Re: Response curves of gyrator bandpass filter

Discussion in 'Electronic Design' started by analog, Apr 21, 2005.

  1. analog

    analog Guest

    The Phantom wrote:

    > These two images show the response of the gyrator bandpass filter
    > shown in the file, "GyratorFilters-BP-LP-HP", on Jim Thompson's web
    > site. I used an op amp model with DC gain of 1000000 and a single
    > pole rolloff. I then swept the GBW from 25 kHz to 100 MHz and
    > plotted the circuit response. One of the plots had excess phase
    > added to the op amp transfer function using the excess gain block
    > from Jim's OP27 model.
    >
    > The difference caused by excess phase is not visible on these plots,
    > but a careful examination of the circuit response at any particular
    > frequency, carefully compared with and without excess phase, does
    > show a small change caused by the excess phase.
    >
    > [Image]


    Hi Phantom,

    Beautiful images. Great work!

    Once again you set the newsgroup straight.

    It looks like JT's favorite gyrator filter configuration is rather
    sensitive to gain bandwidth non-idealalities in the opamps. It also
    looks like excess phase (delay) plays very little roll (contrary to
    the claims of another poster).

    Of the various opamp/gyrator configurations/combinations are there
    any that are less sensitive to opamp GBW than Jim's "optimum"
    configuration? Maybe with some, opamp non-idealalities tend to
    cancel each other?

    Regards -- analogspiceman
     
    analog, Apr 21, 2005
    #1
    1. Advertising

  2. analog

    Jim Thompson Guest

    On Thu, 21 Apr 2005 07:05:56 -0700, analog <> wrote:

    [snip]
    >
    >It looks like JT's favorite gyrator filter configuration is rather
    >sensitive to gain bandwidth non-idealalities in the opamps. It also
    >looks like excess phase (delay) plays very little roll (contrary to
    >the claims of another poster).
    >
    >Of the various opamp/gyrator configurations/combinations are there
    >any that are less sensitive to opamp GBW than Jim's "optimum"
    >configuration? Maybe with some, opamp non-idealalities tend to
    >cancel each other?
    >
    >Regards -- analogspiceman


    If you'll recall, many posts ago, my gyrator configuration was
    compared to a popular configuration (Antoniou ??).

    For a given GBW, my configuration fared better.

    Somehow I didn't read Phantom's posts as saying excess-phase played no
    role (watch your spelling :)

    But it's for certain that GBW does. Scratching my head, and trying to
    recall old discussions, I believe that inserting a series R, where

    R = 1/(2*pi*GBW*C)

    improves peaking tremendously.

    ...Jim Thompson
    --
    | James E.Thompson, P.E. | mens |
    | Analog Innovations, Inc. | et |
    | Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
    | Phoenix, Arizona Voice:(480)460-2350 | |
    | E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
    | http://www.analog-innovations.com | 1962 |

    I love to cook with wine. Sometimes I even put it in the food.
     
    Jim Thompson, Apr 21, 2005
    #2
    1. Advertising

  3. analog

    Joel Kolstad Guest

    "analog" <> wrote in message news:...
    > It looks like JT's favorite gyrator filter configuration is rather
    > sensitive to gain bandwidth non-idealalities in the opamps.


    It's probably safe to say that in ANY active filter design, GBW is the most
    critical parameter as to whether or not thet hing's going to work.

    > It also
    > looks like excess phase (delay) plays very little roll (contrary to
    > the claims of another poster).


    I'll take a look at Phantom's results later today and see what I can glean...

    > Of the various opamp/gyrator configurations/combinations are there
    > any that are less sensitive to opamp GBW than Jim's "optimum"
    > configuration?


    Sensitive in what way? In Jim's case, he's using a gyrator to simulate an
    inductor, so about the only sensitivty you can talk about is that of the
    inductor's inductance and loss (or Q, if you'd like) with respect to GBW.
    However, in the more general case where you're trying to build an active
    filter (using a gyrator, a bi-quad, etc.), you usually see people computing
    sensitivites with respect to center frequency, bandwidth (or Q again), etc.

