A few more ideas...

1. Low cost current sense transformers are available at
http://www.toroid.com/standard_transformers/current_sensing_transformers/current_sensing.htm
.

2. You can minimize nonlinearity due to rectification by putting the
"burden resistor" after the bridge rectifier. The transformer secondary
forces a current proportional to the primary current through the burden
resistor even though there's a couple diode drops in the way.

3. You can do low side sensing in the neutral side of the power line
making the circuit nonisolated. For that matter, you can also do it on the
high side (have an ADC on the high side, then opto couple the data down.
You need a floating power supply for the ADC).

4. You can use high common mode voltage differential amplifiers (AN117 or
similar) to bring a high side sample down to ground.

5. I like rectification instead of centering the sample in the ADC range
since precise centering is difficult. If you can use half wave
rectification, have the transformer secondary (with burden resistor across
it) drive a current limit resistor into the ADC. Put a clamp diode on the
ADC input so it cannot be driven below ground (much). Drop across the
current limit resistor will be minimal during positive half cycles.

Good luck!

Harold


> You have a few options:
> You can wind the transformer yourself on a ferrite core. You will have
> to calibrate it with a known current.
> Rectification is necessary only if you need a positive sense voltage. If
> you are feeding an ADC you can reference one side of the output to your
> Vref and the other will vary proportional to the current on the line.
> You can make a 'perfect' rectifier with an opamp if you need unipolar
> output.
> There are Allegro hall-effect current sensors that sense the magnetic
> field around a current carrying conductor. They work well but I don't
> know if they have the 100mA resolution that you want. They may send you
> free samples.
> Good luck.
> --
> Martin K
>
> Alessandro Queri wrote:
>
>>Hello guys.
>>
>>I need to monitor the power consumption on the mains. I googled and found
>>one can use current transformers but they are very expensive and not
>> easily
>>available in my region (Italy). Besides with this technique a full bridge
>>rectifier is also needed and as I understand, there's a treshold current
>> you
>>cannot measure.
>>My need is to measure from 100mA (that is 22 W @ 220V) to 30A (actually I
>>need 5KW) with a resolution of 100mA. should I wind the transformer
>> myself,
>>and how? Any other idea?
>>
>>Many TIA.
>>
>>Alessandro Queri
>>
>>
>>
>
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>


-- 
FCC Rules Updated Daily at http://www.hallikainen.com - Advertising
opportunities available!
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist