Posted by Johny B Good on March 8, 2014, 1:12 am
If the effective length of cable section being scanned by the
detector arrays is limited to just a few millimetres, I doubt the
usual rates of twist in a typical power cord would add more than a 1%
Even so, if such a sophisticated bunch of sensor arrays are being
used, it seems very likely that the twist rate can be determined and
this information used to cancel out any such 'error'.
It seems to be, as I said before, an awful lot of trouble just avoid
seperating out the hot conductor to take electric field (voltage) and
magnetic field (current) sample data to calculate a wattage reading.
However, if this tech is real and based on 'de-weaponised' 'Spook'
technology that's been released into the private sector, then why not?
Regards, J B Good
Posted by mike on March 6, 2014, 5:58 am
On 3/5/2014 6:15 PM, amdx wrote:
You're thinking far field.
Up close and personal, near fields have more information that
you can get from measuring at more than one point in space.
Posted by RobertMacy on March 6, 2014, 12:21 pm
repost a reply here....
Fields from a SINGLE long wire drop off as the inverse distance, 1/S. Ah,
but usually you have two wires, one with current + and the other with
current - If the wires are REALLY close together you can see how the plus
and minus end up with almost zero field.
However, most wires have a bit of size to them, thus kind of close to them
you get 1/S for the positive flow and -1/(S+dS), where dS is that smidgeon
of extra distance. now you see, NOT zero field...
What you get here is
1/S - 1/(S+dS) = (S+dS-S)/(S^2+dS*S), for really small dS, throw it away
and you get
dS/S^2, thus everybody tells you how the field drops off as the inverse
SQUARE of the distance.
Now, with judicious measurements made in a 'real' scenario; it is possible
to find the current flow in the two wires just by examining the magentic
field around the pair, albeit a bit tricky. For example, measuring close,
and further, you can find the 'distance' between the wires and calibrate
your measurement. In other words, measure current without knowing the
exact structure of your wires. and determine the structure of the wires
all at once. But, as I said, a bit tricky.
Most clamp on meters use an 'iron' core to distort and 'short' [as in
concentrate] the field around whatever it is clamped onto. That type meter
dtermines the TOTAL current flowing through whatever it's clamped onto, so
if clamped on both AC wires, you get zero.
However, what I described is measuring the field WITHOUT distorting the
field [well, not distorting too much]
As an extreme example of the effect of 'spatially' measuring fields to
then determine current flow:
My magnetic field measuring circuitry has extremely low noise floor. Using
that circuitry I once designed a non-contact 'clamp on' multi-sensor
system for measuring the independent current flows inside a 25 pair cable.
...for tapping any phone line you want. Each pair when 'active' is
anywhere from 10-75mA and produces a noticeable field out to around 3 ft.
If your noise floor is low enough.
Posted by Johny B Good on March 6, 2014, 12:38 pm
On Thu, 06 Mar 2014 05:21:37 -0700, RobertMacy
So, a bunch of electronically selectable and 'steerable magnetic
detector arrays' plus a shitload of electronics then?
I suppose it might be do-able today (Hell! who'd have thought we'd
have microprocessor chips with transistor counts measured in hundreds
of millions (last figure I saw was 135 Million nearly a decade ago).
Seems to be an incredibly 'over-engeered' solution' just to save the
need to isolate the conductors to take a current measurement though.
Regards, J B Good
Posted by RobertMacy on March 6, 2014, 1:43 pm
On Thu, 06 Mar 2014 05:38:31 -0700, Johny B Good
My guess at material cost for using 'available' components, around $00,
and DSP processing at much less than $00
Well it's for people who want to do something and not be noticed doing
it. Especially undetectably tapping a single phone line from 1 to 3 feet
away, through a wall for example, then use a short range 'retransmit' to
get it out without tracking the contact back to yourself. package
something like that in a wad of tar, so it looks innocuous, like tar
dripped from the ceiling onto the cabling, etc.