baxandall tone circuit design ?
Walter Harley
I've found many schematics for the Baxandall circuit, but no description of
how to choose component values to obtain particular filter parameters.
Trial and error in PSpice is not helpful here, and I'm finding that the
circuit is a bit more complex than my mediocre attempts at algebraic
analysis can help me with (out of school for too long, I guess).
Can anyone help me understand how to select component values, and
potentiometer tapers, for this circuit?
Thanks,
--Walter Harley - http://www.cafewalter.com/~walterh
John Woodgate
inimitably wrote:
>Hi. I'm trying to design a Baxandall tone section for a bass guitar preamp.
>I've found many schematics for the Baxandall circuit, but no description of
>how to choose component values to obtain particular filter parameters.
>Trial and error in PSpice is not helpful here, and I'm finding that the
>circuit is a bit more complex than my mediocre attempts at algebraic
>analysis can help me with (out of school for too long, I guess).
>
>Can anyone help me understand how to select component values, and
>potentiometer tapers, for this circuit?
>
original publication. The thing is **difficult* to analyse, especially
if you wnat to see what happens at intermediate positions of the
controls. Also, be careful: there are at least four different
configurations, only two of which are genuine Baxandall. I would venture
a modest bet that the one you are using is not actually one of
Baxandall's.
--
Regards, John Woodgate, OOO - Own Opinions Only.
Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
Did you hear about the hungry genetic engineer who made a pig of himself?
Nick Dolling
> Hi. I'm trying to design a Baxandall tone section for a bass guitar
> preamp.
> I've found many schematics for the Baxandall circuit, but no
> description of how to choose component values to obtain particular
> filter parameters. Trial and error in PSpice is not helpful here, and
> I'm finding that the circuit is a bit more complex than my mediocre
> attempts at algebraic analysis can help me with (out of school for too
> long, I guess).
>
> Can anyone help me understand how to select component values, and
> potentiometer tapers, for this circuit?
You might check out
http://www.duncanamps.simplenet.com/technical/technical.html
Duncan Munro has a few articles on the Baxandall tone stack on his site.
He seems to have done some pspice modelling of the circuit, and also
invites questions about the articles he has there.
Hope this helps,
Nick
--
o--- Nick Dolling ------------------------------------------o
o--- Apexus Engineering Pty Ltd ----------------------------o
o--- ndol...@apexus.com.au --------------------------------o
o--- Ph : +618 8266 6222 Fax : +618 8266 6333 --------------o
Martin Griffith
scribbled:
>Hi. I'm trying to design a Baxandall tone section for a bass guitar preamp.
>I've found many schematics for the Baxandall circuit, but no description of
>how to choose component values to obtain particular filter parameters.
>Trial and error in PSpice is not helpful here, and I'm finding that the
>circuit is a bit more complex than my mediocre attempts at algebraic
>analysis can help me with (out of school for too long, I guess).
>
>Can anyone help me understand how to select component values, and
>potentiometer tapers, for this circuit?
>
>Thanks,
> --Walter Harley - http://www.cafewalter.com/~walterh
>
>
The Baxandall Tone control (1952) is well explained in the Natsemi
audio radio handbook, my copy dates from 1980, so i presume its no
longer available.
basically the turnover frequency is a simple 6dB/octave filter the 3dB
point is given by "one over 2 pi * R *C"
The maximum or minimum gain is given by the end resistors, so if you
want a lot of boost or cut these have to be small wrt to pot
Choosing R and C, find a value of a pot say 50K think of a 3dB
frequency and work out C.
You can't use low value pots, say 100R because this means your input Z
will be very low and will load the preceding stage, and distort badly.
if you use a bipolar opmp you really want to minimise the thermal
noise generated by the pot, so a 10M pot is out of the question, and
this would mean really low value caps
so its basically all compromise, typically the pots are 25K or 50K,
always linear,
email me if you want a scan of the natsemi stuff
Three things are certain
Death,Taxes and Lost Data
Guess which has occurred
Martin
John Woodgate
<wave...@SPAMbigfoot.com> inimitably wrote:
>The Baxandall Tone control (1952) is well explained in the Natsemi
>audio radio handbook, my copy dates from 1980, so i presume its no
>longer available.
If that's the same as in the MCMLXXVI (;-) edition, it isn't a true
Baxandall, and it isn't an 'Americanized' anything: it's a feedback
version of a configuration probably first described in English by E. J.
James in 'Wireless World', February 1949 and claimed to have been
invented by one Michael Volkhoff in July 1939, in Brussels. See 'WW',
April 1949. Used in passive form in the Mullard 5-10 and 5-20 preamps.
