Cables and Currents

[Chris]

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

 

[Al]

On Fri, 03 Mar 2017 16:14:15 -0500, Ralph Mowery wrote:

Battery jumper cables for the cars are an example. The wire will not
carry large ammounts of current for very long, but long enough to jump
off a car with a dead battery.

So when you see these jumper cables rated at 400A or 600A or whatever,
that's just for 30s or something? I must admit I've never thought about
it. Come to think about it, a continuous duty 600A cable would be far
thicker than the 1/2" diameter stuff they typically label as 600A jumper
cable I would imagine.

 

[Al]

On Fri, 03 Mar 2017 15:19:13 -0800, John Larkin wrote:

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

How so?

I look at some high power MOSFETs/"HEXFETSs" you see in TO-220 package
IIRC and wonder how those slender leads are supposed to carry the very
high currents the devices are often rated for. There must be something
going on there - some effect or other of which I'm ignorant - which
enables that to happen.

 

[Jeroen Belleman]

On 03/03/17 21:20, Chris wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

Not a stupid question at all.

Yes, you can exceed the normal rating of the cable while
reducing the duty cycle. The rating is usually based on
heating due to ohmic loss. The upshot is that you'll have
to reduce the duty cycle quadratically: You pass twice the
rated current, you should use a duty cycle of 1/4. Also,
the pulse length should be short compared to the thermal
time constant of the cable. Shorter still if you exceed
the cable rating grossly.

This being Usenet, yes, I know there are limits to everything.

Jeroen Belleman

 

[Ralph Mowery]

In article <o9cj5r$bd5$17@dont-email.me>, cbx@noreply.com says...

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

In many cases yes. The current rating of cables is often dependent on
how hot the wire will get and the insulation heat rating.

Sooner or later you will get enough current to melt the insulation, melt
the wire, or too much voltage drop.

Battery jumper cables for the cars are an example. The wire will not
carry large ammounts of current for very long, but long enough to jump
off a car with a dead battery.

 

[Ralph Mowery]

In article <o9cquq$bd5$18@dont-email.me>, iqbalali898@noreply.com
says...

On Fri, 03 Mar 2017 16:14:15 -0500, Ralph Mowery wrote:

Battery jumper cables for the cars are an example. The wire will not
carry large ammounts of current for very long, but long enough to jump
off a car with a dead battery.

So when you see these jumper cables rated at 400A or 600A or whatever,
that's just for 30s or something? I must admit I've never thought about
it. Come to think about it, a continuous duty 600A cable would be far
thicker than the 1/2" diameter stuff they typically label as 600A jumper
cable I would imagine.

Thats right. The wire to handle the full current of the starting of a
car , if it had to carry that much for say an hour or all day would be
much larger.

I worked as an electrician at a plant that used many circuits of 200 or
more amps. Large wire and sometimes 2 wires in parallel to handle that
current full time. Unless made of many fine wires like welding cable
they would be so big and stiff they would be hard to handle. Even at
that they would be very bulky to store in a car.

Many of the cables are only 6 or 8 guage. Some are even 10 gauge of
copper coated aluminum. While they may be rated for 200 amps to start a
car, the # 10 would only be rated for 30 amps in a house if they were
all copper. Good enough to turn over a 4 or 6 cyclinder engine for the
short time it should take to start it.

 

[John Larkin]

On Fri, 3 Mar 2017 20:20:11 -0000 (UTC), Chris <cbx@noreply.com>
wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

Sure. The cable heats up from the current (current squared,
approximately) and has some heat storage capacity. So you can really
whack it for a short time, milliseconds to tens of seconds maybe,
before the copper gets too hot.

Wire can handle a lot of current if you cool it, too. Most power
wiring stuff assumes that wires are inside jackets, inside walls
maybe, where there's not much cooling. So power wire is conservatively
rated for current.

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

(Which could restart the argument about "average RMS current.")


