A 0.05 Hz Integrated Low Pass Filter

[timo67]

300 hz low pass filter

Has any one an idea about how to design a LPF (1st. order) with a 0.05Hz cut-off frequency... But in an integrated circuit (0.35u AMS techno)?
To reach such frequency you need huge capacitor or huge resistor
Now I am trying to design this filter in swiched cap
To reach the time constatnt of ≈3s needed and with a sampling clock of 300 Hz frequency (less than 300 Hz reduces the banwidth...) you have to draw two capacitors with a ratio of 1000!
Parasitic capacitors induce output offset ...

Has any one an advice?
Thanks

 

[borislee]

high pass filter 0.05hz

just a thought for your information:
change the input signal to a digital signal, use a digital filter to realize your requirement of 0.05Hz cut-off LPF, then change the result to analog signal and output

 

[sengyee88]

2hz low pass filter

To perform 0.05Hz cut-off LPF, in general it is difficult if no external components are allowed.

For discrete time approach, eg switched cap, you will need a low sampling clock and small switched capacitor. Low sampling clock itself will consume large area.

For continuous time approach, eg gm-C filter, you will need a very low gm linear OTA, which is at nA/V range. This requires nA range current biasing. Some low gm techniques need to apply and transistor may need to operate at weak inversion.
You may look for the following papers if you decided to go for continuous time approach:
1. A 60-db dynamic-range CMOS sixth-order 2.4-hz low-pass filter for Medical Applications, IEEE Transactions on Circuits and Systems -II, Vol 47 No.12 Dec 2000
2. Transconductance Amplifier Strucutres with very small transconductances: A comparative design approach, IEEE Journal of Solid-State, Vol 37 No.6 June 2002

 

[wylee]

low pass filter 3 hz

2. Transconductance Amplifier Strucutres with very small transconductances: A comparative design approach, IEEE Journal of Solid-State, Vol 37 No.6 June 2002

 

[cougar168]

2 hz digital low pass filter

At first, I thought this was a joke, but ...

I would use a DC amplified first, then, use an AD converter, then, use DSP
filtering

I doubt any analog based filter will work well at such low frequency

 

[mince]

1 hz low pass filter

Some guys in my class did something similar in the spring. They used long channels for high output imedances and for the capacitor they used a capacitor multiplier circuit.

 

[timo67]

low pass filter at 1 hz cutoff

To cougar168
No it is not a joke
Such filter is used in ECG measurment to supress a DC or very low frequency parasitic signal du to electrodes

 

[okguy]

cut-off frequency 0.05hz

Use the Miller effect to amplify a capacitor.
It works well, but it is not accurate due to the process variation on the gain.

 

[wenadinho]

2 hz low-pass filter

try pseudoresistor configuration, in Reid Harisson paper. technically for small signal swing, a linear terra ohm resistance is achievable with the pseudoresistor. The integrated capacitor, is thus, can be made very small, e.g. below 20 pF

 

[okguy]

20 hz lpf

But then, the problem will be leakage currents, espacially for sub-micron processes :|

 

[laglead]

how to design a 10hz bpf

huge capacitor is a must.

 

[blue_moon]

1 hz low pass filter discrete

Search on IEEE for a paper on Filter&Hold I think Iit might help...