Re: the danger of higher power levels, voltage vs. current
Phil Scott
<phil-new...@ipal.net> wrote in message
news:ckk6n...@news3.newsguy.com...
> Which do you consider to be the more dangerous? Or at what
cutoff or
> ratio would one be more dangerous than the other?
>
> High voltage
> High available amperage
Depends on the scenario...high school physics classes have
static electricity demonstations of lighting a fluorescent
tube by holding one end in your hand and the other on a high
voltage static electricity generator.... its harmless. very
little amperage...all voltage.. watts are way down.
Add some amperage (millions of electrons) then you get
power..and thats dangerous...the more power the more
dangerous... the higher the voltage the farther it will arc.
50 amps at 12 volts will not arc very far..or have the force
behind it go through your body producing severe damage at
least.
50 amps at 110 volts will kill your... at 240 it will do it
faster...at 480 it will cook you well done.
>
> For example, a certain voltage, like 480 volts or 600 volts
has a
> certain level of danger to it. But at what point of
available fault
> current would you consider the amperage to exceed the
voltage as a
> hazard (such as the arc flash hazard)?
That was posted on last week a great reference with the charts
and numbers well worth a read.
in summary the closer you are to the line transformers the
more dangerous... the heavier the service feed wire the more
dangerous... the higher the rating of the transformer the more
dangerous... the slower the trip time on the fault interrupt
device (by tenths of a second) the more dangerous...linearly
a half second interupt time way over twice as dangerous as 1/4
second interupt time.
If your theoretical bolted dead short (the power company
provides that spec to you based on the transformer, feeds and
interrupt they have supplied) can draw say 10,000 amps and
the interupt is 1/4 of a second...and you get a path of
current though that device and through your body at only 1,000
amps and the device takes 6 seconds to interupt...it will have
interrupted long after your body meat has gone crispy.... and
is too tough to serve with breakfast.
So its tricky and a compromise...the safer it it the more
tripped breakers and this upsets customers... no tripped
breakers, heavy feeds, everything oversize...and its a dry
location.. no one hosing the place down.. its not real safe
but not the danger it would be in a marine environment with
salt water around.
< When would you raise the system
> voltage/impedance (at design time) in order to decrease the
available
> fault current to decrease the hazard?
At the line transformers, interrupt devices and feeders...and
you talk this over with the utility engineers... and in sizing
the feeders from the transformers to the main distribution
panel...and in selecting the main breaker to interupt as fast
as possible at its max amp rating, and as instantaeously as
possible on a dead short. Devices vary in their performance
you need to compare them.
Then you would leave the main distribution panel from a
breaker sized for the expected peak load of the panel it
supplies (with diversity figured in) .
for instance if you were feeding a 500 amp sub panel ....
which with diversity would never draw more than 200 amps...
you would supply that panel with a 250 amp main breaker.. with
as fast a trip time at 250 amps as possible.... and as fast
as possible dead short interrupt. In reality at that level you
are going to use whats available from the primary vendors etc.
GE, Seimens, Square D etc. they will all be close but you
still have choices.
all this adds impedance to the system that minimizes the arc
flash hazard or explosion and starts to get anally retentive
as you get way from from the main distribution panel to small
and smaller panels... the primary hazard is at the main, and
these line transformer and main distribution breaker and
feeder sizes etc
If I were in a salt water marine environment I would have many
smaller panels, low amperage not oversized on the wire...or
breakers... rather than several very large ones...this cuts
the feeder size and main breaker size to each sub panel and
makes the entire arrangement safer.
I designed and built a shipboard system with two Cat gen sets,
250 and 137 kw and a 150 ton refrigeration system a few years
ago. (blast freezing lobsters in the south pacific).. I sized
the gen sets so that only one refer package could be started
at a time.. a staged start... this cut the size of all the
wire and panels and reduced the hazard potentials... I also
stayed with 230 v 3 ph. running my cost of starter and
breakers back up, but reducing the hazards dramatically.
With motor panels you have to take into account start surges
as you know (locked rotor amps) and that will affect your
breaker size and interupt delay on overload characteristics
serving a particular motor, but should not be cumulative for
all the motors ...just full load amps but not locked rotor
amps... so you end up with an oversize main breaker for that
sub panel since all motors will not be starting at once..
these are factors to consider...not set in stone though...the
NEC addresses these issues extensively.
for instance if you have 8ea 50 amp 3ph 460v breakers in a sub
panel...and these serve a mix of motors, air conditioners and
say a trash compactor.. you know they will not all be starting
at once so if you want maximum safe operation you would not
size that panels main at 400 amps...though its very common.
you would size it closer to the diveristy load.. say 200 amps
max...and specify maybe 250 amps on the main breaker...with
wire feeding that panel sized the same..again the NEC defines
this pretty well...but the guess on diversity is largely up to
the electrical engineer...most size not for maximum safely but
to minimize main breaker trips from what Ive seen, that might
start changing as we get litigation on the issue.
Ive never done it but I might well start using several smaller
sub panels instead of one larger panel for a single group of
large motors.. just to get the wire size down to each one and
smaller main breakers at the sub panels to reduce flash
hazards...that would be a stretch in most dry locations...it
would not be in wet locations.
In the past with a more stable grid, and with smaller service
drops and a single small transformer.... hazards were still
somewhat contained... but as the utilities start paralleling
transfomers and beefing up the service feeders to satisfy a
customer ... the arc flash danger goes up by 10 to 1 or
so...massively...if its ratty old equipment no originally
arranged to handle that much potential you can have big
problems... if its a cannery with some guy from Peru hosing
the place down at night the risks are high...and way out the
top if its a 480v main and subs and thats often the case.
I also see grounding that meets code but not any more than
that and in many cases 5 times the bonding is required to be
actually safe... bonding goes ignored by many it
seems...because its not hot its almost considered peripheral
by some people..
