internet's largest depository.
http://avanimation.avsupport.com

Your only internet source for spiral staircase plans.
http://jself.com/stair/Stair.htm

Experimental Aircraft Association (EAA)
Technical Counselor
 
W

w_tom

You tell me what that plug-in protector is protecting from.
You tell me what the manufacturer claims to protect from.
Your assumption that any plug-in protector is additional
protection is based upon what?

Earth ground is the protection from typically destructive
transients. The protector is only a connection to
protection. Where is the other earth ground that a plug-in
protector connects to? Without some dedicated earth ground,
then a plug-in protector provides nothing additional.

Do not assume a protector is protection. Some circuits need
no protector to be protected. A wire connects to protection
(ie cable TV). The protector simply replaces that wire when a
direct wire connection (ie to AC hot wire) cannot be
installed.

A 4KVA UPS is completely different from a plug-in UPS. The
plug-in UPS only claims blackout and brownout protection. The
4KVA building unit costs more than a few $100 because it has
other power functions. What do the specification numbers
say? For example, does it list THD? Just another
embarrassing number that some plug-in UPS manufacturer will
'forget' to mention. Another function probably listed on that
4KVA UPS.

To assume that 4KVA UPS is at all similar to the APC UPS is
like saying a shark and a trout are same.

A UPS must be properly grounded. I don't understand what
you mean by "One that grounds, etc. were not able to do." But
a building UPS would include 'whole house' protection; would
have superior earthing.

Even a single ground stake provides major earthing
improvement. Earthing and how earthing is connected being the
protection. A 4KVA building UPS would have superior earthing
due to its location and other requirements. Earthing defines
the protection. The plug-in UPS has all but no earth ground.
Just another reason why a building UPS also does what a 'whole
house' protectors does; provides superior transient protection
even to computers with unacceptable power supplies.

The 4KVA building UPS would 'fix' problems created by
'defective by design' power supplies that are missing
essential functions. These are not just functions defined by
Intel. Industry standard functions that existed long before
Intel wrote their spec. Specs that other companies such as
AMD, IBM, Motorola, Conexant, TI, EPRI PEAC Corp., Compaq,
Central Hudson Power, Toshiba, TXU Electric, National
Semiconductor, Sony, Public Service of New Mexico, Gateway,
HP, Duke Power, and Dell demand. Disparaging Intel only
because they too make demands often not found in clone power
supplies is myopic. Those 'defective by design' power
supplies so often found in clone computers tend to violate
numerous standards. Then others fix that problem with
additional equipment.

When the computer suffers hardware failure, too often others
blame power rather than a bean counter who assembled that
computer. Then the naive 'feel' additional protection is
necessary.

When hardware fails, one good starting point is the power
supply. If AC mains power problems are damaging disk drives,
then a power supply missing essential functions is a most
likely suspect. Disk drive should never be damaged by
anything that passes through a minimally acceptable power
supply.

Meanwhile, do you remember the number for those other three
vacuum tubes used in 1950s AM radios?
 
W

w_tom

Equipment case is safety grounded. This is not electrically
same as earth ground. For safety grounding, wire resistance
is relevant. 50 feet of 12 AWG wire will be less than 0.2
ohms resistance. That same 12 AWG wire could be 130 ohms
impedance to transients. Wire length and other conditions
such as sharp wire bends being a critical parameter when
discussing transient protection.

A wall receptacle is typically too far to earth even a
trivial 100 amp transient. The critical number is 'less than
10 feet'. Even 6 foot of power cord on a protector
compromises the protector - power strip or UPS. We install a
'whole house' protector to make a 'less than 10 foot'
connection to earth ground. The protector is not protection.
Protector is simply a connection from utility wire to
protection - earth ground.

A protector that 'fails if struck' is not acceptable. That
'Pepsi can' protector is probably same as properly sized
protectors sold in Home Depot and Lowes. Both effective and
ineffective protectors were listed in that other discussion
entitled "Lightning and computer?" on 20 April 2005 (in a
first reply to Susan). Further information on what is and is
not effective earthing was detailed in same discussion (that
begins "Impedance has little to do with the size...") on 21
April 2005. A response after Leythos claimed he has an EE
degree but did not understand resistance and impedance.

Again, if any protector fails during a transient, then the
protector was grossly undersized - insufficient joules. Look
at joules number for a plug-in UPS. Also undersized.
Effective protectors degrade - do not fail. Manufacturers
charts joules verses number of transients in data sheets. The
electrical conditions that cause total failure? Not even on
those charts because such failure is not a normal nor
acceptable event. Joules: what ineffective protectors have
too few of and therefore fail catastrophically. A properly
sized protector has sufficient joules to earth the direct
strike and remain functional.

