Inverting Sallen-Key low pass filter

[Robert Stevens]

Can someone please describe the circuit topology for an inverting two
pole Sallen-Key (or similar) low pass filter, or point me to a
reliable diagram?

There are plenty of circuits online for the non-inverting type, but I
am unclear about how the inverting type is built.

I would also like to know how to caluclate the component values for
500Hz cut-off, if this process differs from the non-inverting variety.

Robert Stevens

[John Larkin]

On Mon, 05 Oct 2015 13:42:14 +1100, Robert Stevens <rste...@tsci.edu>
wrote:

S-K is inherently non-inverting. Maybe you mean the MFB (multiple
feedback) topology.

TI's FilterPro program is handy. It accounts for opamp GBW, and can
select standard R and C values.

Google MFB filter

[Robert Stevens]

It appears the MFB filters are bandpass though, not the low pass I was
looking for.

Is there any configuration of two pole active low pass filter that is
inverting?

Thanks,

Robert Stevens

[piglet]

On 05/10/2015 07:23, Robert Stevens wrote:
>
> It appears the MFB filters are bandpass though, not the low pass I was
> looking for.
>
> Is there any configuration of two pole active low pass filter that is
> inverting?
>

MFB can be Low Pass. For example see fig 2. page 7 of this:

<http://www.ti.com/lit/an/sloa049b/sloa049b.pdf>

piglet

[Robert Stevens]

On Mon, 5 Oct 2015 11:53:30 +0100, piglet <erichp...@hotmail.com>
wrote:

Thanks. That is exactly what I was looking for.

Robert Stevens

[bitrex]

John Larkin <jjla...@highlandtechnology.com> Wrote in message:

With a 5 volt power supply, you can make a fairly decent MFB
filter using the TL431 in a pinch!

--




----Android NewsGroup Reader----
http://usenet.sinaapp.com/

[George Herold]

The state variable filter can be made both inverting and non-inverting.
but it takes three opamps.

George H.

[Phil Hobbs]

MFB filters can have pretty high component sensitivity if you do them
wrong. Unity-gain SKs with equal or nearly-equal resistors are very
well behaved unless the Q is large.

It's pretty educational to re-calculate the filter with components at
the tolerance limits, of course including the tempcos of the Rs and Cs,
and the voltage coeffients of the Cs.

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

[John Larkin]

On Mon, 5 Oct 2015 10:09:30 -0400, Phil Hobbs
<pcdhSpamM...@electrooptical.net> wrote:

>On 10/05/2015 07:58 AM, Robert Stevens wrote:
>> On Mon, 5 Oct 2015 11:53:30 +0100, piglet <erichp...@hotmail.com>
>> wrote:
>>
>>> On 05/10/2015 07:23, Robert Stevens wrote:
>>>>
>>>> It appears the MFB filters are bandpass though, not the low pass I was
>>>> looking for.
>>>>
>>>> Is there any configuration of two pole active low pass filter that is
>>>> inverting?
>>>>
>>>
>>> MFB can be Low Pass. For example see fig 2. page 7 of this:
>>>
>>> <http://www.ti.com/lit/an/sloa049b/sloa049b.pdf>
>>>
>>> piglet
>>>
>>
>> Thanks. That is exactly what I was looking for.
>
>MFB filters can have pretty high component sensitivity if you do them
>wrong. Unity-gain SKs with equal or nearly-equal resistors are very
>well behaved unless the Q is large.
>
>It's pretty educational to re-calculate the filter with components at
>the tolerance limits, of course including the tempcos of the Rs and Cs,
>and the voltage coeffients of the Cs.
>
>Cheers
>
>Phil Hobbs

High-Q active filters will always be twichey, some less than others.

Apparently there's some magical interaction inside an LC filter that
tames the sensitivities.

[John Larkin]

And get a lifelong exclusion from polite society.

