For an Overcurrent detector, the present solution is a full wave bridge of MBR0530 diodes, and the forward drop of these diodes is the cause of increasing error towards the low setting.
The current signal is presented across a 5.1 ohm resistor from a 1:1000 current transformer, rectified, and compared to an adjustable and metered reference, with an AD311 comparator.
There is a 5 volt and a 9 volt supply available, and the frequency of the signal can vary from 20kHz and up to 1 MHz.
Is it possible to build a precision rectifier that runs from a single supply, and functions up to 1 MHz?
Cheers, Finn Hammer
It is always nice to be able to answer one's own questions, since it usually means, that the effort to find out has been succesful.I tried to build the circuit mentioned in this publication:
but instead of the AD8277 I used AD8028 and 2.2k resistors instead of 40k, I got this result at 500kHz, which is sort of ok, I think
At higher frequencies, the inverted part of the signal gets more ragged, and obtains a peak at the leading edge, This peak exeeds the absolute value around 800kHz, and this is exactly within speck, so I will stick with this solution.