The power company ground wire is your primary protection.
An inspection of that primary protector is demonstrated in:
http://www.tvtower.com/fpl.html

Your building earth ground and associated connections (ie
the protector) are secondary protection.

Much reading and technical information was posted in that
other discussion. A long list of manufacturer app notes,
industry professional
experiences, utility recommendations, NIST figure, and other
underlying concepts (probably a full days worth of reading)
was posted in reply to H. W. Stockman in
alt.comp.periphs.mainboard.asus on 30 Mar 2005 entitled "UPS
unit needed for the P4C800E-Deluxe" at
http://makeashorterlink.com/?X61C23DCA
 
W

w_tom

'UPS connected' devices were protected just like the
"missing UPS" devices were protected. What protected Leythos'
smoke detector, bathroom and kitchen GFCIs, clock radio, and
dishwasher. Clearly Leythos forgot to mention he also
installed "missing UPS" protectors. How else can Leythos
explain those other undamaged electronics not connected to a
UPS?

This was explained previously in that discussion entitled
"Lightning and computer?". Back then, I was replying to
someone who claimed to have an EE degree. Now that we know
Leythos lied about the EE degree, well then of course he
doesn't understand why some things were damaged and others
were not. Of course he never understood the answer. The
technical education he claimed did not exist.

In the meantime, a "missing UPS" protected the TV while an
adjacent VCR was damaged. That example also answered Leythos
question a second time. A more technical answer was posted
in "Lightning and computer?". But that means Leythos must
learn some basic electrical concepts. Clearly, Leythos wants
to argue rather than learn.

Leythos again misrepresents what was posted only to keep
arguing. He again posts:
... as tom does, that a UPS provides no surge protection, ...

When Leythos misrepresented what I posted above on 28 April
2005, the reply (that Leythos still does not comprehend) was:
Leythos should read with care. The UPS claims to protect
from types of transients that are typically not destructive.
UPS does claim to protect from something. But anything
effective inside that UPS is already accomplished inside the
adjacent appliance. Somehow, Leythos dumbs this down and
distorts reality into "a UPS does NOT protect anything."

Again Leythos. Remember those numerical specification you
could not provide after how many requests? Fifteen? The UPS
does claim protection from transients. Provide the numbers.
It does not claim to protect from the type of transient that
typically damages electronics. Again, I am only reposting
what you intentionally distort to keep arguing.

So for the sixteenth time, Leythos: post the manufacturer's
spec for that protection you claim is provided by an APC UPS.
You cannot? Why? Numbers are what junk scientists fear.
But prove me wrong. Show me. Show me this manufacturer spec
that even claims to protect from each type of transient. Show
me where the manufacture even claims to do as you claim - in
that same post where you also claim an EE degree. Show me the
numbers - rather than repeatedly misrepresent what I have
posted and misrepresent what a UPS manufacturer claims.

Leythos - where are your numbers? Or do you just know these
things.
 
L

Leythos

On Mon, 02 May 2005 06:25:11 -0400, w_tom wrote:
[snip]
Leythos - where are your numbers? Or do you just know these
things.

How can I show you any more NUMBERS than many devices connected to UPS's
that were protected and many devices that were connected to the same
electrical outlets that were not protected at the same time?

Here is a typical UPS that we would install in a facility for workstations:
http://www.apc.com/resource/include/techspec_index.cfm?base_sku=BR1000

Here is another:
http://www.apc.com/resource/include/techspec_index.cfm?base_sku=SU700J

Here is a typical Server UPS:
http://www.apc.com/resource/include/techspec_index.cfm?base_sku=SU2200RMXLNET

Now, you have the links to typical UPS units that we would/have installed,
you have my explanation of devices being protected while others failed
that were not on the UPS.

Explain it away if you can.
 
L

Leythos

A 4KVA UPS is completely different from a plug-in UPS. The
plug-in UPS only claims blackout and brownout protection. The 4KVA
building unit costs more than a few $100 because it has other power
functions. What do the specification numbers say? For example, does
it list THD? Just another embarrassing number that some plug-in UPS
manufacturer will 'forget' to mention. Another function probably listed
on that 4KVA UPS.

To assume that 4KVA UPS is at all similar to the APC UPS is
like saying a shark and a trout are same.

Does this mean that your saying that a UPS (which the APC Smart-UPS XL)
line (that covers the 4KVA range, really does provide protection and that
you were wrong or misstated yourself?