    > Maybe with some, opamp non-idealalities tend to
    > cancel each other?


    The Moss-Ackerberg arrangement uses this approach, although (like many
    bi-quads) there are multiple components that need to track each other to
    change the center frequency.

    ---Joel Kolstad
     
    Joel Kolstad, Apr 21, 2005
    #3
  4. analog

    Joel Kolstad Guest

    Jim,

    "Jim Thompson" <> wrote in message
    news:...
    > But it's for certain that GBW does. Scratching my head, and trying to
    > recall old discussions, I believe that inserting a series R, where
    >
    > R = 1/(2*pi*GBW*C)
    >
    > improves peaking tremendously.


    Yes it does; I've verified this in my own simulations, and of your your PDF
    file shows it as well.

    You actually used 2/(2*pi*GBW*C) in your PDF file's circuit. I was hoping you
    happened to already have a derivation of where this came from so that I could
    be lazy and not have to try to analyze it myself. :) Actually, I don't mind
    the analyzing it, but I'm not really sure how to 'isolate' the op-amp that has
    the capacitor in it from everything else (since, unlike the integrator, where
    R=1/(2*pi*GBW*C) is correct as shown in Schaumann/Van Valkenburg, in the
    Gyrator the V+ input of the op-amp isn't grounded).

    ---Joel
     
    Joel Kolstad, Apr 21, 2005
    #4
  5. analog

    The Phantom Guest

    On Thu, 21 Apr 2005 07:17:36 -0700, Jim Thompson <> wrote:

    >On Thu, 21 Apr 2005 07:05:56 -0700, analog <> wrote:
    >
    >[snip]
    >>
    >>It looks like JT's favorite gyrator filter configuration is rather
    >>sensitive to gain bandwidth non-idealalities in the opamps. It also
    >>looks like excess phase (delay) plays very little roll (contrary to
    >>the claims of another poster).
    >>
    >>Of the various opamp/gyrator configurations/combinations are there
    >>any that are less sensitive to opamp GBW than Jim's "optimum"
    >>configuration? Maybe with some, opamp non-idealalities tend to
    >>cancel each other?
    >>
    >>Regards -- analogspiceman

    >
    >If you'll recall, many posts ago, my gyrator configuration was
    >compared to a popular configuration (Antoniou ??).
    >
    >For a given GBW, my configuration fared better.
    >
    >Somehow I didn't read Phantom's posts as saying excess-phase played no
    >role (watch your spelling :)


    That's exactly what I'm saying. The two .jpg images attached to the first post in this
    thread (on ABSE; I was thinking the discussion might stay on SED, but now it's both
    places) show no visible difference in the peaking (or frequency shift) with or without
    excess phase. And just to be very careful about what I'm saying, it is that when I insert
    the excess gain block in the OP27 model shown on your web page in a model that otherwise
    has only a single pole rolloff as the op amp's frequency dependence, I see no difference.
    I said further on SED:

    "I simulated Jim's bandpass filter with and without the excess phase block he uses in his
    OP27 model. That excess phase block doesn't start to add excess phase until you get up to
    about 1 MHz, so I was wondering how it could cause peaking down at 20 kHz, which is the
    center frequency of the bandpass filter. After I got the model working, I decided to try
    to duplicate the family of response curves shown on page 3 of the .pdf file. I swept the
    GBW with smaller steps and noticed even more peaking at frequencies that Jim didn't show.
    In fact, with certain op amp parameters, you can get peaking of greater than 130 db (I got
    peaking of a factor of 3.5 million!). Just set the op amp DC gain to 1000000, the GBW to
    279069.248 Hz and you should see a huge peak at 17437.0011 Hz (this is without excess
    phase in the op amp model). With excess phase, the huge amount of peaking remains, but is
    *slightly* changed in frequency."