Heavy transformer lighting for stages etc presents all sorts
of problems with neutrals and bonding.. I run 2 or 3x the
neutral wire for those applications. and heavy grounds at
multiple locations. (stage lighting people know this, most
other electricians cant fathom the range of issues that can
develop with these arrangements)
Location is important...if you are in moist area, or a marine
salt water environment all of these issues become a lot more
important as the path to ground becomes a lot more conductive
and available. Personnel proximity to the equipment is a big
issue...on ships for instance it can be tight... and its salt
water... fault interupts and wire sizing for impedance is even
more of an issue.
> A few years ago a friend told me about a place that had a
lot of lighting
> that was set up on low voltage because someone thought that
would be
> safer. And then because it was believed to be safe (perhaps
they used
> only 12 volts ... I never found out exactly what it was)
electricians
> would regularly work on it energized. That was, until one
of them
> somehow caused a bolted fault that didn't get cleared, and
caused the
> transformer powering it to eventually explode with the
ensuing fire
> destroying a building.
>
> Obviously voltage is unsafe to humans until the system
impedance gets
> high enough that the human is an insignificant part of the
circuit.
> But amperage is certainly unsafe when out of control as it
can cause
> serious arc and heat problems.
>
> Years ago I was considering building a house with low
voltage DC wiring
> throughout for almost all lighting. I'm now reconsidering
that, both
> due to my better understanding of fault conditions, as well
as the
> problem with clearing faults in DC (no zero crossing).
Combining DC
> with high available currents could be serious ... at least
that is my
> thinking and concern right now.
stay with AC use the commercially available components..you
dont have arc flash issues with home type systems, panels and
feeder sizes to any great extent...and those are very slight
on the branches...
12vac wire has to be ten times the size per watt than 120vac
wire its not economical and the components are not available
commonsly.. if you want transformer lighting in spots use
whats available, plug it into a 110 source... those small
transformers are dry and impedance protected they dont blow
up.
>
> --
> ------------------------------------------------------------
-----------------
> | Phil Howard KA9WGN | http://linuxhomepage.com/
http://ham.org/ |
> | (first name) at ipal.net | http://phil.ipal.org/
http://ka9wgn.ham.org/ |
> ------------------------------------------------------------
-----------------
Phil Scott
<phil-new...@ipal.net> wrote in message
news:ckmhp...@news2.newsguy.com...
> On Thu, 14 Oct 2004 00:48:51 GMT Phil Scott
<philsc...@sf.sbcglobal.net> wrote:
>
> |> For example, a certain voltage, like 480 volts or 600
volts
> | has a
> |> certain level of danger to it. But at what point of
> | available fault
> |> current would you consider the amperage to exceed the
> | voltage as a
> |> hazard (such as the arc flash hazard)?
> |
> | That was posted on last week a great reference with the
charts
> | and numbers well worth a read.
>
> even more insight than I had before.
:-)
>
>
> | So its tricky and a compromise...the safer it it the more
> | tripped breakers and this upsets customers... no tripped
> | breakers, heavy feeds, everything oversize...and its a dry
> | location.. no one hosing the place down.. its not real
safe
> | but not the danger it would be in a marine environment
with
> | salt water around.
>
> tripped breakers. And that comes down to, at least in the
> cases I can think of, the startup current surges. That
happens
> even on an incandescent light bulb. Power companies do want
> customers with big motors to use slow start controls to keep
> the surge of current down so it doesn't blink lights
elsewhere.
> But this can also allow the use of overcurrent protection
that
> can respond faster and at lower currents. It can make
things
> safer that way.
and
> the transformer may not be very much. So even a 20 amp
branch
> circuit could pull some serious amperage for an instant if
> the building is served by a big transformer (say 750 kVA).
> If the breakers fail, the branch circuit itself could well
> be vaporized while the guy who touched a screwdriver to the
> wrong spot gets his face cooked. I'd want fast trip
breakers,
> but also current limiting that won't prevent that fast trip.
> The article previously posted shows how this kind of choice
> is not easy to make because time is perhaps an even larger
> factor tha current.
What your remark here tells me is that with heavy
potentials involved we may need to think about standing 120
volt panels off alone and at a distance from the main service
and on lighter wire... and even think about running some 14
gage wire on circuits that serve single devices or controllers
in close proximity to the primary service..
Maybe the NEC will require signs as transformers get
paralleled onto antiquated existing building services. Floor
in the area painted red,,,with conductive paint so the victim
won't suffer.
"You are at the buildings primary electrical service
location... ground fault amperage is 10,000 times higher here,
on the same voltage than more distant from these feeders on
lighter conductors.... Utility company ground fault potential
here is EXPLOSIVE... rated _________ killowatts" Utility
company fills in the blank to match thier latest grid and
transformer upgrades in the area
> | I designed and built a shipboard system with two Cat gen
sets,
> | 250 and 137 kw and a 150 ton refrigeration system a few
years
> | ago. (blast freezing lobsters in the south pacific).. I
sized
> | the gen sets so that only one refer package could be
started
> | at a time.. a staged start... this cut the size of all the
> | wire and panels and reduced the hazard potentials... I
also
> | stayed with 230 v 3 ph. running my cost of starter and
> | breakers back up, but reducing the hazards dramatically.
>
> You'd have 400 volts between phases in that, right?
230 the way the gen set was wired.
>
>
> | for instance if you have 8ea 50 amp 3ph 460v breakers in a
sub
> | panel...and these serve a mix of motors, air conditioners
and
> | say a trash compactor.. you know they will not all be
starting
> | at once so if you want maximum safe operation you would
not
> | size that panels main at 400 amps...though its very
common.
