Isolating Ac and Dc Signal

[mustafapk]

Hello,
I am working on a project in which i have to isolate the Ac voltages of low amplitude and low frequency (100uV and 1Hz) from a continously decreasing Dc value. i have built just a smiple High pass filter type circuit. but the problem is that due to very low value of amplitude and frequency either the signal doesnt passes or when i reduce the cutoff freqency of the high pass filter the effect of Continously reducing dc voltages also prduces its effect on the output.
One more thing that i also mention here is, when i isolates these voltages they are effected from the Hum noise(50Hz Ac noise.) even i have used 8th order low pass filter of 10Hz cutoff frequency.
I want a technique through which i can completly isloate the Ac signal which isnt effected by the Dc reduction value.
Anyone please help me in solving it.
Thanks

 

[mustafapk]

i am surprised why people havent answered such an interesting question on this forum.

 

[LvW]

Hello,
..............
I want a technique through which i can completly isloate the Ac signal which isnt effected by the Dc reduction value.
Thanks

What do you mean with "isolate" ? Is it your task only to REMOVE the ac signal from the decreasing DC voltage or to extract it for further processing ?
As a first guess I would recommend a high-Q active bandpass circuit. But its not easy to design it for a center frequency of 1 Hz.

 

[kumarghz]

Mr Mustafa,
Can you upload your circuit and waveforms if you have. It will make discussion easier.

 

[mustafapk]

Thanks for replying
Mr. Lvw. Yes i have to get the ac for processing. i have connected a high pass circuit. but the problem is that it doesnt exactly seperate the ac voltages and the slope of dc also gives some changes in it.

Mr. Kumarghz. i didnt make any circuit diagram on jpg. but i think its very simple i am getting my ouput from a sensor which ranges from 0.1 to 2V DC.

 

[LvW]

To separate an ac signal from a (slighty changing) dc voltage requires a highpass or a bandpass filter.
The problem is the rather low frequency of 1 Hz as it is not easy exactly to design a filter for such a low corner frequency. Because of the varying dc component the Q of the filter must be rather high (let´s say: 10 for instance).
Perhaps a good and exact solution is a commercially available switched-capacitor filter.

 

[mustafapk]

Perhaps a good and exact solution is a commercially available switched-capacitor filter.

Do you think i will be successful in my target after using switched capacitor filter. as i have already used high pass active filter.
Thanks

 

[LvW]

What is the dc decreasing rate in comparison to the 1 Hz signal ?
This figure determines the minimum Q of the filter.
Did you perform already some circuit simulations (with high pass or bandpass) for different pole Q´s ?

 

[simen]

hello,
i dont knw whether it will help or not..... but i am doin a proj where i had 2 isolate ac and dc from a mixture (low freq ac) .... wht i did ws, and it worked preety good too, i first split the signal and used a simple op-amp LM347 in one branch to build a dc extractor (if u dnt find this in net, mail me, i will snd u the dc extractor ckt) and in the othr branch, i used a large capacitor to extract the low freg ac....