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[John Larkin]

On Fri, 3 Mar 2017 22:32:58 -0000 (UTC), Al <iqbalali898@noreply.com>
wrote:

On Fri, 03 Mar 2017 16:14:15 -0500, Ralph Mowery wrote:

Battery jumper cables for the cars are an example. The wire will not
carry large ammounts of current for very long, but long enough to jump
off a car with a dead battery.

So when you see these jumper cables rated at 400A or 600A or whatever,
that's just for 30s or something? I must admit I've never thought about
it. Come to think about it, a continuous duty 600A cable would be far
thicker than the 1/2" diameter stuff they typically label as 600A jumper
cable I would imagine.

They lie, too.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[Phil Allison]

Chris wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

** Cable current ratings are given in amps rms - so the heating effect of the current is what matters, not its peak or average value.

For the same reason, circuit breakers are thermally operated, responding to the rms value of whatever current is flowing in the wire.

The most efficient use of a cable occurs with DC current.

..... Phil

 

[DJ Delorie]

Al <iqbalali898@noreply.com> writes:

I look at some high power MOSFETs/"HEXFETSs" you see in TO-220 package
IIRC and wonder how those slender leads are supposed to carry the very
high currents the devices are often rated for. There must be something
going on there - some effect or other of which I'm ignorant - which
enables that to happen.

I think that's something different - current handling in house wiring is
limited by the insulation. When it melts and exposes wiring... fail.
Part leads are limited by the melting point of the metal, which is I
think steel (vs copper) so the metal part of the conductor can handle a
much higher current. Inner-chip wiring is gold? That is an even better
conductor, so even more current. Etc.

 

[John Larkin]

On Sat, 4 Mar 2017 01:35:51 -0000 (UTC), Al <iqbalali898@noreply.com>
wrote:

On Fri, 03 Mar 2017 15:19:13 -0800, John Larkin wrote:

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

How so?

By making peak temperatures higher than what you'd get if the current
were steady. Peak temperatures melt things. And heat increases the
wire's resistance (well, for most materials) which causes more heat.

I look at some high power MOSFETs/"HEXFETSs" you see in TO-220 package
IIRC and wonder how those slender leads are supposed to carry the very
high currents the devices are often rated for. There must be something
going on there - some effect or other of which I'm ignorant - which
enables that to happen.

IR invented the trick of wildly overstating the current and power
capability of small mosfets, like by radically cooling them in exotic
boiling fluids. Everybody else had to tell the same lies to be
competitive.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[George Herold]

On Friday, March 3, 2017 at 6:19:23 PM UTC-5, John Larkin wrote:

On Fri, 3 Mar 2017 20:20:11 -0000 (UTC), Chris <cbx@noreply.com
wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

Sure. The cable heats up from the current (current squared,
approximately) and has some heat storage capacity. So you can really
whack it for a short time, milliseconds to tens of seconds maybe,
before the copper gets too hot.

Wire can handle a lot of current if you cool it, too. Most power
wiring stuff assumes that wires are inside jackets, inside walls
maybe, where there's not much cooling. So power wire is conservatively
rated for current.

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

(Which could restart the argument about "average RMS current.")


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

For thing like magnet coils (we do mostly air coils)
you can totally run 'em high, we have one instrument,
that limits the duty cycle... up to a 20 second period.

For a hunk of copper, there should be some current, that
raises the piece 1 deg K/ sec. (Well at least for small changes in T)

A related question, (and currently of more interest to me. NPI)
is how much current can a wire carry in vacuum.
I've got some graphs on my computer at work, but I'm not sure I believe
them.... The wire is phosphor-bronze,
this looked good, but they didn't model radiation..?
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090032058.pdf

I've never heard of this, but do people paint their
wires black? Better radiators.

George H.