> | you would size it closer to the diveristy load.. say 200
amps
> | max...and specify maybe 250 amps on the main
breaker...with
> | wire feeding that panel sized the same..again the NEC
defines
> | this pretty well...but the guess on diversity is largely
up to
> | the electrical engineer...most size not for maximum safely
but
> | to minimize main breaker trips from what Ive seen, that
might
> | start changing as we get litigation on the issue.
>
> And someone might also get upset if they do happen to try to
start
> too many loads at once and trip the breaker. You're damned
if you
> do and you're damned if you don't.
correct. Lawsuits will determine the eventual choices.
Starting them manually isnt typically a problem since LRA is
only for a second or two... it takes that long to reach the
next breaker. But in the event of a power failure, and power
coming back on...you would trip a lot of breakers... so...
the people just have to go reset them one at a time...that
works. I will going a little lighter in the future...more
but smaller sub panels... lighter wire to max divirsity load
especially in high risk locations.
>
>
> | In the past with a more stable grid, and with smaller
service
> | drops and a single small transformer.... hazards were
still
> | somewhat contained... but as the utilities start
paralleling
> | transfomers and beefing up the service feeders to satisfy
a
> | customer ... the arc flash danger goes up by 10 to 1 or
> | so...massively...if its ratty old equipment no originally
> | arranged to handle that much potential you can have big
> | problems... if its a cannery with some guy from Peru
hosing
> | the place down at night the risks are high...and way out
the
> | top if its a 480v main and subs and thats often the case.
>
> Or like network services in many downtown areas where they
have
> added more parallel transformers, while many buildings still
> have lower rated breakers or fuses. Imagine the utility
beefing
> up their network to 6000 amps, while many customers have
10000
> AIC rated 1950's era breakers.
Nothing will be done until Underwriters Lab gets on
it...along with the NEC...and that might happen in a few years
if there are a few more bad accidents or one spectacular one
.... and an attorney wins big on this combination of
oversights.
It seems the utility could be held reponsiible as most of
thier customers are not qualified and registered EE's. .. the
reasonable man priniciple dictates that the utility has the
responsibliity to advise its customers and see to it that the
system they are feeding is suitable... EE's will in order to
protect thier interests, liability etc will have to start
paying attention to these issues. So will building inspectors.
> | I also see grounding that meets code but not any more than
> | that and in many cases 5 times the bonding is required to
be
> | actually safe... bonding goes ignored by many it
> | seems...because its not hot its almost considered
peripheral
> | by some people..
>
> And it can fail without being noticed. Redundancy would be
> good, too.
>
>
> | stay with AC use the commercially available
components..you
> | dont have arc flash issues with home type systems, panels
and
> | feeder sizes to any great extent...and those are very
slight
> | on the branches...
>
> Data centers are, however, a different issue. And some are
> run on DC (telco ones tend to be). That 48 VDC stuff can
> deliver a load of amps.
Im not a telco guy... what size wire are we talking at the
vdc source?
typical range for what you do.
Phil Scott
>
>
> | 12vac wire has to be ten times the size per watt than
120vac
> | wire its not economical and the components are not
available
> | commonsly.. if you want transformer lighting in spots use
> | whats available, plug it into a 110 source... those small
> | transformers are dry and impedance protected they dont
blow
> | up.
>
> I'll probably leave the low voltage lights to places like
pools
> and such.
Brian
smoking.
<phil-new...@ipal.net> wrote in message news:ckk6n...@news3.newsguy.com...
Which do you consider to be the more dangerous? Or at what cutoff or
ratio would one be more dangerous than the other?
High voltage
High available amperage
For example, a certain voltage, like 480 volts or 690 volts has a
certain level of danger to it. But at what point of available fault
current would you consider the amperage to exceed the voltage as a
hazard (such as the arc flash hazard)? When would you raise the system
voltage/impedance (at design time) in order to decrease the available
fault current to decrease the hazard?
A few years ago a friend told me about a place that had a lot of lighting
that was set up on low voltage because someone thought that would be
safer. And then because it was believed to be safe (perhaps they used
only 12 volts ... I never found out exactly what it was) electricians
would regularly work on it energized. That was, until one of them
somehow caused a bolted fault that didn't get cleared, and caused the
transformer powering it to eventually explode with the ensuing fire
destroying a building.
Obviously voltage is unsafe to humans until the system impedance gets
high enough that the human is an insignificant part of the circuit.
But amperage is certainly unsafe when out of control as it can cause
serious arc and heat problems.
Years ago I was considering building a house with low voltage DC wiring
throughout for almost all lighting. I'm now reconsidering that, both
due to my better understanding of fault conditions, as well as the
problem with clearing faults in DC (no zero crossing). Combining DC
with high available currents could be serious ... at least that is my
thinking and concern right now.
--
-----------------------------------------------------------------------------
operator jay
<phil-new...@ipal.net> wrote in message
news:ckkua...@news3.newsguy.com...
>
> All the data centers I have designed were all based on 120/240 or
208Y/120
> for power source. Usually various kinds of UPS systems are
involved.
> I was talking with a friend who has a friend who runs a very large
data
> center. They have a pair of 500 kVA UPS systems running the place,
and
> behind that a single transfer switch between a fault tolerant
utility
> feed (unknown primary voltage) and a 1500 kW generator.
>
> Technicians NOT trained in power engineering would often be working
> around inside the cabinets of the computer and network systems.
They
> would be dealing with the secondary LV power. While 120 volts is
not
> much, that 500 kVA UPS is delivering 4166 amps and potentially as
much
4166? Would this not be a 3phase unit @ 1400A per phase?
> as 80 kA of fault current. That means the distribution panels will
> have to be rated to interrupt perhaps at least 100 kA (depending on
> lots of things such as actual UPS fault availability, series ratings
> of the whole system, etc).
>
> My tendency is to go with smaller UPS systems on the order of 10 kVA
> to 30 kVA at most. Enough for 1 to 3 cabinets. But these can be
had
> with 120 volt output and either 208Y/120 in, or 480Y/277 in. I have
> been thinking of starting to go with the latter. But at least with
> the smaller UPSes, fault currents within a data cabinet will be
small.