So, it seems that you're suggesting that if I purchase a UPS, that it
doesn't protect me, but if I purchase a better UPS (since you had no idea
which ones we use, and is still a UPS) that you're changing your story
now?

If you look on most UPS sites, there are clear ranges of features, for you
to make a blanket assumption that a 4KVA UPS is not similar to a APC UPS
is ludicrous, as we buy a lot of 2+KVA UPS's that PLUG-IN to wall outlets
from APC vendors.

It seems to me you're a little wishy-washy on this, either keep your story
that UPS's don't protect us or change your statement to reflect that a
select class of UPS's don't protect us while also mentioning that some XXX
UPS's do protect devices.
 
M

Michael W. Ryder

w_tom said:
You tell me what that plug-in protector is protecting from.
You tell me what the manufacturer claims to protect from.
Your assumption that any plug-in protector is additional
protection is based upon what?

Earth ground is the protection from typically destructive
transients. The protector is only a connection to
protection. Where is the other earth ground that a plug-in
protector connects to? Without some dedicated earth ground,
then a plug-in protector provides nothing additional.

Do not assume a protector is protection. Some circuits need
no protector to be protected. A wire connects to protection
(ie cable TV). The protector simply replaces that wire when a
direct wire connection (ie to AC hot wire) cannot be
installed.

A 4KVA UPS is completely different from a plug-in UPS. The
plug-in UPS only claims blackout and brownout protection. The
4KVA building unit costs more than a few $100 because it has
other power functions. What do the specification numbers
say? For example, does it list THD? Just another
embarrassing number that some plug-in UPS manufacturer will
'forget' to mention. Another function probably listed on that
4KVA UPS.

The 4KVA unit I am talking about is a plug in UPS. It plugs into a wall
socket (granted it uses a locking plug) and the equipment plugs into it.

To assume that 4KVA UPS is at all similar to the APC UPS is
like saying a shark and a trout are same.
Which APC unit are you talking about? They make a very large number of
different ones.

A UPS must be properly grounded. I don't understand what
you mean by "One that grounds, etc. were not able to do." But
a building UPS would include 'whole house' protection; would
have superior earthing.

The unit I am talking about is not a building UPS, but only used for the
computer and the drives attached to it. ALL of our wiring was new and
up to or above the latest codes. The computer room was on a separate
wire from the power company and was tested multiple times for faults.
Until we installed the UPS to "clean" up the power we were having hard
drive failures at least once a month. We had the factory reps for the
company check out the drives and could find no problems other than the
transients in the power.

Even a single ground stake provides major earthing
improvement. Earthing and how earthing is connected being the
protection. A 4KVA building UPS would have superior earthing
due to its location and other requirements. Earthing defines
the protection. The plug-in UPS has all but no earth ground.
Just another reason why a building UPS also does what a 'whole
house' protectors does; provides superior transient protection
even to computers with unacceptable power supplies.

Again, the UPS was plugged into the wall and the power circuit was
properly "grounded". The problem is that a ground can NOT remove
"noise" from a signal.
 
W

w_tom

What were numbers for transient on AC mains? Transient peak
voltage and transient width. Were those transients observed
on output of the disk drive's power supply - or did the
company rep check?

If transients are "noise", then the APC UPS (as Leythos
defines) also would not eliminate that noise. Such noise is
suppose to be eliminated by the disk drive's power supply.
That would be noise; not surges.

What is the manufacturer (or type) of this 4KVA UPS?
 
W

w_tom

Had Leythos understood what was posted, then weeks ago he
knew there are many different types of UPSes. Serious UPSes
do provide effective transient protection. They specifically
state such protection in their numerical specs. But a $100
UPS is nothing more than stripped down UPS to only protect
data from blackouts and brownouts; only for a very limited
time.

Does a plug-in UPS do same as the UPS inside telephone
switching stations? Apparently so from what Leythos has
posted. But then he never needed any specification to know
better.
 
W

w_tom

Leythos finally provides the manufacturer numbers (as
requested almost 20 times). The surge protector is:
Surge Protection and Filtering
Surge energy rating 420 Joules
Filtering Full time multi-pole noise filtering : 5% IEEE surge
let-through : zero clamping response time : meets UL 1449
Dataline protection RJ-45 Modem/Fax protection (two wire
single line) , RJ45 10/100 Base-T Ethernet protection

So the UPS surge protector is same circuit in power strip
protectors - a paltry 420 joules. Let's assume that all
joules are installed on AC mains - give it the best possible
protection circuit. That means UPS provides as little as 140
joules and never more than 280 joules of protection. Using
MOV manufacturer charts to compare this protection to a
minimally sized 'whole house' protector for a classic 30
usecond, 1000 amp transient. Above 420 joule protector would
be sufficient for 2 transients. The minimally sized 'whole
house' protector (that costs two to five times less money) is
rated for about 200 or 300 of these same transients. For
larger transients, the UPS would catastrophically fail on the
first transient whereas the 'whole house' protector would
remain functional. That is the difference between a grossly
undersized protector and a minimally sufficient ('whole
house') protector. The latter actually provides protection.
The former only speculates protection.

As repeatedly stated and now demonstrated by numbers, the
protection inside that plug-in UPS is undersized - too few
joules. Same protector circuit provided by $0.10 component
inside power strip protectors.

Filtering is for AC noise elimination as required by FCC.
If similar to other UPSes that were disassembled, the
filtering is only on control electronics; not connected to
load (computer). IOW no filtering is applied to the load
(computer) as is routine in so many other plug-in UPSes. AC
mains connects directly to load when not in battery backup
mode. No filtering to the computer as others could so easily
assume based upon how the spec is deceptively worded.

Zero Clamping response time? I assume they claim 0
nanoseconds response time which what previous APC specs stated
and which is what that same circuit does inside power strip
protectors. Again, the UPS has same protector circuit found
in power strip protectors using same $0.10 components.

UL1449 does not mean the surge protector works. UL1449
(which is obsolete - must be UL1449 2nd edition) only says the
UPS will not harm humans. Protector circuit can completely
disintegrate on first transient. That protector circuit still
qualifies for UL1449 approval as long as the catastrophic
failure does not threaten human life.

What dataline protection? Where is one number that claims
such protection? For example, is capacitance low enough for
DSL lines? Again this APC spec intentionally shorts us of
useful information so that the consumer will 'assume' things
such as effective surge protection. What are they trying to
hide?

More numbers they don't provide for good reason. Where is
the THD number? A high THD means the UPS output could damage
small electric motors because the output has spikes and
harmonics. Best when selling a 'computer grade' UPS, do not
provide much information. Computers already have internal
protection that makes those harmonics and spikes irrelevant.

APC once provided better specs for UPSes.
Normal mode clamping response time 0 ns, instantaneous
Normal mode surge voltage let through <5% of test peak
voltage when subjected to IEEE 587 Cat. A 6kVA test

IOW they admit to protection from only one type of transient
- normal mode. So that you don't ask embarrassing questions,
latest specs forget to even mention various transient modes.
Better to not discuss what is not really provided. A plug-in
UPS does not even claim to protect from the other typically
destructive type of transients. Now they ignore many types of
transients exist - including a destructive type that APC never
claimed to protect from. Numbers provided by Leythos still
don't claim to protect from typically destructive types of
transients.

The numbers demonstrate what kind of protection is
provided. First, undersized - too few joules. Second, it
connects the load directly to AC mains. Where is the
isolation or filtering? No such isolation exists. Third,
whereas APC once claimed protection from normal mode
transients (those that require no earth ground), they now
avoid any mention of the different types of transients. This
so that the consumer will 'feel' it protects from all types of
transients.

APC does not even provide THD numbers for Outputs. Numbers
they forget to provide are most damning. If we saw high THD
numbers, then it would be obvious there is no filtering to the
load (computer). APC does not even state type of protection
on telephone and data lines. Protection that could actually
degrade DSL service. Better to not say anything - provide no
numbers. Again, more embarrassing numbers they would rather
forget to provide.

Leythos recommended this UPS for surge protection that even
its manufacturer does not claim to provide. With or without
that UPS, his computer would have probably survived just
fine. The "missing UPS" would have protected just as good as
his existing UPS since computers already have effective
internal protection. The computer would have protected
itself.

Why the term 'computer grade'? Because other appliances can
even be damaged by a UPS output in battery backup mode.
Output that can damage other appliances will not overwhelm
internal computer protection. That is how resilient computers
can be. IOW his computer connected directly to AC mains via a
UPS protected itself. Instead, Leythos credits the UPS. The
UPS manufacturer tells us much about the transient protection
provided. Protection that pragmatically does not exist.
Still, Leythos know more than the manufacturer<g>.

First demand the numbers. That APC UPS provides data
protection from blackouts and brownouts. It does not claim
protection from the typically destructive transients. It
claims to protect like a power strip protector with the same
'too few' joules. And it provides misleading statements (ie
filtering) so that one (ie Leythos) will actually believe the
filter exists and therefore provides protection.