 

[John Larkin]

On Sat, 4 Mar 2017 11:45:26 -0800 (PST), George Herold
<gherold@teachspin.com> wrote:

On Friday, March 3, 2017 at 6:19:23 PM UTC-5, John Larkin wrote:
On Fri, 3 Mar 2017 20:20:11 -0000 (UTC), Chris <cbx@noreply.com
wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

Sure. The cable heats up from the current (current squared,
approximately) and has some heat storage capacity. So you can really
whack it for a short time, milliseconds to tens of seconds maybe,
before the copper gets too hot.

Wire can handle a lot of current if you cool it, too. Most power
wiring stuff assumes that wires are inside jackets, inside walls
maybe, where there's not much cooling. So power wire is conservatively
rated for current.

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

(Which could restart the argument about "average RMS current.")


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

For thing like magnet coils (we do mostly air coils)
you can totally run 'em high, we have one instrument,
that limits the duty cycle... up to a 20 second period.

Big (non-superconducting) electromagnets are usually water cooled.
They have a lot of copper volume to surface area ratio, so get hot.

For a hunk of copper, there should be some current, that
raises the piece 1 deg K/ sec. (Well at least for small changes in T)

A related question, (and currently of more interest to me. NPI)
is how much current can a wire carry in vacuum.
I've got some graphs on my computer at work, but I'm not sure I believe
them.... The wire is phosphor-bronze,
this looked good, but they didn't model radiation..?
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090032058.pdf

I've never heard of this, but do people paint their
wires black? Better radiators.

Most shiny metals have low emissivities at thermal wavelengths. Copper
is an almost perfect mirror at thermal IR. So in a hard vacuum,
practically the only cooling will be conduction to the end
terminations.

Making the wire black (at thermal wavelengths!) would really help.
Smashing it into a ribbon would increase the surface area, too.

Painted or insulated wire is better than bare metal, unless you can
run literally red hot. Most organics have high emissivity.

Interesting experiment: try bare copper wire vs magnet wire, same
sizes, same current, in vacuum. Inferr the temperature from the
resistance.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[George Herold]

On Tuesday, March 7, 2017 at 8:50:58 PM UTC-5, John Larkin wrote:

On Tue, 7 Mar 2017 12:17:40 -0500, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 03/07/2017 11:49 AM, John Larkin wrote:
On Tue, 7 Mar 2017 06:51:48 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Tuesday, March 7, 2017 at 9:33:38 AM UTC-5, Phil Hobbs wrote:
On 03/06/2017 05:17 PM, John Larkin wrote:
On Mon, 6 Mar 2017 09:42:27 -0500, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 03/04/2017 07:22 PM, John Larkin wrote:
On Sat, 4 Mar 2017 11:45:26 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Friday, March 3, 2017 at 6:19:23 PM UTC-5, John Larkin wrote:
On Fri, 3 Mar 2017 20:20:11 -0000 (UTC), Chris <cbx@noreply.com
wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

Sure. The cable heats up from the current (current squared,
approximately) and has some heat storage capacity. So you can really
whack it for a short time, milliseconds to tens of seconds maybe,
before the copper gets too hot.

Wire can handle a lot of current if you cool it, too. Most power
wiring stuff assumes that wires are inside jackets, inside walls
maybe, where there's not much cooling. So power wire is conservatively
rated for current.

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

(Which could restart the argument about "average RMS current.")


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

For thing like magnet coils (we do mostly air coils)
you can totally run 'em high, we have one instrument,
that limits the duty cycle... up to a 20 second period.

Big (non-superconducting) electromagnets are usually water cooled.
They have a lot of copper volume to surface area ratio, so get hot.


For a hunk of copper, there should be some current, that
raises the piece 1 deg K/ sec. (Well at least for small changes in T)

A related question, (and currently of more interest to me. NPI)
is how much current can a wire carry in vacuum.
I've got some graphs on my computer at work, but I'm not sure I believe
them.... The wire is phosphor-bronze,
this looked good, but they didn't model radiation..?
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090032058.pdf

I've never heard of this, but do people paint their
wires black? Better radiators.

Most shiny metals have low emissivities at thermal wavelengths. Copper
is an almost perfect mirror at thermal IR. So in a hard vacuum,
practically the only cooling will be conduction to the end
terminations.

