office wiring

[[email protected]]

The following web page:

http://www.joelonsoftware.com/articles/BionicOffice.html

describes an office setup for software developers. Well into the page it
tells about how the offices have 20 outlets powered through UPSes located
in the server room.

What I'd like to know is what code compliance wiring issues might be
involved in connecting the office outlets back to the UPSes. I would
think that whenever power emerges from a separately derived system and
then becomes building wiring again, it must meet all code requirements
including proper overcurrent protection (not the supplementary type of
protection found in the UPS units).

 

[[email protected]]

In alt.engineering.electrical [email protected] wrote:

| "Power. Every desk has twenty, that's right, twenty outlets. Four of
| them are colored orange and have uninterruptible power coming off of a
| UPS in the server closet, so you don't need a UPS in every office.
| The outlets are right below desk level in a special trough which runs
| the entire length of the desk, about six inches deep and six inches
| wide. The trough is a place to hide all your cables neatly and has a
| handy cover which blends in with the desk."
|
| Comment:
| Flexible cords must be exposed and cannot be run in troughs as
| described in the bionic office. There is a reason for this. Several

Would it really be classified as a trough if it's just an area behind
the desk where the receptacles are? When I envisioned from this is a
space for the cords to lay out of the way, effectively protected from
most causes of damage. Basically the desk would have a platform for
the cords to lay on, instead of dangling to the floor. If the cords
are reachable from the under-desk area when dangling, there is now the
added hazard of feet causing either direct damage, or damage due to
be ing pulled by entanglement with feet (shoes on, preventing the
sensation of knowing that the cord is entangled).

So I would think that what still might officially be a trough could
in this case actually be safer by providing added protection to the
cords.

 

[Charles Perry]

The following web page:

http://www.joelonsoftware.com/articles/BionicOffice.html

describes an office setup for software developers. Well into the page it
tells about how the offices have 20 outlets powered through UPSes located
in the server room.

What I'd like to know is what code compliance wiring issues might be
involved in connecting the office outlets back to the UPSes. I would
think that whenever power emerges from a separately derived system and
then becomes building wiring again, it must meet all code requirements
including proper overcurrent protection (not the supplementary type of
protection found in the UPS units).

A UPS is not a separately derived source. The neutral passes straight
through. If the UPS is large, as I expect this one is, it will have output
breakers. If it is really large, which is likely, it probably feeds a PDU
(power distribution unit) which is basically a breaker panel for a UPS.

Charles Perry P.E.

 

[Andrew Gabriel]

A UPS is not a separately derived source. The neutral passes straight
through.

Interesting -- in the UK, you are not permitted to assume any
supply side connections remain intact, i.e. the UPS can't assume
that its neutral or ground connections to the supply still work.
(One reason it might come on is if someone were to cut through
all of them.) UPS's usually have provision for a separate ground
connection for this reason.

If the UPS is large, as I expect this one is, it will have output
breakers. If it is really large, which is likely, it probably feeds a PDU
(power distribution unit) which is basically a breaker panel for a UPS.

People sometimes overlook a UPS's [in]ability to supply enough
current to cause a breaker to trip in a fault condition.

 

[[email protected]]

|
| |> The following web page:
|>
|> http://www.joelonsoftware.com/articles/BionicOffice.html
|>
|> describes an office setup for software developers. Well into the page it
|> tells about how the offices have 20 outlets powered through UPSes located
|> in the server room.
|>
|> What I'd like to know is what code compliance wiring issues might be
|> involved in connecting the office outlets back to the UPSes. I would
|> think that whenever power emerges from a separately derived system and
|> then becomes building wiring again, it must meet all code requirements
|> including proper overcurrent protection (not the supplementary type of
|> protection found in the UPS units).
|
| A UPS is not a separately derived source. The neutral passes straight
| through. If the UPS is large, as I expect this one is, it will have output
| breakers. If it is really large, which is likely, it probably feeds a PDU
| (power distribution unit) which is basically a breaker panel for a UPS.

This is not true with all UPSes. I've talked with tech support engineers
at a couple UPS makers who indicate that this aspect (neutral passing
through) varies by model. Specifically they said ALL 230 volt units do
NOT pass neutral through because of the variations of wiring that are
standard in the world. But even many 120 volt units do not pass neutral
through.