    In another post (on SED), I asked:
    "Are you saying that the peaking is caused by the excess phase shift?
    So that if we remove the excess phase shift from the op amp model, the peaking will go
    away?"

    And Joel Kolstad replied: "Yes it will!"

    But I don't get that result as can be seen in the images attached to the first post of
    this thread (in ABSE). I was hoping somebody else would run simulations and either verify
    or falsify my results.

    >
    >But it's for certain that GBW does. Scratching my head, and trying to
    >recall old discussions, I believe that inserting a series R, where
    >
    >R = 1/(2*pi*GBW*C)
    >
    >improves peaking tremendously.


    I see a mention of this on the fourth page of the file, "GyratorFilters-BP-LP-HP.pdf", on
    your web site although the formula given there is R = 2/(2*pi*GBW*C). But the first three
    pages show a filter with a nominal center frequency of 20 kHz whereas the fourth page
    shows a center frequency of about 30 kHz.

    On page 4 it says to insert a resistor in series with C1 of 2/(2*pi*GBW*C1) ohms, which
    works out to 3183 ohms for a GBW of 100 kHz, for example. When I do this, the peaking is
    much reduced, but the frequency shift is hardly reduced at all, and the center frequency
    for various GBWs remains at 20 kHz and below, and does not shift to around 30 kHz, as
    shown on page four.
    >
    > ...Jim Thompson
     
    The Phantom, Apr 21, 2005
    #5
  6. analog

    The Phantom Guest

    On Thu, 21 Apr 2005 09:00:12 -0700, "Joel Kolstad" <> wrote:

    >"analog" <> wrote in message news:...
    >> It looks like JT's favorite gyrator filter configuration is rather
    >> sensitive to gain bandwidth non-idealalities in the opamps.

    >
    >It's probably safe to say that in ANY active filter design, GBW is the most
    >critical parameter as to whether or not thet hing's going to work.
    >
    >> It also
    >> looks like excess phase (delay) plays very little roll (contrary to
    >> the claims of another poster).

    >
    >I'll take a look at Phantom's results later today and see what I can glean...
    >
    >> Of the various opamp/gyrator configurations/combinations are there
    >> any that are less sensitive to opamp GBW than Jim's "optimum"
    >> configuration?

    >
    >Sensitive in what way? In Jim's case, he's using a gyrator to simulate an
    >inductor, so about the only sensitivty you can talk about is that of the
    >inductor's inductance and loss (or Q, if you'd like) with respect to GBW.


    Sensitive in just those ways you've mentioned. Even without excess phase shift, the
    center frequency of the complete filter shifts from a nominal 20 kHz with GBW of 100 MHz
    down to 14.34 kHz with op amp GBW of 100 kHz because *both* the real and imaginary parts
    of the simulated inductance change with op amp GBW. So that to speak of the sensitivities
    of the simulated inductor is equivalent to speaking of the sensitivities of the complete
    filter, since it dominates the stability behavior of the filter (for any reasonable
    quality of R3 and C1).

    >However, in the more general case where you're trying to build an active
    >filter (using a gyrator, a bi-quad, etc.), you usually see people computing
    >sensitivites with respect to center frequency, bandwidth (or Q again), etc.
    >
    >> Maybe with some, opamp non-idealalities tend to
    >> cancel each other?

    >
    >The Moss-Ackerberg arrangement uses this approach, although (like many
    >bi-quads) there are multiple components that need to track each other to
    >change the center frequency.
    >
    >---Joel Kolstad
    >
     
    The Phantom, Apr 21, 2005
    #6
  7. analog

    Terry Given Guest

    Joel Kolstad wrote:
    > "analog" <> wrote in message news:...
    >
    >>It looks like JT's favorite gyrator filter configuration is rather
    >>sensitive to gain bandwidth non-idealalities in the opamps.

    >
    >
    > It's probably safe to say that in ANY active filter design, GBW is the most
    > critical parameter as to whether or not thet hing's going to work.