Making the wire black (at thermal wavelengths!) would really help.
Smashing it into a ribbon would increase the surface area, too.

Painted or insulated wire is better than bare metal, unless you can
run literally red hot. Most organics have high emissivity.

Interesting experiment: try bare copper wire vs magnet wire, same
sizes, same current, in vacuum. Inferr the temperature from the
resistance.



Thick black copper oxide has an emissivity of about 0.78 in the thermal
IR, according to
http://www-eng.lbl.gov/~dw/projects/DW4229_LHC_detector_analysis/calculations/emissivity2.pdf

Cheers

Phil Hobbs

I guess that copper will get hot and tarnish some.

Not to change the subject, but regular office white-out has a very
high emissivity. So you can dab it on shiny things, like the metal
tops of some FPGAs and such, to read the temps better.

Kapton tape is pretty good.

https://dl.dropboxusercontent.com/u/53724080/Thermal/Cool1.JPG

https://dl.dropboxusercontent.com/u/53724080/Thermal/Cool2.jpg



Just about any dielectric at least a few mils thick has a thermal
emissivity of about 0.95. The rest is Fresnel reflection at the surface.

Fresnel reflection... I had to look it up. (dielectric mismatch)
Is this for IR wavelengths? Certainly off white paint is higher than
that in the visible.
Maybe I can get the wire with a triple or quadruple build
of insulation. That adds... (checks MWS catalog) ~30 mil to the diameter.

George H.

The more insulation, the cooler the wire! A universe full of plastic
conducts heat better than a universe full of vacuum.



T'other way round. Near room temperature, any thickness of vacuum looks
like about 5 mm of air, iirc, which is 0.5 mm of plastic.


I don't get that. Vacuum conducts heat better than air? 0.5 mm of
plastic conducts heat as well as 5 mm of air?

I think Phil is calculating the thermal conductivity of air,
then equating that to some thermal loss from radiation in vac.
But he has to pick some temperature rise... 1 deg K or something.

And the air number has to be for low temperature differences..
or convection kicks in...

But it is complex. More insulation increases the radiation surface,
which works against the T^4 radiation curve. And more insulation
conducts heat out to the surface of a given radius, better than a
vacuum gap would... depending on the thermal conductivity of the
insulation. All that math is way past my pay grade.
Grin... right do the measurement and use that to check/adjust your math. :^)

George H.

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[Phil Hobbs]

On 03/07/2017 08:50 PM, John Larkin wrote:

On Tue, 7 Mar 2017 12:17:40 -0500, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 03/07/2017 11:49 AM, John Larkin wrote:
On Tue, 7 Mar 2017 06:51:48 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Tuesday, March 7, 2017 at 9:33:38 AM UTC-5, Phil Hobbs wrote:
On 03/06/2017 05:17 PM, John Larkin wrote:
On Mon, 6 Mar 2017 09:42:27 -0500, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 03/04/2017 07:22 PM, John Larkin wrote:
On Sat, 4 Mar 2017 11:45:26 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Friday, March 3, 2017 at 6:19:23 PM UTC-5, John Larkin wrote:
On Fri, 3 Mar 2017 20:20:11 -0000 (UTC), Chris <cbx@noreply.com
wrote:

This may be a stupid question, but here goes.
We all know that cable is graded for its current carrying capabilities
according to its cross-sectional area. BUT, could one conceivably pass
excessive amounts of current through a cable not rated to carry it by
pulsing the current in short bursts at a very low duty cycle?

Sure. The cable heats up from the current (current squared,
approximately) and has some heat storage capacity. So you can really
whack it for a short time, milliseconds to tens of seconds maybe,
before the copper gets too hot.

Wire can handle a lot of current if you cool it, too. Most power
wiring stuff assumes that wires are inside jackets, inside walls
maybe, where there's not much cooling. So power wire is conservatively
rated for current.

