[Prev][Next][Index][Thread]

Calibrating I Transformer

Mark;

For the most part, your calibration procedure seems to be a reasonable one.
However, it leaves out an important point. Your statement:

>Now you can adjust the turns on the coil to result in a .25 volt or 2.5 volt
>or 25 volt reading on the meter connected to the winding which will be in
>proportion to the current flowing through the line that runs through the
>center of the core.

is correct in the absolute literal sense, but may lead others to draw a false
conclusion. More strictly, the voltage induced in the coil is not proportional
to i, but to di/dt. True, if we increase i by a factor, di/dt will increase
by the same factor, but di/dt will also increase in proportion to frequency,
even if i is not changed. This means that a calibration done at one frequency
will be invalid at any other frequency, unless a frequency correction factor
is employed. Fortunately, the correction factor is simple to use. If the
signal being measured has a frequency that's n times the calibration
frequency, then the actual current will be n times the meter reading. This
shouldn't detract from the fine job you did in describing the calibration
procedure, but I hope it will help make it as useful as it deserves to be.

Along another slightly different, but intimately related topic, it might be
useful to employ the current sensor at various points along the ground wire,
to look for variations. I'm wondering wether the Tesla secondary might not
couple into the winding of a moving coil-type meter, and disrupt it's
accuracy. I say this based on previous mention here of current measurement with
an AC ammeter, whose numbers seemed way, way too high for my own gut feeling.
A quick "back of the envelope" calculation seemed to verify that impression,
and I'm at least vaguely suspicious that something adversely disturbed the
measurement. Wether or not it really did, and wether or not this is the
reason remains to be seen, but moving the sensor & meter might be an
educational thing to try, and will not have wasted an unreasonable amount of
time, should it prove unimportant.

Wes B.