>
> Going from 208Y/120 to 480Y/277 raises the system impedance to 5.333
> times.
Agreed.
Fault currents would be 43.3% as much on the higher voltage
> feeds.
Agreed, but, available fault VA is the same either way. Can you tell
me whether arc hazard depends on fault VA or fault current?
I find myself wishing 600Y/346 was more common in the US as
> it is in Canada.
Yay Canada.
The thing is, I have heard of (and under my watch
> happened once, but no one was hurt ... just one guy's pride lost)
> computer technicians shorting something out much more often than
> getting themselves shocked or electrocuted. So it seems to me that
> voltage isn't as much of an issue as amperage, as long as the
voltage
> is within what can be safely handled through insulation (e.g. up to
> 600 volts perhaps).
Do you think you WOULD hear about the 120V shocks that technicians
might accidentally get? I wonder if they'd go unmentioned for
whatever reasons. I have not gotten into the arc hazard stuff at all.
So, of course I still instinctively think of lower voltages as safer,
and I still think of shock hazard being way more of a threat than
arcs. Maybe I'll learn otherwise.
j
TimPerry
<phil-new...@ipal.net> wrote in message
news:ckmfr...@news2.newsguy.com...
> On Wed, 13 Oct 2004 21:57:20 -0400 TimPerry <timp...@noaspamadelphia.net>
wrote:
>
> | <phil-new...@ipal.net> wrote in message
> | news:ckk6n...@news3.newsguy.com...
> |> Which do you consider to be the more dangerous? Or at what cutoff or
> |> ratio would one be more dangerous than the other?
> |>
> |> High voltage
> |> High available amperage
> |
as
> | it is only a theoretical concept. current however is the movement of
> | electrons and does supply energy to a load (i squared r).
>
> There is a real voltage at the source.
For example batteries will
> have a specific voltage based on their type of construction.
but their capacity is rated in ampere-hours not volt-hours
And
> I presume a given voltage will be present on generator windings under
> certain conditions. But beyond there, it's all about system impedance
> and what proportion of that impedance your load is (or what proportion
> of that impedance your body is in instances where that matters) which
> comes down to how the current dissipation is distributed (as r goes
> up, i goes down, so it's not like r can't have significant influence).
>
>
this does not prove that voltage exists as anything other then a convenient
theoretical concept.
> |> For example, a certain voltage, like 480 volts or 690 volts has a
> |> certain level of danger to it. But at what point of available fault
> |> current would you consider the amperage to exceed the voltage as a
> |> hazard (such as the arc flash hazard)? When would you raise the system
> |> voltage/impedance (at design time) in order to decrease the available
> |> fault current to decrease the hazard?
> |>
> | electronic parameters are set to accomplish a given task. then
> | insulation/shielding/grounding is employed to minimize hazard
>
> There's really a lot of technical flexibility in choosing a voltage.
> What influences that choice so heavily, however, is what you cannot
> readily change, such as the standard voltage that loads want to have.
> I'd rather go with lights on 277 volts at home instead of 120 volts,
> but the standard is 120 and that extensively influences the choices.
> Incandescent bulbs for 277 volts are hard to find, have less variety
> of wattage, and are more expensive. HID lights are what get used at
> that voltage for the most part. But I don't consider HID to be viable
> lighting for the home (I favor incandescent even over CF and LED).
>
>
> |> A few years ago a friend told me about a place that had a lot of
lighting
> |> that was set up on low voltage because someone thought that would be
> |> safer. And then because it was believed to be safe (perhaps they used
> |> only 12 volts ... I never found out exactly what it was) electricians
> |> would regularly work on it energized. That was, until one of them
> |> somehow caused a bolted fault that didn't get cleared, and caused the
> |> transformer powering it to eventually explode with the ensuing fire
> |> destroying a building.
> |>
> | seems to me that this could happen at any voltage level when a breaker
fails
> | to function (or when the system is improperly designed)
>
> What came across to me was that the people working on it apparently
> presumed it was "safe" due to the low voltage. I think that assumption
> just simply cannot be allowed to be made. Still, I'd use voltages like
> that for special situations like swimming pool lighting. But there
> you can at least put each light on its own transformer to be extreme.
>
how about phosphorescent strips? no heat, no moving parts
>
> |> Obviously voltage is unsafe to humans until the system impedance gets
> |> high enough that the human is an insignificant part of the circuit.
> |> But amperage is certainly unsafe when out of control as it can cause
> |> serious arc and heat problems.
> |>
> | you are trying to separate 2 factors that need to be considered
together.
>
> Both need to be considered dangerous in their own way. Circumstances do
> vary, and in some places high voltage works because the voltage hazard
> is less likely to be realized (where people cannot touch live wires).
high voltage works because it is the most efficient way to accomplish a
given task. safety precautions are then effected to minimize the hazard.
some guy didnt just come along one day and say "wow, lookie at those wires
up on those poles. i think it would be a great way to run high voltage
lines"
I
> think far too few people realize that high currents are just as dangerous
> as high voltage.
any combination of factors that manipulates energy can be "dangerous".
without voltage there is no current... just a bunch of unemployed electrons
hanging around waitihg for the government check to come.
It's just a matter of what is the best balance in some
> particular situation, and how close to that you can get to given limits
> like budgets and available equipment.
>
> How often do you see signs saying "Danger High Amperage".
>
i have one that says "danger 1,000,000 ohms"
i guess i could make one that says "danger: red herrings" :)
>
> |> Years ago I was considering building a house with low voltage DC wiring
> |> throughout for almost all lighting. I'm now reconsidering that, both
> |> due to my better understanding of fault conditions, as well as the
> |> problem with clearing faults in DC (no zero crossing). Combining DC
> |> with high available currents could be serious ... at least that is my
> |> thinking and concern right now.
