# Measuring current using inductance coil

#### Parthasarathy

Joined Feb 15, 2004
2
Dear Friends,
We have implmented a crypto system in a 8051 micro controller.Now we are going to carry out ,what are known as power attacks on the cryptosystem.ie whenever a bit in the KEY string is '1' then certain operations will be performed by the microcontroller otherwise those operations will be skipped.We expect a small surge in current when the KEY bit is '1'.We would like to measure this current and hence determine the KEY.The processor runs at 12 MHz and each instruction cycle requirs 4 clock cycles.We put a toroidal wound inductor around the power cable of the microcontroller and couldn't measure any significant current changes.Also there is a large noise pickup.
What could be ideal core and coil material ,dia for this measurement.Can we make use of a differential amplifier?
Kindly help.

#### Battousai

Joined Nov 14, 2003
141
If you used an op-amp in a current buffer configuration that might help reject the noise. Why not measure the voltage change across the inductor during the current surge instead of trying to measure the current itself?

#### Parthasarathy

Joined Feb 15, 2004
2
Dear sir,
Thanks.That is what we have been trying to do - measuring the voltages across the inductance coil.The problem is as below :
The micro controller kit is powered by the USB card.USB contains 4 wires of which one is 5V power lead.We have isolated that lead and put a toroidal wound coil around it.But the voltage developed accross the coil is very small since the current fluctuations are tiny.
So noise overpowers the meaningful information leakage by the way of current fluctuations.
Actually we ran a computationally intensive (periodic) code and expected to see some periodic fluctuation in the voltage but due to the noise(>20mV) we could not observe anything.

Now we would like to know two things:
1.What type of core and gauge size should be used to pickup 0 -
12MHz current fluctuations.
2.How to minimise noise ?