And then there's the matter of when there is no power. It certainly is a
separately derived system if the source supply is cut off.

 

[[email protected]]

| People sometimes overlook a UPS's [in]ability to supply enough
| current to cause a breaker to trip in a fault condition.

Having seem this happen a couple times, at least I know the internal
breaker they provide can trip on the current available. In one case a
sysadmin working for me had rewired a power switch incorrectly on an
older AT-style power supply, and short circuited the power which was
being fed from a UPS. It tripped virtually instantly. The UPS was a
10 kVA three phase 120/208 version, so it probably had the umph to do
the job. I then made a new rule for the shop: all rewired computers
are to be first tested on a non-UPS circuit.

 

[Charles Perry]

In alt.engineering.electrical Charles Perry <[email protected]>
wrote:
|
| |> The following web page:
|>
|> http://www.joelonsoftware.com/articles/BionicOffice.html
|>
|> describes an office setup for software developers. Well into the page
it
|> tells about how the offices have 20 outlets powered through UPSes
located
|> in the server room.
|>
|> What I'd like to know is what code compliance wiring issues might be
|> involved in connecting the office outlets back to the UPSes. I would
|> think that whenever power emerges from a separately derived system and
|> then becomes building wiring again, it must meet all code requirements
|> including proper overcurrent protection (not the supplementary type of
|> protection found in the UPS units).
|
| A UPS is not a separately derived source. The neutral passes straight
| through. If the UPS is large, as I expect this one is, it will have
output
| breakers. If it is really large, which is likely, it probably feeds a
PDU
| (power distribution unit) which is basically a breaker panel for a UPS.

This is not true with all UPSes. I've talked with tech support engineers
at a couple UPS makers who indicate that this aspect (neutral passing
through) varies by model. Specifically they said ALL 230 volt units do
NOT pass neutral through because of the variations of wiring that are
standard in the world. But even many 120 volt units do not pass neutral
through.

Powerware 9125, neutral straight through. I have a lab full of european
spec UPS that pass the neutral straight through. I have them from several
manufacturers and in sizes from 2kVA up to 50kVA, single phase and three
phase. Some do have a contactor to open the neutral when the UPS is turned
off.

Charles Perry P.E.

 

[[email protected]]

|
| |> In alt.engineering.electrical Charles Perry <[email protected]>
|> wrote:
|> |
|> | |> |> The following web page:
|> |>
|> |> http://www.joelonsoftware.com/articles/BionicOffice.html
|> |>
|> |> describes an office setup for software developers. Well into the page
|> it
|> |> tells about how the offices have 20 outlets powered through UPSes
|> located
|> |> in the server room.
|> |>
|> |> What I'd like to know is what code compliance wiring issues might be
|> |> involved in connecting the office outlets back to the UPSes. I would
|> |> think that whenever power emerges from a separately derived system and
|> |> then becomes building wiring again, it must meet all code requirements
|> |> including proper overcurrent protection (not the supplementary type of
|> |> protection found in the UPS units).
|> |
|> | A UPS is not a separately derived source. The neutral passes straight
|> | through. If the UPS is large, as I expect this one is, it will have
|> output
|> | breakers. If it is really large, which is likely, it probably feeds a
|> PDU
|> | (power distribution unit) which is basically a breaker panel for a UPS.
|>
|> This is not true with all UPSes. I've talked with tech support engineers
|> at a couple UPS makers who indicate that this aspect (neutral passing
|> through) varies by model. Specifically they said ALL 230 volt units do
|> NOT pass neutral through because of the variations of wiring that are
|> standard in the world. But even many 120 volt units do not pass neutral
|> through.
|
| Powerware 9125, neutral straight through. I have a lab full of european
| spec UPS that pass the neutral straight through. I have them from several
| manufacturers and in sizes from 2kVA up to 50kVA, single phase and three
| phase. Some do have a contactor to open the neutral when the UPS is turned
| off.

Is that a 120 volt or 240 volt model?

So _why_ does the neutral need to be passed through on models that are
dual conversion continuous online? Since not all models do this, there
must be some reason that varies depending on targeted uses for various
models.