> |>
> |
> | for the same amount of power loads the cost of larger copper wire and
heavy
> | duty switches may make LVDC lighting unattractive. might be nice for
> | emergency backup lighting... personally i just use a flashlight :)
>
> And Edison was distributing DC around the streets at utilization voltage.
> At least there weren't so many loads in that day that he would have needed
> the enormous amperage that would have to be present to supply the same few
> city blocks today at 120/240 utilization.
>
> I am looking at using small low power LED lights for emergency lighting.
> But the cost of a power supply for each one becomes the big issue.
>
>
the install cost of LED lighting is still outrageous when compared with
incandescent. the power savings can be 90% using LEDs.
the estimated (predicted) life of a LED FAA tower light flasher (obstruction
becon) is 7 years (if operated only in the nightime). unless it gets hit by
lightning.
i think the true answer to the question posed by the origional poster is:
both or neither equally.
*.*
<phil-new...@ipal.net> wrote in message
news:ckk6n...@news3.newsguy.com...
> Which do you consider to be the more dangerous? Or at what cutoff or
> ratio would one be more dangerous than the other?
>
> High voltage
> High available amperage
>
If the voltage is greater than about 50 volts or so.
Voltage is the key.
If you plot voltage vs amperage, the safe zone is below 50 volts, and below
10m amps in most all areas.
Phil Scott
"Don Kelly" <dh...@peeshaw.ca> wrote in message
news:W%ibd.95885$a41.58178@pd7tw2no...
>
>
>
> <phil-new...@ipal.net> wrote in message
> news:ckk6n...@news3.newsguy.com...
> > Which do you consider to be the more dangerous? Or at
what cutoff or
> > ratio would one be more dangerous than the other?
> >
> > High voltage
> > High available amperage
> >
volts has a
> > certain level of danger to it. But at what point of
available fault
> > current would you consider the amperage to exceed the
voltage as a
> > hazard (such as the arc flash hazard)? When would you
raise the system
> > voltage/impedance (at design time) in order to decrease
the available
> > fault current to decrease the hazard?
> >
lot of lighting
> > that was set up on low voltage because someone thought
that would be
> > safer. And then because it was believed to be safe
(perhaps they used
> > only 12 volts ... I never found out exactly what it was)
electricians
> > would regularly work on it energized. That was, until one
of them
> > somehow caused a bolted fault that didn't get cleared, and
caused the
> > transformer powering it to eventually explode with the
ensuing fire
> > destroying a building.
> >
impedance gets
> > high enough that the human is an insignificant part of the
circuit.
> > But amperage is certainly unsafe when out of control as it
can cause
> > serious arc and heat problems.
> >
voltage DC wiring
> > throughout for almost all lighting. I'm now reconsidering
that, both
> > due to my better understanding of fault conditions, as
well as the
> > problem with clearing faults in DC (no zero crossing).
Combining DC
> > with high available currents could be serious ... at least
that is my
> > thinking and concern right now.
>
> You are looking at two different conditions: High current
and the danger
> from arcs where the source impedance is relatively low as
the arc does not
> limit the current- the system does. As you indicate this
can occur at low
> voltages as well as at high voltages.
> The other is not unsafe "voltage" but unsafe currents in the
range of 100 or
> so ma (5% probability of fibrillation in an adult males is
at about 110ma.
> Obviously this then becomes a problem of both voltage and
contact resistance
> as in typical 60Hz systems, the system will not limit the
current at this
> level. The body will be the limiting factor. At higher
voltages there is
> often a better chance of living (not that I want to try
it) - possibly with
> burn damage- as the body's reaction can cause a person to be
thrown clear -
> heart stoppage may occur but spontaneous restarting often
occurs and has
> resulted with linemen surviving (the fall might do them
in.) - with
> fibrillation- you need help.
Ultra hard self forced caughing can stop fibrillation in
many but of couse not all cases....its worth a search on
google of you can remember the drill in such a situation. I
found your points on fibrilation thresholds useful.
Phil Scott
>
> --
> Don Kelly
> dh...@peeshaw.ca
> remove the urine to answer
>
> > --
>
> ------------------------------------------------------------
--------------
Dale Farmer
phil-new...@ipal.net wrote:
> On Wed, 13 Oct 2004 21:01:24 -0500 operator jay <no...@none.none> wrote:
>
> | Does arc hazard depend on available fault current or available fault
> | VA? (Pls let me know). If VA, then changing secondary voltage won't
> | help much. Available secondary VA would be a function of txf rated VA
> | and txf %Z (per unit impedance). So you might have to go with the
> | increased impedance route you mentioned. Or, split loads up onto
> | smaller txfs if that is an option. Sounds messy. I bet a lot of
> | designers wouldn't like selecting voltage based on arc hazard ... when
> | they want 120, they want 120 - and when they want 480, they want 480.
>
> All the data centers I have designed were all based on 120/240 or 208Y/120
> for power source. Usually various kinds of UPS systems are involved.
> I was talking with a friend who has a friend who runs a very large data
> center. They have a pair of 500 kVA UPS systems running the place, and
> behind that a single transfer switch between a fault tolerant utility
> feed (unknown primary voltage) and a 1500 kW generator.
>
> Technicians NOT trained in power engineering would often be working
> around inside the cabinets of the computer and network systems. They
> would be dealing with the secondary LV power. While 120 volts is not
> much, that 500 kVA UPS is delivering 4166 amps and potentially as much
> as 80 kA of fault current. That means the distribution panels will
> have to be rated to interrupt perhaps at least 100 kA (depending on
> lots of things such as actual UPS fault availability, series ratings
> of the whole system, etc).