[Phil Hobbs]

The magical interaction is the square-root dependence, i.e. a
sensitivity coefficient of 1/2. You never get below 1.0 for active
filters AFAIK.

omega_0 = 1/sqrt(LC)

Q = sqrt(L/(R**2 C))

[Phil Hobbs]

One of my daughters used to live in the Vieux Carr\'{e}, at Dauphine &
St. Peter. She said that it was very public-spirited of some of the
male tourists to wear T-shirts covered with various misogynistic
slogans, because it saved women the unpleasantness of actually meeting them.

[John Larkin]

New Orleans attracts all the redneck party-types from Texas to
Florida.

[Frank Miles]

On Mon, 05 Oct 2015 10:09:30 -0400, Phil Hobbs wrote:

> On 10/05/2015 07:58 AM, Robert Stevens wrote:
>> On Mon, 5 Oct 2015 11:53:30 +0100, piglet <erichp...@hotmail.com>
>> wrote:
>>
>>> On 05/10/2015 07:23, Robert Stevens wrote:
>>>>
>>>> It appears the MFB filters are bandpass though, not the low pass I was
>>>> looking for.
>>>>
>>>> Is there any configuration of two pole active low pass filter that is
>>>> inverting?
>>>>
>>>
>>> MFB can be Low Pass. For example see fig 2. page 7 of this:
>>>
>>> <http://www.ti.com/lit/an/sloa049b/sloa049b.pdf>
>>>
>>> piglet
>>>
>>
>> Thanks. That is exactly what I was looking for.
>
> MFB filters can have pretty high component sensitivity if you do them
> wrong. Unity-gain SKs with equal or nearly-equal resistors are very
> well behaved unless the Q is large.
>
> It's pretty educational to re-calculate the filter with components at
> the tolerance limits, of course including the tempcos of the Rs and Cs,
> and the voltage coeffients of the Cs.
>
> Cheers
>
> Phil Hobbs

As you suggest, the S-K component sensitivities for Q scale with Q,
i.e. they get fairly bad for Qs higher than a small fraction. This
is consistent with Budak (see p. 181).

I disagree about MFB sensitivities - wo and Q sensitivities for the
MFB are |S|<= 0.5, though gain sensitivities are (surprise!) |S|=1.
Gain sensitivites for S-K forms are no better excepting gains near unity.

Of course if low sensitivities are desired, the dual-integrator forms are
best.

I'm sure you'll be happy to tell me where I'm wrong about these...

[krw]

On Mon, 05 Oct 2015 17:23:13 +1100, Robert Stevens <rste...@tsci.edu>

wrote:

>On Sun, 04 Oct 2015 20:07:39 -0700, John Larkin
><jjla...@highlandtechnology.com> wrote:
>
>>On Mon, 05 Oct 2015 13:42:14 +1100, Robert Stevens <rste...@tsci.edu>
>>wrote:
>>
>>>
>>>Can someone please describe the circuit topology for an inverting two
>>>pole Sallen-Key (or similar) low pass filter, or point me to a
>>>reliable diagram?
>>>
>>>There are plenty of circuits online for the non-inverting type, but I
>>>am unclear about how the inverting type is built.
>>>
>>>I would also like to know how to caluclate the component values for
>>>500Hz cut-off, if this process differs from the non-inverting variety.
>>>
>>>Robert Stevens
>>
>
>>S-K is inherently non-inverting. Maybe you mean the MFB (multiple
>>feedback) topology.
>>
>>TI's FilterPro program is handy. It accounts for opamp GBW, and can
>>select standard R and C values.
>>
>>Google MFB filter
>>
>
>It appears the MFB filters are bandpass though, not the low pass I was
>looking for.

It's whatever you want, in two poles. Take John's advice and get
FilterPro from TI. It'll do whatever SK or MFB filter you want.

>
>Is there any configuration of two pole active low pass filter that is
>inverting?
>

Yes. Look!

[Simon S Aysdie]

With regard to "matched" doubly terminated passive filters, they are also insensitive because this is the point of "maximum transfer of power." The slope of deltaP is zero, or near zero, for most of the passband. Therefore, any deltaL or deltaC with respect to the zero, or near zero, slope of power transfer (deltaP) is necessarily insensitive.