 

[Matthew Beasley]

In alt.engineering.electrical Charles Perry <[email protected]>
wrote:
|
| |> In alt.engineering.electrical Charles Perry
<[email protected]>
|> wrote:
|> |
|> | |> |> The following web page:
|> |>
|> |> http://www.joelonsoftware.com/articles/BionicOffice.html
|> |>
|> |> describes an office setup for software developers. Well into the
page
|> it
|> |> tells about how the offices have 20 outlets powered through UPSes
|> located
|> |> in the server room.
|> |>
|> |> What I'd like to know is what code compliance wiring issues might be
|> |> involved in connecting the office outlets back to the UPSes. I
would
|> |> think that whenever power emerges from a separately derived system
and
|> |> then becomes building wiring again, it must meet all code
requirements
|> |> including proper overcurrent protection (not the supplementary type
of
|> |> protection found in the UPS units).
|> |
|> | A UPS is not a separately derived source. The neutral passes
straight
|> | through. If the UPS is large, as I expect this one is, it will have
|> output
|> | breakers. If it is really large, which is likely, it probably feeds
a
|> PDU
|> | (power distribution unit) which is basically a breaker panel for a
UPS.
|>
|> This is not true with all UPSes. I've talked with tech support
engineers
|> at a couple UPS makers who indicate that this aspect (neutral passing
|> through) varies by model. Specifically they said ALL 230 volt units do
|> NOT pass neutral through because of the variations of wiring that are
|> standard in the world. But even many 120 volt units do not pass
neutral
|> through.
|
| Powerware 9125, neutral straight through. I have a lab full of european
| spec UPS that pass the neutral straight through. I have them from
several
| manufacturers and in sizes from 2kVA up to 50kVA, single phase and three
| phase. Some do have a contactor to open the neutral when the UPS is
turned
| off.

Is that a 120 volt or 240 volt model?

So _why_ does the neutral need to be passed through on models that are
dual conversion continuous online? Since not all models do this, there
must be some reason that varies depending on targeted uses for various
models.

It makes the bypass simpler since it only has to switch the line conductors.

 

[[email protected]]

|
| |> In alt.engineering.electrical Charles Perry <[email protected]>
|> wrote:

[...]

|> | Powerware 9125, neutral straight through. I have a lab full of european
|> | spec UPS that pass the neutral straight through. I have them from
|> several
|> | manufacturers and in sizes from 2kVA up to 50kVA, single phase and three
|> | phase. Some do have a contactor to open the neutral when the UPS is
|> turned
|> | off.
|>
|> Is that a 120 volt or 240 volt model?
|>
|> So _why_ does the neutral need to be passed through on models that are
|> dual conversion continuous online? Since not all models do this, there
|> must be some reason that varies depending on targeted uses for various
|> models.
|>
|
| It makes the bypass simpler since it only has to switch the line conductors.

You used plural. Are you assuming TWO line conductors?

In some locations, 220 to 240 volts is carried with two conductors where
each is about equal distant from ground potential, sometimes at 180 degrees
phase vector, sometimes at 120 degrees (and possibly even at 60 degrees in
an uncommon setup). In other locations, 220 to 240 volts is carried with
two conductors where one of the is grounded. Some of those locations may
have a correctly wired polarized outlet and plug which identifies which of
the two conductors is grounded. Others may have unpolarized plugs making
it possible that either is grounded.

If the UPS assumes one of the input lines is grounded, and makes a new
hot voltage relative to it, it could result in as much as 360 volts to
ground on the output in the case of a split 120/240 volt system coming
in, and it could result in as much as 480 volts to ground on the output
in the case of straight 240 volts in where it assumed the neutral on
wrong conductor, depending on the phase angle of input and produced
voltages.

I talked to a support engineer at one of the UPS companies, which I
beleive was Powerware, who told me that indeed most, but not all, 120
volt models did indeed pass the neutral through, and the 220-240 volt
models specifically did not due to the variety of electrical systems
around the world. Much of the world that normally utilizes 220 volts
still does so with a L-L connection, either from a single phase 110/220
volt system, or from a three phase 127/220 volt system where usually
just two of the phases are used. The only way to make a 220-240 volt
UPS work in all locations, even if North America is not considered, is
to NOT pass the neutral through, and switch all conductors in a bypass
switch.

