...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 |
+1 m
With two poles and a cutoff at 35Hz, you're not going to see much attenuation at 60Hz, but you _will_ see some significant phase effects.
-- Tim Wescott Wescott Design Services http://www.wescottdesign.com I'm looking for work -- see my website!
What's often bigger than 60 Hz is the harmonics of 60 Hz. So lowpass is appropriate.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Thread drift... does anyone know what the step (or impulse) response of a notch filter looks like?
George H.
It's about the same as the ringing of the complementary resonator of the notch. Besides, there are plenty of phase / group delay contortions around the notch.
I just wonder what may the OP's signal be, EKG?
-- -TV
Oh, that's a bit disappointing.
Yes, it is an "EKG type" signal. But for a unique purpose I am trying to find an analog design that introduces the least phase shift and distortion. That's not usually a criteria for similar signals. Hence, no ready made design available.
Seems like there is always a trade-off.
I can't get enough Q without more poles which result in more phase shift ... which is what I am trying to avoid.
Cut off frequency being close to the bandwidth of interest, etc. As usual I have painted myself into a dilemma.
Keven Foster
An elliptical filter, a lowpass with a notch or two, might be better. You could combine a soft, Bessel-like lowpass with a notch around 60 Hz. That would be a nonstandard (Bessel-like-elliptical) filter.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
+1. I'd consider notches at 60, 120, and 180Hz -- it seems like whenever you suppress 60Hz, you just find all the other harmonics vying for your attention.
Come to think of it, if you're going into a microprocessor and if it's sufficient, you might want to consider sampling at some robust multiple of 60Hz, and then averaging and decimating down to 60Hz in software. This automagically gives you a comb filter with zeros at 60Hz and all of its harmonics, with very regular phase characteristics.
-- Tim Wescott Control systems, embedded software and circuit design I'm looking for work! See my website if you're interested http://www.wescottdesign.com
A notch filter will do a lot more for 60 Hz than a LPF, and if it's sharp enough, it won't contribute much phase shift at all below 30 Hz. I've used that trick fairly often over the years.
If you can make one with a Q of 30 (which will require good parts), the phase shift will all be happening within a few hertz of 60. I suggest a gyrator to make a grounded inductance, with a capacitor in series, and a resistor from there to the filter input. You'll want to keep the LPF, of course, and may need a 120-Hz notch as well.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
if it doesn't have to be realtime another trick is to run it through a filter twice once forward once reverse, the squares the amplitude response response and removes the phase shift
There are fundamental limitations to normal linear filtering approaches. This includes notch filters. One different approach (but may be more complicated than you can tolerate) is exemplified by the "Humbug" product:
There have been some fairly cheap, moderately simple approaches that have been published on the internet (can't find my references to these at the moment, sorry). They're all more complicated than a 2-pole analog filter.
The lowpass part could kill the higher harmonics.
Digital filtering can do most anything. But we don't know if the OP is even digitizing.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
An ideal lowpass filter (exactly unity gain and no phase shift below Fc, zero output above Fc) is impossible because it's not causal; it lets you buy stocks that you know will go up. But a digital approximation of an ideal filter, with enough time delay to avoid paradoxes, is legal.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
If a digital filter is outside your domain of experience, then maybe that isn't the way to go. But the circuit would be very simple....
R1 +---------+ R2 Input --->>---/\/\/---+---| Digital |---\/\/\---+--->>--- Output | +---------+ | C1 === | === C2 | --- | --- - --- - -
R1-C1 and R2-C2 are anti-alias filters which should be sized to not interfere with the pass band characteristics. With a sample rate well above the Nyquist determined rate, say 100 kHz, the anti-alias filters would not need to be more than one pole unless you have very stringent out of band rejection criteria.
The digital portion of the filter can be either the switched capacitor filter you originally asked about. Or if you are willing to learn how to code a filter, you can use an MCU or DSP for the digital block. An MCU could do the job easily and give a very low cost solution.
-- Rick C
Maybe what you are asking for is not the right way to go about things. What do you plan to do with the signal once you have it? If the 60 Hz interference and harmonics aren't too large, perhaps there is a way to do what you want without removing them up front?
-- Rick C
** The OP has been here before with EXACTLY the same idiot question - has no-one here got a good memory ?
He is trying to take electrical readings from a plant.
He has almost zero electronics knowledge.
He appears to be mentally defective as well.
All he needs to do is surround his stupid plant with a Faraday cage of some kind - it will block the 60Hz based interference and still let light through.
Of course, he has no understanding of the need for an instrumentation amplifier or the benefits of a good CMRR.
His imagined need for low phase shift is bunkum as well.
.... Phil
A lot of times that's the point to take a step back, and ask if you can stop the 60 Hz from getting into signal path, rather than filter it later.... attack the source, if you can.
George H.
+1,+1... I like it all. I'm not much on sampling, but I'll often set "magnetic" triggering rates at 6.000... Hz (digital function generator's are great.) I guess I'm sampling with my 'scope.
George H.
My memory sucks,
If it's basically electro-statics, then as you say does he have a "chicken wire" screen at least?
If he's really measuring some voltage/ current in plants, then I'm much more concerned about his probes... DC and water becomes chemistry in a hurry,
George H.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.