>
> My tendency is to go with smaller UPS systems on the order of 10 kVA
> to 30 kVA at most. Enough for 1 to 3 cabinets. But these can be had
> with 120 volt output and either 208Y/120 in, or 480Y/277 in. I have
> been thinking of starting to go with the latter. But at least with
> the smaller UPSes, fault currents within a data cabinet will be small.
>
> Going from 208Y/120 to 480Y/277 raises the system impedance to 5.333
> times. Fault currents would be 43.3% as much on the higher voltage
> feeds. I find myself wishing 600Y/346 was more common in the US as
> it is in Canada. The thing is, I have heard of (and under my watch
> happened once, but no one was hurt ... just one guy's pride lost)
> computer technicians shorting something out much more often than
> getting themselves shocked or electrocuted. So it seems to me that
> voltage isn't as much of an issue as amperage, as long as the voltage
> is within what can be safely handled through insulation (e.g. up to
> 600 volts perhaps).
>
> | Re dc fault clearing. I've seen breaker catalogs / data sheets where
> | they list the dc duty a breaker can handle. Maybe you can get what
> | you need off the shelf & cheap. Are you going to be able to find
> | appliances / devices that will take dc at the voltage you have chosen?
> | Would you expect average field strengths to be (a) higher, (b) lower,
> | or (c) don't care, in a dc-wired home?
>
> Incandescent lighting. The decision is 12 volts vs. 120 volts.
>
> --
> -----------------------------------------------------------------------------
> | Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
> | (first name) at ipal.net | http://phil.ipal.org/ http://ka9wgn.ham.org/ |
> -----------------------------------------------------------------------------
Another interesting issue is that the last time I was setting up a computer
center, it was less money to buy a 700 VA individual UPS for each system
than to buy the big system and all the distro to go with it. This removes a
single point of failure. Since the particular application was a linux server
farm, any one machine dropping out was not a big deal. Got a 1400 VA
UPS to operate the KVM switch, hubs and the common monitor... and a
small fluorescent desk lamp.
--Dale
phil-new...@ipal.net
|> would be dealing with the secondary LV power. While 120 volts is
| not
|> much, that 500 kVA UPS is delivering 4166 amps and potentially as
| much
|
| 4166? Would this not be a 3phase unit @ 1400A per phase?
You're right. I was using a new program to get the numbers and forgot
to put the /3 in there. OK, so it's not quite as dangerous as
previously thought.
phil@altair:/home/phil 1217> electric 500000w 120v
amps: 4166.66666 ohms: 0.02880000 volts: 120.000000 watts: 500000.000
phil@altair:/home/phil 1218> electric 500000/3w 120v 277v
amps: 1388.88888 ohms: 0.08640000 volts: 120.000000 watts: 166666.666
amps: 601.684717 ohms: 0.46037400 volts: 277.000000 watts: 166666.666
phil@altair:/home/phil 1219>
| Fault currents would be 43.3% as much on the higher voltage
|> feeds.
|
| Agreed, but, available fault VA is the same either way. Can you tell
| me whether arc hazard depends on fault VA or fault current?
I think the answer would be, it depends. The higher voltage can draw a
longer arc for sure.
| I find myself wishing 600Y/346 was more common in the US as
|> it is in Canada.
|
| Yay Canada.
I've seen 600Y/346 in one place when I was in Texas. I've seen it
nowhere else in the US. My grandfather used to work on 1000Y/577 but
that was at 25 Hz.
| The thing is, I have heard of (and under my watch
|> happened once, but no one was hurt ... just one guy's pride lost)
|> computer technicians shorting something out much more often than
|> getting themselves shocked or electrocuted. So it seems to me that
|> voltage isn't as much of an issue as amperage, as long as the
| voltage
|> is within what can be safely handled through insulation (e.g. up to
|> 600 volts perhaps).
|
| Do you think you WOULD hear about the 120V shocks that technicians
| might accidentally get? I wonder if they'd go unmentioned for
| whatever reasons. I have not gotten into the arc hazard stuff at all.
| So, of course I still instinctively think of lower voltages as safer,
| and I still think of shock hazard being way more of a threat than
| arcs. Maybe I'll learn otherwise.
You may have a point there. Such things could go unreported since no
one dies or goes to the hospital, and no equipment is damaged.
phil-new...@ipal.net
| this does not prove that voltage exists as anything other then a convenient
| theoretical concept.
Semiconductors and insulation (including air) have specific breakdown
voltage levels. I don't think electrical pressure is just pure theory.
| how about phosphorescent strips? no heat, no moving parts
Powered how? The previously powered lighting? I might need to run
down a hallway that had been dark all night, and would have lighted up
automatically from motion detectors.
phil-new...@ipal.net
| Another interesting issue is that the last time I was setting up a computer
| center, it was less money to buy a 700 VA individual UPS for each system
| than to buy the big system and all the distro to go with it. This removes a
| single point of failure. Since the particular application was a linux server
| farm, any one machine dropping out was not a big deal. Got a 1400 VA
| UPS to operate the KVM switch, hubs and the common monitor... and a
| small fluorescent desk lamp.
There are some advantages to smaller. Even the servers cost less these
days when you have many smaller ones to do the work of a few big ones.
But putting several small UPSes in a rack cabinet is not a good fit.
Maybe 2 or 3 small 2U versions designed for the rack. But you can't
get these small ones in 277 volt or three phase input. And they often
have as much harmonic problems as the PC itself. At least the big ones
try to remove harmonics some.
phil-new...@ipal.net
|> You'd have 400 volts between phases in that, right?
|
| 230 the way the gen set was wired.
But if that was 230 volts 3 phase, and each phase was on a common
neutral, you have 400 volts between phases. Was it wires wye/star or
delta? Or maybe 230Y/133?
| It seems the utility could be held reponsiible as most of
| thier customers are not qualified and registered EE's. .. the
| reasonable man priniciple dictates that the utility has the
| responsibliity to advise its customers and see to it that the
| system they are feeding is suitable... EE's will in order to
| protect thier interests, liability etc will have to start
| paying attention to these issues. So will building inspectors.
If the utility hooks me up and says available fault current is 12 kA
and I have breakers rated 25 kA, I'm OK and they are OK. But if years
down the road they upgrade the network and there is then 36 kA of
available fault current, and an accident happens, and the 25 kA breaker
blows up or melts a contact closed, it's sure not the breaker
manufacturer at fault.
|> Data centers are, however, a different issue. And some are
|> run on DC (telco ones tend to be). That 48 VDC stuff can
|> deliver a load of amps.
|
| Im not a telco guy... what size wire are we talking at the
| vdc source?
| typical range for what you do.
The one I saw had some 5 inch wide bus bars that looked to be maybe 3/8
inch thick. I don't know if they were aluminum or plated copper.
The guy that showed me around said there were 960 single battery cells
that weighed in at 300 pounds each. I saw a few of them and volume wise
each was about the size of 4 to 6 car batteries. I presume 2 volts each
so that should be 24 in series, so there would be 40 in parallel. I
don't know if everything was all parallel or if there were separate
isolated circuits. But I'd imagine all of them in parallel could do
some serious vaporizing.
I've never had to work on DC systems, and I don't think I want to.
--
-----------------------------------------------------------------------------
TimPerry
> | how about phosphorescent strips? no heat, no moving parts
>
> Powered how? The previously powered lighting? I might need to run
> down a hallway that had been dark all night, and would have lighted up
> automatically from motion detectors.
>
in the pool, with "downloaded" energy from the sun
Beachcomber
>If the voltage is greater than about 50 volts or so.
>Voltage is the key.
>
>If you plot voltage vs amperage, the safe zone is below 50 volts, and below
>10m amps in most all areas.
>
>
Not necessarily true. I have seen 5 VDC Power supplies for banks of
memory in video equipment that have vaporized circuit boards due to
the arc-flash and the large current capabilites available. The arc
was literally as hot as a commercial arc-welder. All it takes is a
short circuit in the wrong place (in one case, it was a technicians
screwdriver).
The key is to subdivide the current branches with individual,
well-designed fuses or circuit breakers. Edision understood this when
he was constructing the first DC Central Station system in New York
and had to be concerned about safety.
Another example...In the case of a telephone company it is very common
to use large DC supplies common to hundreds or thousands of customers.
In this case, individual impedences are placed in every subscriber
line to limit the current in case of a direct short.
Beachcomber
Michael Moroney
>| I find myself wishing 600Y/346 was more common in the US as
>|> it is in Canada.
>|
>| Yay Canada.
>I've seen 600Y/346 in one place when I was in Texas. I've seen it
>nowhere else in the US. My grandfather used to work on 1000Y/577 but
>that was at 25 Hz.
I asked an electrician neighbor if he ever got to work on something
big and bad like "480 volts at 4000 amps or anything like that" and he
responded that he worked on 600V and 550V systems at times, so there
may be some 600V systems around in Mass. (I asked further about the
550V systems, he said it was common in the old mills and factories
around here. I once worked in an old mill building and there was a
conduit marked 550 VOLTS running around feeding the ancient elevators.
Parts were still knob-and-tube)
--
-Mike
Nicholas O. Lindan
> > | how about phosphorescent strips? no heat, no moving parts
> > Powered how? The previously powered lighting?
> in the pool, with "downloaded" energy from the sun
Tritium. That's what the two T's in the "T SWISS T" mean at the
bottom of some watch dials.
Tritium has been replaced with something else now, can't remember
what ...
--
Nicholas O. Lindan, Cleveland, Ohio
Consulting Engineer: Electronics; Informatics; Photonics.
Remove spaces etc. to reply: n o lindan at net com dot com
psst.. want to buy an f-stop timer? nolindan.com/da/fstop/
Phil Scott
<phil-new...@ipal.net> wrote in message
news:cknq8...@news4.newsguy.com...
> On Thu, 14 Oct 2004 20:49:06 GMT Phil Scott
<philsc...@sf.sbcglobal.net> wrote:
>
> |> You'd have 400 volts between phases in that, right?
not right.
> |
> | 230 the way the gen set was wired.
>
> But if that was 230 volts 3 phase, and each phase was on a
common
> neutral, you have 400 volts between phases.
not correct.
> Was it wires wye/star or
> delta? Or maybe 230Y/133?
Delta. 230 between phases. any line to ground or grounded
neutral is 115 volts. If it was a wye, you would get 115
any line to ground and 230 or whataver its wound for between
phases. yer 400 volt notion does not align.
If you have a combination wye / delta (two diffferent
transformers in series) with any range of taps mid phase then
you can get all sorts of strange anomalies of which you are
beating around the edges of.... 277 volts primarily for
lighting... 115 to ground for recepticles, and some single and
3 phase 230 for motors....still 400v would not be commonly
arranged,..you might get a stinker close to that range...a
single line to ground...not used for anything in that case by
itself.
Phil Scott
phil-new...@ipal.net
|
| <phil-new...@ipal.net> wrote in message
| news:cknq8...@news4.newsguy.com...
|> On Thu, 14 Oct 2004 20:49:06 GMT Phil Scott
| <philsc...@sf.sbcglobal.net> wrote:
|>
|> |> You'd have 400 volts between phases in that, right?
|
| not right.
|
|> |
|> | 230 the way the gen set was wired.
|>
|> But if that was 230 volts 3 phase, and each phase was on a
| common
|> neutral, you have 400 volts between phases.
|
| not correct.
|
|
|> Was it wires wye/star or
|> delta? Or maybe 230Y/133?
|
| Delta. 230 between phases. any line to ground or grounded
| neutral is 115 volts. If it was a wye, you would get 115
| any line to ground and 230 or whataver its wound for between
| phases. yer 400 volt notion does not align.
