Precise zero crossing detector with zero phase shifting filter

[udhay_cit]

I'm in need of a very precise (zero phase shifting) zero crossing detector for 50Hz AC line. I have tried with some of ZCD circuit but the output square wave has some distortion (jitter like effect).
The application of this circuit is TanDelta measurement. The ZCD circuit required for both voltage & current waveform so there should not be phase shifting by introducing a pass filter (capacitor) or step down transformer (inductor) for the filtering application.

I consider three way of eliminating the noise.
1. Using a DSP to remove the noise at the zero crossing level by sample & average.
2. Converting the AC waveform into dc by using ZCD circuit. The jitter present in the zero cross level can be eliminated by a microcontroller at digital level without involving high speed ADC (or DSP).
3. Feeding the output of ZCD to a jitter attenuator IC chip. but i can't find a jitter attenuator IC for 50Hz yet.

I don't know which method is good. Please advice me to handle the problem.

Regards
Udhay

 

[FvM]

If the measurement is fixed to 50 Hz, a bandpass filter can provide zero phaseshift. Identical low-pass filters for current and voltage will also result in zero differential phase error, even for varying frequency.

 

[schmitt trigger]

I'm with FvM on this one.
If the current and voltage measurement channels have identical filters, the relative phase between the two will remain at zero, or extremely close to zero.

By using a pair of switched capacitor filters, driven from the same clock source, you can easily achieve that.

 

[crutschow]

Have you considered a PLL with a zero phase-shift (digital memory type) phase detector such as a CD4046 to minimize jitter?

 

[udhay_cit]

Hi FvM, schemitt trigger, Thanks for the reply. I will try with your suggestion. The simulation circuit gives filter output with near to zero offset. But if we use nearest standard value resistor & capacitor, there we can see the phase shift. Anyway i will build two different circuit & compare the result.

Have you considered a PLL with a zero phase-shift (digital memory type) phase detector such as a CD4046 to minimize jitter?

Can you please explain little more about your suggestion? I studied PLL circuit can be used to eliminate the jitter but i don't know how to do?

 

[crutschow]

..............................
Can you please explain little more about your suggestion? I studied PLL circuit can be used to eliminate the jitter but i don't know how to do?

On the CD4046 you use a squared input (from a comparator connected to the 50Hz) to control the internal VCO oscillator set to a center frequency of 50Hz using the the type II digital phase detector. You use a long time-constant filter on the phase-detector output to minimize jitter on the VCO output.