The UPS as a protector is only as effective as its earth
ground. That UPS has no effective earthing. Therefore its
own specs confirm the obvious. That UPS provides ineffective
protection that is also grossly undersized - too few joules.
Just a short description of what that UPS does not provide is
in direct contradiction to what Leythos claims. These are
damning numbers - especially those too few joules. Leythos
recommended protection that is mostly mythical.
 
L

Leythos

Had Leythos understood what was posted, then weeks ago he
knew there are many different types of UPSes. Serious UPSes do provide
effective transient protection. They specifically state such protection
in their numerical specs. But a $100 UPS is nothing more than stripped
down UPS to only protect data from blackouts and brownouts; only for a
very limited time.

Nice to see you changing your story to match the real-world now. Since
there was never a mention of a UPS model, and since you said that UPS's
don't protect from transients, you appear to be changing your story now.
Does a plug-in UPS do same as the UPS inside telephone
switching stations? Apparently so from what Leythos has posted. But
then he never needed any specification to know better.

I've never suggested anything about a telco loc, only what I've seen first
hand when using APC UPS units at the desktop and in our server farms.

I see you have to post diversionary comments now that you've been shown to
be wrong about UPS's not providing transient protection.
 
L

Leythos

Leythos recommended this UPS for surge protection that even
its manufacturer does not claim to provide. With or without that UPS,
his computer would have probably survived just fine. The "missing UPS"
would have protected just as good as his existing UPS since computers
already have effective internal protection. The computer would have
protected itself.

I never suggested that the vendors information stated anything, only that
I've seen it protect devices while others not connected to the UPS were
damaged. Can I explain it, no, don't even care too. I've seen it enough
times that I'm going to trust my judgment in purchasing a quality UPS
device for protection.
Why the term 'computer grade'? Because other appliances can
even be damaged by a UPS output in battery backup mode. Output that can
damage other appliances will not overwhelm internal computer protection.
That is how resilient computers can be. IOW his computer connected
directly to AC mains via a UPS protected itself. Instead, Leythos
credits the UPS. The UPS manufacturer tells us much about the transient
protection provided. Protection that pragmatically does not exist.
Still, Leythos know more than the manufacturer<g>.

Again, you use diversionary measures to keep from understanding what I've
stated - I've seen, time and time again, where a UPS has saved devices
from damage while other devices have been damaged that were not connected
to a UPS.

It seems clear that you are speculating that a UPS will not protect
devices as you are completely stuck on numbers and not real-world
experiences. It would be interesting to see you in a server/system farm
where there are only plug-in UPS devices protecting them and never having
a fault/failure related to electrical issues or transients. Come back
after you visit the reality of the real world.
 
M

Michael W. Ryder

w_tom said:
What were numbers for transient on AC mains? Transient peak
voltage and transient width. Were those transients observed
on output of the disk drive's power supply - or did the
company rep check?

I was not directly involved in the measuring of the power problems as I
had my work to do. I do know that the power was Much dirtier than that
in other parts of town due to the number of Truck repair firms, etc. in
the area. Yes, the power was metered going into the disk drives and at
its output. We often had recorders running for weeks on the lines.

If transients are "noise", then the APC UPS (as Leythos
defines) also would not eliminate that noise. Such noise is
suppose to be eliminated by the disk drive's power supply.
That would be noise; not surges.

What is the manufacturer (or type) of this 4KVA UPS?

The UPS we purchased was from Exide. We have since replaced it with a
APC Smart UPS XL after the Exide finally died. Plus the APC unit is
much smaller and lighter.
 
L

Leythos

I was not directly involved in the measuring of the power problems as I
had my work to do. I do know that the power was Much dirtier than that
in other parts of town due to the number of Truck repair firms, etc. in
the area. Yes, the power was metered going into the disk drives and at
its output. We often had recorders running for weeks on the lines.

We experienced the same issues in many industrial locations around the USA
and Argentina. In all cases you could watch recorders/scopes connected to
the lines and see the problems, you could watch system problems increase
during electrical storms, track their history in maintenance actions,
etc...

Installation of UPS units at the desktop level, server level, and at the
control systems eliminated much downtime and also eliminated all of the
faults that indicated a major power spike (blown power supply units along
with damaged boards, etc...). Heck, we even installed small UPS units
inside scale systems in order to eliminate batching problems on remote
lines.

Tom can spout his numbers all he wants, and I have never disputed them,
but he's never going to convenience me that a UPS does not protect
equipment from transients or surges, and you get the added benefit of
power loss/surge regulation/protection too.
 

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. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads

Living With a Computer 3

Top