Pulse bursts don't increase the long-term RMS current capacity of a
wire. They actully reduce it.

(Which could restart the argument about "average RMS current.")


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

For thing like magnet coils (we do mostly air coils)
you can totally run 'em high, we have one instrument,
that limits the duty cycle... up to a 20 second period.

Big (non-superconducting) electromagnets are usually water cooled.
They have a lot of copper volume to surface area ratio, so get hot.


For a hunk of copper, there should be some current, that
raises the piece 1 deg K/ sec. (Well at least for small changes in T)

A related question, (and currently of more interest to me. NPI)
is how much current can a wire carry in vacuum.
I've got some graphs on my computer at work, but I'm not sure I believe
them.... The wire is phosphor-bronze,
this looked good, but they didn't model radiation..?
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090032058.pdf

I've never heard of this, but do people paint their
wires black? Better radiators.

Most shiny metals have low emissivities at thermal wavelengths. Copper
is an almost perfect mirror at thermal IR. So in a hard vacuum,
practically the only cooling will be conduction to the end
terminations.

Making the wire black (at thermal wavelengths!) would really help.
Smashing it into a ribbon would increase the surface area, too.

Painted or insulated wire is better than bare metal, unless you can
run literally red hot. Most organics have high emissivity.

Interesting experiment: try bare copper wire vs magnet wire, same
sizes, same current, in vacuum. Inferr the temperature from the
resistance.



Thick black copper oxide has an emissivity of about 0.78 in the thermal
IR, according to
http://www-eng.lbl.gov/~dw/projects/DW4229_LHC_detector_analysis/calculations/emissivity2.pdf

Cheers

Phil Hobbs

I guess that copper will get hot and tarnish some.

Not to change the subject, but regular office white-out has a very
high emissivity. So you can dab it on shiny things, like the metal
tops of some FPGAs and such, to read the temps better.

Kapton tape is pretty good.

https://dl.dropboxusercontent.com/u/53724080/Thermal/Cool1.JPG

https://dl.dropboxusercontent.com/u/53724080/Thermal/Cool2.jpg



Just about any dielectric at least a few mils thick has a thermal
emissivity of about 0.95. The rest is Fresnel reflection at the surface.

Fresnel reflection... I had to look it up. (dielectric mismatch)
Is this for IR wavelengths? Certainly off white paint is higher than
that in the visible.
Maybe I can get the wire with a triple or quadruple build
of insulation. That adds... (checks MWS catalog) ~30 mil to the diameter.

George H.

The more insulation, the cooler the wire! A universe full of plastic
conducts heat better than a universe full of vacuum.



T'other way round. Near room temperature, any thickness of vacuum looks
like about 5 mm of air, iirc, which is 0.5 mm of plastic.


I don't get that. Vacuum conducts heat better than air? 0.5 mm of
plastic conducts heat as well as 5 mm of air?

But it is complex. More insulation increases the radiation surface,
which works against the T^4 radiation curve. And more insulation
conducts heat out to the surface of a given radius, better than a
vacuum gap would... depending on the thermal conductivity of the
insulation. All that math is way past my pay grade.

Yup. In vacuo, the heat just radiates away into space, so any thickness
of vacuum has the same thermal resistance, namely the derivative of the
Stefan-Boltzmann law, i.e.

L = epsilon sigma T**4,

where epsilon is the thermal emissivity. Two parallel surfaces with
different temperatures will exhibit a power transfer per unit area of

Delta L = epsilon_1 epsilon_2 sigma (T_1**4 - T_2**4),

which for small delta-T is

alpha = dL/dT = 4 epsilon_1 epsilon_2 sigma T**3.

It's modified some by the thermal emissivity of the emitter and
surroundings.

That's how superinsulation works--you have many layers of metallized
Kapton, spaced out so that they don't touch. Works great, but it's an
absolute bear to bake out--all that surface area, the constricted space
for gas to diffuse out, and the superior insulation making it hard to
get it all hot.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net