Of course, it is all relative. The nearer the natural modes are to the jw axis, the more sensitive it will be, ceteris paribus. "Unmatched" LC filters are more sensitive than matched ones, again ceteris paribus.

By matched I mean there is ideally no reflected power in the passband aside from the unavoidable dips in the eps ripples.

[Phil Hobbs]

I've never gone through the math of MFB filters, but I don't think we
actually disagree. I was comparing MFB filters "if you do them wrong"
with "unity-gain SKs with equal or nearly-equal resistors".

The OP seemed to be building his first active filter, or very nearly, so
a few cautions seemed in order.

[Tim Wescott]

"Passive and Active Network Analysis and Synthesis" by Aram Budak,
Houghton-Mifflin 1974. It was old when I was told off to buy it for a
class, but I still pull it out from time to time today to get valuable
information. If you can get it -- get it.

It goes into detail on the filter you're considering making.

(That was a great class. Usually gut classes make me want to barf, but
that was an easy A that taught me something new every day I was in the
class -- which was a godsend because I was taking the World's Hardest
Class at the same time, and I needed something easy to offset it.)

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

[George Herold]

OK I'll bite, what was your "Worlds Hardest Class"?
(your caps.)
(I got a C in organic chem.)
George H.

[Carl Ijames]

"George Herold" wrote in message
news:a68799dd-86de-4125...@googlegroups.com...

=============================================================

There's a reason the only "Honk if you passed xxx" bumper stickers the
American Chemical Society sells is for pchem, physical chemistry :-) :-). I
worked as hard or harder in pchem than any other class.

-----
Regards,
Carl Ijames

[Phil Hobbs]

> There's a reason the only "Honk if you passed xxx" bumper stickers the
> American Chemical Society sells is for pchem, physical chemistry :-) :-). I
> worked as hard or harder in pchem than any other class.

As an old friend of mine (a very accomplished physical chemist) used to
say,

"A physical chemist is someone who talks physics to chemists,
chemistry to physicists,
and baseball to other physical chemists."

[Bill Sloman]

I ended up talking electronics to all of them. My physical chemistry classes did include thermodynamics, and it took me longer than usual - not all that long - to get my head around that, but when both your parents have bachelors degrees in chemistry, you do hit the ground running.

My only post-doc supervisor finally made Fellow of the (British) Royal Society a few moths ago, and he is a physical chemist/photochemist. As a post-doc I built some electronics for him and his graduate students, and rather more after I'd moved into industry, strictly as hobby projects.

--
Bill Sloman, Sydney

[Carl Ijames]

"Bill Sloman" wrote in message
news:460972ed-2bf2-4fd4...@googlegroups.com...

==============================================================================

My pchem classes (3 quarters my junior year) included thermodynamics, and
then finished with quantum mechanics. As an undergrad I really liked lots
of different things and had a hard time deciding on what to major in between
chemistry, computer science, electrical engineering, and maybe optics. I
finally decided that if I went to grad school as an analytical chemist I
could build and work with nice instrumentation, and do all of the
electronics, programming, uhv vacuum system design, etc. that I wanted, and
any of it I didn't want to do I could just say "I'm a chemist, you need a
____ for that" :-), so I wound up with a double major in chemistry and
computer science (took all the ee classes a comp sci major would take, just
skipped a few math and physics classes or I'd of had a double degree).
Between grad school and later I've gotten to do a lot of chromatography-mass
spectrometry interfacing, a good bit of uhv design and construction, some
ion optics work (mostly transporting ions into a superconducting magnet for
trapping and analysis), a good bit of control electronics (mostly
straightforward digital stuff) and programming, and I've designed and built
two complete mass specs from scratch so far. First quarter of grad school I
was told to take the graduate quantum mechanics and advanced graduate
organic courses since they were all that were available at the time; they
weren't teaching any analytical courses that quarter. The first day the
prof told us that only pchem grad students ever got an A from him, the best
the rest of us could hope for was a B. It took a lot of work, but I got a
B+ and considered it a victory; I was very happy that that was my last pchem
class :-). Anyway, enough reminiscing, back to mostly lurking and learning.