The 400 volt was for 230 between phase and ground. The exact
values are closer to 230.94 and 400, or 230 and 398.37. You
must be describing a single phase Edison style system to get
115 and 230 volts.
Since you said 230 volts and 3 phase, what came to mind was the
400Y/230 system standardized in Europe (with actual voltages in
various places ranging from 381Y/220 to 416Y/240).
But if you get 115 and 230 in a 3 phase delta, then it sounds
like you have the kind of system where one of the delta windings
is center tapped and grounded so you can get 115 volts between
ground and TWO of the phases. The third (often called high leg
or wild leg) phase would be 200 volts to ground. Lots of 3 phase
power is delivered in the US like that, but the voltages are
standardized at 240, 120, and 208.
| If you have a combination wye / delta (two diffferent
| transformers in series) with any range of taps mid phase then
| you can get all sorts of strange anomalies of which you are
| beating around the edges of.... 277 volts primarily for
| lighting... 115 to ground for recepticles, and some single and
| 3 phase 230 for motors....still 400v would not be commonly
| arranged,..you might get a stinker close to that range...a
| single line to ground...not used for anything in that case by
| itself.
I have no idea how you could wire a wye transformer in series
with a delta transformer. I assume thats what you mean by "two
different transformers in series".
What is the frequency of the system? 50 Hz? 60 Hz? That might
be more telling about what kinds of configuration you are designing
for. Since you mention 277 volts, it sounds like a 60 Hz system
as I have seen that voltage only in 60 Hz systems.
What did the generators produce? Wye? Delta? From there you can
transform it to other voltages and other configurations. If the
transformers have delta primaries you could be running the generators
in either delta or wye.
--
-----------------------------------------------------------------------------
Phil Scott
<phil-new...@ipal.net> wrote in message
news:ckujl...@news2.newsguy.com...
It could have been 240 .. that was 20 years ago.. I know
it wasnt 400v to ground, and that we had both single phase
230, three phase, and 110 from the generators.. and yes there
is stinger with that arrangement.
>
>
> | If you have a combination wye / delta (two diffferent
> | transformers in series) with any range of taps mid phase
then
> | you can get all sorts of strange anomalies of which you
are
> | beating around the edges of.... 277 volts primarily for
> | lighting... 115 to ground for recepticles, and some single
and
> | 3 phase 230 for motors....still 400v would not be commonly
> | arranged,..you might get a stinker close to that range...a
> | single line to ground...not used for anything in that case
by
> | itself.
>
> I have no idea how you could wire a wye transformer in
series
> with a delta transformer. I assume thats what you mean by
"two
> different transformers in series".
look up various transformer configurations in a basic
power distribution manual, there are many options... you sound
like you would be fully familiar with these strategies.
>
> What is the frequency of the system? 50 Hz? 60 Hz? That
might
> be more telling about what kinds of configuration you are
designing
> for.
I designed and built it in 1982 aboard the Hawaiian Princess
(Honollulu, 200' fisher processor)..its not a project in
progress.
> Since you mention 277 volts, it sounds like a 60 Hz system
> as I have seen that voltage only in 60 Hz systems.
the 277 was mentioned in context with other systems
entirely... commercial lighting on land... not on this boat at
all.
>
> What did the generators produce? Wye? Delta? From there
you can
> transform it to other voltages and other configurations. If
the
> transformers have delta primaries you could be running the
generators
> in either delta or wye.
We wired it delta as discussed above.
Phil Scott
>
> --
> ------------------------------------------------------------
phil-new...@ipal.net
| It could have been 240 .. that was 20 years ago.. I know
| it wasnt 400v to ground, and that we had both single phase
| 230, three phase, and 110 from the generators.. and yes there
| is stinger with that arrangement.
Then it sounds like 230 delta with one side center tapped
for 115 volts, and a 200 volt wild leg.
| look up various transformer configurations in a basic
| power distribution manual, there are many options... you sound
| like you would be fully familiar with these strategies.
I've seen many. Probably just about anything can be done with
enough creativity. But making it economical is another matter.
| the 277 was mentioned in context with other systems
| entirely... commercial lighting on land... not on this boat at
| all.
Ah, OK. The common 480Y/277 where industrial motors use 480 in
a delta connection to the wye, and lighting circuits connected to
get 277 volts to ground (common for fluorescent and HID lighting
that voltage is available).
--
-----------------------------------------------------------------------------
Dale Farmer
phil-new...@ipal.net wrote:
> On Fri, 15 Oct 2004 01:50:28 GMT Dale Farmer <da...@cybercom.net> wrote:
>
> | Another interesting issue is that the last time I was setting up a computer
> | center, it was less money to buy a 700 VA individual UPS for each system
> | than to buy the big system and all the distro to go with it. This removes a
> | single point of failure. Since the particular application was a linux server
> | farm, any one machine dropping out was not a big deal. Got a 1400 VA
> | UPS to operate the KVM switch, hubs and the common monitor... and a
> | small fluorescent desk lamp.
>
> There are some advantages to smaller. Even the servers cost less these
> days when you have many smaller ones to do the work of a few big ones.
> But putting several small UPSes in a rack cabinet is not a good fit.
> Maybe 2 or 3 small 2U versions designed for the rack. But you can't
> get these small ones in 277 volt or three phase input. And they often
> have as much harmonic problems as the PC itself. At least the big ones
> try to remove harmonics some.
I priced rack mount computers versus regular AT chassis. The AT
machines cost way less, and I put them on 'bakers rack' wire shelving.
They are inexpensive and since they had wheels, I could pack them all
into the back of the computer room, just had to roll out the shelf
unit I needed when I needed physical access to them. Not quite as many
systems per square foot as racked systems, but it saved me way more
money than the cost of the additional square feet of computer room.
--Dale