Need for an accuarate Peak/Amplitude detector

[vahid_ff]

Hi folks,

I designed a signal amplifier and need to measure the amplitude of the output accurately (less that %5 error) to be independent of oscilloscopes. Since output signal has some sort of distortion, especially at high frequencies, the idea of using RMS converters is useless. Furthermore, I tried to simulate other ideas based on fast ECL comparators like (**broken link removed**) and (**broken link removed**), but I didn't succeed. The frequency range is 100 KHz up to 6-7 MHz.

Any help is appreciated...

 

[crutschow]

The circuit in Link 1 would seem that it should work for you. Why did it not? Describe how you built it.

 

[chuckey]

What you do is to feed your signal into a pin of a voltage comparator chip, the other has a capacitor to ground with a high value DC leak (bias) across it. Into this input you couple via a small capacitor the output from a monostable which is fired by the output of the comparator. So no charge on storage cap, input pulse comes along, mono fires and puts a little voltage into the storage cap. Next peak comes along, it repeats. Eventually the voltage on the storage cap exceeds the input voltage peak and the mono does not fire until the storage cap has discharged a bit. So the absolute accuracy is coupling cap/storage cap which can be .1% with excellent linearity. Its the BBCs idea from the 1960s.
Frank