    I dont know what "thet" is, but I have yet to use Hing (Asofoetida) in
    any of my active filters. Its very nice in meals though.


    >
    >
    >>It also
    >>looks like excess phase (delay) plays very little roll (contrary to
    >>the claims of another poster).

    >
    >
    > I'll take a look at Phantom's results later today and see what I can glean...
    >
    >
    >>Of the various opamp/gyrator configurations/combinations are there
    >>any that are less sensitive to opamp GBW than Jim's "optimum"
    >>configuration?

    >
    >
    > Sensitive in what way? In Jim's case, he's using a gyrator to simulate an
    > inductor, so about the only sensitivty you can talk about is that of the
    > inductor's inductance and loss (or Q, if you'd like) with respect to GBW.
    > However, in the more general case where you're trying to build an active
    > filter (using a gyrator, a bi-quad, etc.), you usually see people computing
    > sensitivites with respect to center frequency, bandwidth (or Q again), etc.
    >
    >
    >>Maybe with some, opamp non-idealalities tend to
    >>cancel each other?

    >
    >
    > The Moss-Ackerberg arrangement uses this approach, although (like many
    > bi-quads) there are multiple components that need to track each other to
    > change the center frequency.
    >
    > ---Joel Kolstad
    >
    >


    Cheers
    Terry
     
    Terry Given, Apr 22, 2005
    #7
  8. I read in alt.binaries.schematics.electronic that Terry Given
    <> wrote (in <ttX9e.538$>)
    about 'Response curves of gyrator bandpass filter', on Fri, 22 Apr 2005:

    >I dont know what "thet" is, but I have yet to use Hing (Asofoetida) in
    >any of my active filters. Its very nice in meals though.


    In accordance with the Usenet Law of Typos in Spelling Corrections, I
    respectfully submit 'asafoetida'.
    --
    Regards, John Woodgate, OOO - Own Opinions Only.
    There are two sides to every question, except
    'What is a Moebius strip?'
    http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk
     
    John Woodgate, Apr 22, 2005
    #8
  9. analog

    Terry Given Guest

    John Woodgate wrote:
    > I read in alt.binaries.schematics.electronic that Terry Given
    > <> wrote (in <ttX9e.538$>)
    > about 'Response curves of gyrator bandpass filter', on Fri, 22 Apr 2005:
    >
    >> I dont know what "thet" is, but I have yet to use Hing (Asofoetida) in
    >> any of my active filters. Its very nice in meals though.

    >
    >
    > In accordance with the Usenet Law of Typos in Spelling Corrections, I
    > respectfully submit 'asafoetida'.


    Oops. Actually a reading comprehension issue (or perhaps its my dreadful
    writing. most likely a combination of the above).

    Cheers
    Trery
     
    Terry Given, Apr 22, 2005
    #9
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Rheilly Phoull

    Receiver bandpass filter questions.

    Rheilly Phoull, Aug 16, 2004, in forum: Electronic Basics
    Replies:
    6
    Views:
    1,453
    Rubicon
    Aug 18, 2004
  2. TRABEM

    Butterworth bandpass filter question

    TRABEM, Sep 11, 2005, in forum: Electronic Components
    Replies:
    6
    Views:
    991
    maxfoo
    Sep 13, 2005
  3. Georgy Onishchukov

    High-Q Bandpass Filter for 10 GHz

    Georgy Onishchukov, Nov 22, 2005, in forum: Electronic Components
    Replies:
    0
    Views:
    327
    Georgy Onishchukov
    Nov 22, 2005
  4. John

    bandpass filter

    John, Sep 28, 2003, in forum: Electronic Design
    Replies:
    5
    Views:
    385
    Jim Thompson
    Oct 1, 2003
  5. N Cook

    Response 400 answer machine - no response

    N Cook, Nov 1, 2005, in forum: Electronic Repair
    Replies:
    2
    Views:
    309
    N Cook
    Nov 2, 2005
Loading...

Share This Page