 

[Charles Perry]

|
| |> In alt.engineering.electrical Charles Perry
<[email protected]>
|> wrote:

[...]

|> | Powerware 9125, neutral straight through. I have a lab full of
european
|> | spec UPS that pass the neutral straight through. I have them from
|> several
|> | manufacturers and in sizes from 2kVA up to 50kVA, single phase and
three
|> | phase. Some do have a contactor to open the neutral when the UPS is
|> turned
|> | off.
|>
|> Is that a 120 volt or 240 volt model?
|>
|> So _why_ does the neutral need to be passed through on models that are
|> dual conversion continuous online? Since not all models do this, there
|> must be some reason that varies depending on targeted uses for various
|> models.
|>
|
| It makes the bypass simpler since it only has to switch the line
conductors.

You used plural. Are you assuming TWO line conductors?

In some locations, 220 to 240 volts is carried with two conductors where
each is about equal distant from ground potential, sometimes at 180
degrees
phase vector, sometimes at 120 degrees (and possibly even at 60 degrees in
an uncommon setup). In other locations, 220 to 240 volts is carried with
two conductors where one of the is grounded. Some of those locations may
have a correctly wired polarized outlet and plug which identifies which of
the two conductors is grounded. Others may have unpolarized plugs making
it possible that either is grounded.

If the UPS assumes one of the input lines is grounded, and makes a new
hot voltage relative to it, it could result in as much as 360 volts to
ground on the output in the case of a split 120/240 volt system coming
in, and it could result in as much as 480 volts to ground on the output
in the case of straight 240 volts in where it assumed the neutral on
wrong conductor, depending on the phase angle of input and produced
voltages.

They pass it through, they do not assume it is grounded. Check european
equipment requirements. Unlike the US it is unusual to find any connection
between neutral and ground (such as filters).

I talked to a support engineer at one of the UPS companies, which I
beleive was Powerware, who told me that indeed most, but not all, 120
volt models did indeed pass the neutral through, and the 220-240 volt
models specifically did not due to the variety of electrical systems
around the world. Much of the world that normally utilizes 220 volts
still does so with a L-L connection, either from a single phase 110/220
volt system, or from a three phase 127/220 volt system where usually
just two of the phases are used. The only way to make a 220-240 volt
UPS work in all locations, even if North America is not considered, is
to NOT pass the neutral through, and switch all conductors in a bypass
switch.

I can say with 100% certainty that the Powerware 9125 European spec, 240V,
UPS passes the neutral straight through. We test them, open then up, etc.

Charles Perry P.E.

 

[[email protected]]

| They pass it through, they do not assume it is grounded. Check european
| equipment requirements. Unlike the US it is unusual to find any connection
| between neutral and ground (such as filters).

That's not the assumption I am talking about. They CANNOT pass it through
unless they are passing BOTH wires through.


|> I talked to a support engineer at one of the UPS companies, which I
|> beleive was Powerware, who told me that indeed most, but not all, 120
|> volt models did indeed pass the neutral through, and the 220-240 volt
|> models specifically did not due to the variety of electrical systems
|> around the world. Much of the world that normally utilizes 220 volts
|> still does so with a L-L connection, either from a single phase 110/220
|> volt system, or from a three phase 127/220 volt system where usually
|> just two of the phases are used. The only way to make a 220-240 volt
|> UPS work in all locations, even if North America is not considered, is
|> to NOT pass the neutral through, and switch all conductors in a bypass
|> switch.
|
| I can say with 100% certainty that the Powerware 9125 European spec, 240V,
| UPS passes the neutral straight through. We test them, open then up, etc.

Well, you are contradicting explicit information from the support engineer
who said: (1) the neutral is not and cannot be passed through because not
all electrical systems in the world are the same ... (2) the unit operates
correctly in continental Europe when the plug is rotated 180 degrees.

So what kind of tests did you do on this model? Did you test it with a
Schuko plug into a Schuko outlet in both way the plug could fit? Did you
test this with the same kind of electrical service in some rural parts
of Europe, including in Spain, Norway, and Russia, as well as in countries
like Saudi Arabia where the 220 volt mains has 2 hots at 127 volts to ground?

Are your test reports available to the manufacturer? How can I cite your
reports to them the next time I call them up to confirm this?