-----
Regards,
Carl Ijames

[Tim Wescott]

I can't remember the exact name of the class, but the book is Van Trees
"Detection Estimation, and Modulation Theory".

In one of the homework problems it has you derive Kalman filtering from
basic principles. In another, Kalman-Bucy filtering.

I still get occasional contracts to do weird stuff at the intersection of
communications and metrology that I wouldn't be able to touch if it
weren't for that class. In fact, I'm working on one right now.

[Jim Thompson]

I had a radar course from Harry L. Van Trees while he was still a grad
student... took us on a field trip to a Nike base in NH... where they
practiced locking-in on aircraft landing at Boston's Logan Airport ;-)

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

[George Herold]

Awesome! thanks. If I ever have any metrology/ communications problems
you'll be the first one I contact. (Don't hold your breathe.)

I did see this Ladyada, Paul Horowitz video today, where he talks about
AoE3.. fun stuff, Appendix I is about TV, digital TV looks like
white noise compared to the old analog AM ... sad news for SETI, unless someone is trying to communicate.

George H.

[Tim Wescott]

Yup. I was listening to some SETI guys talking and they were all over
the notion of detecting alien broadcasts because they'd be narrow band --
we only started in on narrow-band RF comms in 1915; it's only been 100
years and we're going more and more to spread spectrum.

Assuming that the little green men out there are basically like us,
that's a pretty small window of time to catch them in the act.

(and for all we know, narrow band RF signals means "we're tasty and
defenseless right now!" to someone out there).

[Bill Sloman]

Happily, the narrow-band period of signalling tasty and defenseless seems to have taken more than 100 years to get to anybody hungry and offensive-minded.

Maybe some Dubbya-equivalent in a nearby star-system is arguing that we already have weapons of mass-destruction, and needed to be pacified before we get bigger ones. If they can measure our atmospheric CO2 levels they will have already worked out that we are dangerously irresponsible.

--
Bill Sloman, Sydney

[joe hey]

On Tue, 06 Oct 2015 20:53:14 -0700, Bill Sloman wrote:

}snip{

> Maybe some Dubbya-equivalent in a nearby star-system is arguing that we
> already have weapons of mass-destruction, and needed to be pacified
> before we get bigger ones. If they can measure our atmospheric CO2
> levels they will have already worked out that we are dangerously
> irresponsible.

Increasing CO2 levels are the only option to be able to keep feeding our
growing population. ;-)

joe

[Bill Sloman]

Until the population crash that's going to happen when our agriculture - dependent on specific plants carefully selected over ten thousand years of regular interglacial weather - collapses when our plants decide they don't like the new climate.

10 metre higher sea levels - after the Greenland and West Antarctic ice sheets decide that they are going to slide off into the ocean in big chunks, like the Canadian ice sheet did at the end of the last ice age - are going to flood every port we've got, so shipping food from places that can still grow it to places that need it may get difficult too.

I'm probably going to be dead before it gets interesting, but while your ignorance doesn't guarantee that you are young - Jamie is both elderly and ill-informed - it seems likely that you might have to cope with it.

--
Bill Sloman, Sydney

[Tim Wescott]

On Tue, 06 Oct 2015 20:53:14 -0700, Bill Sloman wrote:

Personally I think that there are three possibilities:

1: We really are alone for one reason or another.

2: We're about as interesting to the ones out there as a 1st
grader is to a 6th grader

3: The solar system is a designated wildlife (or maybe game)
preserve.

[Jim Thompson]

And Slowman is insane. Don't feed the troll.

[Phil Hobbs]

Graduate quantum field theory (relativistic quantum mechanics). The
class wasn't the very hardest, except that the guy who cajoled me into
taking it in the first place bailed out after a couple of months,
leaving me with no study partner. Having a wife and daughter, I needed
to graduate ASAP, so I had to bail too. (He's now the beam-time czar at
Brookhaven.)

A pity--it was a great class.

