Thanks Eric,
I am already allowing for this in my software. At the moment I am deducting 2 from the values I am reading.
Plugging my laptop charger: is showing a reading of:
AD736 output......CA3140 output......Software output...reference system
2mv....................11.3/11.4mv..........(4 raw) .............. 0.045KW
My watts would be : ((2/10)*240)/1000 = 0.048 KW
Adding my 2000 watts hair dryer to the load:
77mv....................230mv...............( 76 raw)............ 1.790 to 1830KW
My watts would be : ((74/10)*240)/1000 = 1.776 KW
I think those are pretty good figures especially that they match very closely a reference system I am using.
Is it possible for me to send the schematics to someone who can build the PCB for me without remortgaging my home to pay for it
.
My other alternative is to do it on a veroboard, and I was thinking of adding an LED pushbutton to switch on and off, 2 banana sockets for the clamp and a plug for +5v, Gnd, and output, and place all this in nice case with battery compartment.
Ideally the CA3140 should be powered from +/- 5V supplies in order to get absolute zero output. Most OPA when working on a single supply will not go to absolute zero. The CA3140 is a good choice for single supplies as will go close to zero output.
I would use the program to remove this small zero offset, the offset value will remain constant over the full output range.
The only other alternative is a another small battery, which is a pain.!
The reason for asking about the 100K's if they were low tolerance, its possible that the junction of the two would not be exactly V/2, which would give a slight dc offset.
Personally I think the results you are getting are OK.
Once the project is laid out on a pcb the noise should be reduced.
I would be interested when you have finished building, to see your test results.