Tested one of the current sensors tonight.
http://scienceshareware.com/articles...ent-sensor.pdf
My sensors are CSLA2DK versions and they are good for
400 amps! (3 uSec response time).
I'm using a 12vdc power supply, so my Zero DC bias voltage is at 6 volts. (Not 4v as shown)
This evening, I put 3 turns of wire (125vac @ 10amps) though the core.
My oscilloscope display a 6 volt DC signal with a superimposed sinewave,
of only 2 vac peak to peak.
It was going up to 7 volts and dropping down to 5 volts. (at 60hz).
Since the CAI board's Analog inputs are 0 to 10vdc with 1023 resolution,
That means the resolution is about 0.01v per bit.
So, 2 volts are going to have a range of only 200 bits.
Using 200 bits to monitor 0 to 10 amps would be pretty course.
But, the bad news.. The CAI board can't read AC voltages (peak),
without a lot software overhead. I'll have to convert the AC output from the sensor, to DC.
To capture the +&- magnitude of the whole 60hz waveform, I'm thinking of
feeding the sensor's signal though a capacitor, into a full-wave bridge.
(Voltage doubler).
This should give the CAI board a larger +DC voltage input to work with.
The hole in the core is 0.625" ID. So, adding
up to about 10 of loops of 12G wire is going to be feasible.
Ten turns would really increase the sensitivity. Maybe from 400a down to 40a.?.
I'm hoping to be able to use most of the full range of the analog input,
(Which is clipped at 9v?), because the +6vdc bias will be blocked out.
Only when AC current is passing, will DC signal be produced.
It would be great if I could get these working at 0 to 8vdc, to represent 0 to 16 Amps, ..
Half scale is 512 bits, so if I could get 8A to cause a 512 output..?.