Cheers

Phil Hobbs

[Phil Hobbs]

Or, as seems most likely,

0. As expected from Einstein's causality argument, there really is no
way to accomplish FTL travel, and sending an object interstellar
distances on organism-relevant time scales requires converting a major
fraction of its mass into energy.

[Tim Wescott]

That could be.

I think if we want to better dodge mass extinction events than we can now
-- even ones not of our own creation -- we want to spread off the
planet. If we could colonize an asteroid and power it from fission I'm
pretty sure we could get someplace interesting on a species-relevant time
scale if not an organism-relevant one, thus dodging even more mass
extinction events.

[krw]

On Wed, 07 Oct 2015 15:43:51 -0500, Tim Wescott

4: The universe is big. Really, really, big.

[Tim Wescott]

True, but as we're finding out on a daily basis, there's lots of planets
out there. Really, really, lots.

If nothing else, modesty would insist that we're not unique.

[Jim Thompson]

On Wed, 07 Oct 2015 18:18:38 -0500, Tim Wescott

Heaven help us, but there's probably another Slowman out there >:-}

[Bill Sloman]

> Heaven help us, but there's probably another Sloman out there >:-}

Whereas you can find pond scum like Jim Thompson in any sewage farm anywhere around the world.

--
Bill Sloman, Sydney

[Bill Sloman]

My feeling is that there aren't any hard classes, just a some badly
taught classes.

Some stuff - like thermodynamics and relativistic quantum mechanics - is
hard to teach well, because it does require the student to learn an
unusual way of looking at the problems involved, but good teachers can
do it.

Egomaniac teacher go out of their way to make it hard, because they look
cleverer that way.

--
Bill Sloman, Sydney

[M Philbrook]

In article <inab1bp94qa3vqegn...@4ax.com>, To-Email-Use-
The-Enve...@On-My-Web-Site.com says...

> >
> >True, but as we're finding out on a daily basis, there's lots of planets
> >out there. Really, really, lots.
> >
> >If nothing else, modesty would insist that we're not unique.
>
> Heaven help us, but there's probably another Slowman out there >:-}
>
> ...Jim Thompson
> --
> | James E.Thompson | mens |
>
>

I hope not!

Jamie

[George Herold]

4: We don't know enough yet to "look" in the right place.
(say you wan't to communicate with some planet, ~50
light years away? How many stars are within a
50 light year radius?

http://www.solstation.com/stars3/100-gs.htm
Ask and the web shall deliver.
not many nearby stars.

George H.

[krw]

On Wed, 07 Oct 2015 18:18:38 -0500, Tim Wescott

Not the point. You can't get here from there so who cares how many
there are?

>If nothing else, modesty would insist that we're not unique.

Modesty should tell you to mind your own business. ;-)

[M Philbrook]

In article <bdgb1bpd2gjrlrkab...@4ax.com>, k...@nowhere.com
says...

> >>3: The solar system is a designated wildlife (or maybe game)
> >>> preserve.
> >>
> >> 4: The universe is big. Really, really, big.
> >
> >True, but as we're finding out on a daily basis, there's lots of planets
> >out there. Really, really, lots.
>
> Not the point. You can't get here from there so who cares how many
> there are?
>
> >If nothing else, modesty would insist that we're not unique.
>
> Modesty should tell you to mind your own business. ;-)
>
>

You know what I tell them?

"If you can't hear a zipper, shut up!"

Jamie :)

[Ralph Barone]

So just out of curiosity, if we managed to load 1000 humans onto a nuclear
powered asteroid space ship before whatever catastrophe wiped humanity off
Earth, would that actually make you feel better?

[mixed nuts]

Probably not NH - there weren't any. There were a dozen or more around
Boston. The northern most batteries were in Topsfield, Danvers and
Beverly MA. Pretty rural - cows, trees, pastures and whatnot - easy to
mistake for the southern part of NH where they still talk Boston-funny.

Yup. We'd look at whatever aircraft were about - not close-in though -
didn't want to interfere with navigation or communications. There was a
simulator available but there was only one per battalion so it might be
a month or more between full-scale training sessions where practice
didn't need live radar.

--
Grizzly H.