The National treatment glosses over the problem of interaction between
the controls and disguises that by not showing any frequency responses
where interaction occurs. It's much better than nothing, but it won't
necessarily guide one to a good design.
Go back to WW 1952 to find the real Baxandall, with centre-tapped treble
pot.
Martin Griffith
<j...@jmwa.demon.co.uk> scribbled:
>Go back to WW 1952 to find the real Baxandall, with centre-tapped treble
>pot.
>--
>Regards, John Woodgate, OOO - Own Opinions Only.
>Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
>Did you hear about the hungry genetic engineer who made a pig of himself?
Hi John
sorry for the mis-info
I didn't start collecting WW until 1965, and EW say I should try the
British Lib for a copy.......
unless you've got it!
Walter Harley
>The National treatment glosses over the problem of interaction between
>the controls and disguises that by not showing any frequency responses
>where interaction occurs. It's much better than nothing, but it won't
>necessarily guide one to a good design.
>
>pot.
Respectfully, I'm not sure I actually need a "real" Baxandall, especially if
it requires a centre-tapped pot, which I would probably have difficulty
obtaining. Mostly I just want to tweak the component values of the circuits
I've seen so that they're appropriate for different midrange frequency
(somewhere around 400Hz rather than the more typical 1kHz) and lower overall
impedances (so that, for instance, I can use more easily obtainable pots
than the 1M audio taper pots some designs specify, and so that I can keep
temperature noise low). But it would be nice overall to have a more
analytic sense of what's going on.
A question about tapers: it seems that the feedback designs (true or not)
use linear taper pots, where the passive designs use log taper. Since none
of the references I have discuss intermediate positions, I'm having a hard
time understanding the difference. Can anyone comment?
And, I don't think I can find the 1952 issues of Wireless World. Is there
anyone who could make me a copy, either physical or electronic? If they
can, and if it's legal, I would be glad to turn the results into a web page
accessible to all. Contact me offline if you can help.
Thanks very much for everyone's help!
--Walter
John Woodgate
<wave...@SPAMbigfoot.com> inimitably wrote:
>unless you've got it!
Yes, it is actually October 52, the first issue I bought myself. The
earlier ones (1946 onwards) were given to me. What is it you want? There
are four pages of text and diagrams.
John Woodgate
inimitably wrote:
>John Woodgate wrote in message <1iexG+Av...@jmwa.demon.co.uk>...
>>The National treatment glosses over the problem of interaction between
>>the controls and disguises that by not showing any frequency responses
>>where interaction occurs. It's much better than nothing, but it won't
>>necessarily guide one to a good design.
>>
>>Go back to WW 1952 to find the real Baxandall, with centre-tapped treble
>>pot.
>
>
>Respectfully, I'm not sure I actually need a "real" Baxandall, especially if
>it requires a centre-tapped pot, which I would probably have difficulty
>obtaining.
No, I was just mentioning where it can be found.
> Mostly I just want to tweak the component values of the circuits
>I've seen so that they're appropriate for different midrange frequency
>(somewhere around 400Hz rather than the more typical 1kHz) and lower overall
>impedances (so that, for instance, I can use more easily obtainable pots
>than the 1M audio taper pots some designs specify, and so that I can keep
>temperature noise low). But it would be nice overall to have a more
>analytic sense of what's going on.
To alter the 'hinge' frequency, scale the capacitor values, i.e. for 400
Hz you need caps 2.5 times as big as for 1 kHz. To alter the 'basic
impedance', scale all the components, e.g. to halve the impedance, halve
the resistor values and double the cap values. These changes work
provided the source and load impedances don't compromise things. As with
many circuits, a low source impedance and a high load impedance are
best, but the load impedance should not be nearly infinite.
(Classically, the passive versions were followed by a gain control pot
of the same value as the bass control IIRC.) This is especially
significant for the feedback type. It is not easy to get a low enough
source impedance (at the collector), a high enough load impedance (at
the base) and enough stage gain simultaneously from a single bipolar
transistor. That is why some of the early transistor implementations
were poor.
>
>A question about tapers: it seems that the feedback designs (true or not)
>use linear taper pots, where the passive designs use log taper. Since none
>of the references I have discuss intermediate positions, I'm having a hard
>time understanding the difference. Can anyone comment?
You are correct about the tapers. To fully understand why, you do have
to go to a fairly full analysis.
>
>And, I don't think I can find the 1952 issues of Wireless World. Is there
>anyone who could make me a copy, either physical or electronic?
There are four pages in the October 1952 issue. What do you want?
> If they
>can, and if it's legal, I would be glad to turn the results into a web page
>accessible to all. Contact me offline if you can help.
It is unwise from the legal point of view to post copyright material on
a web site, but for the exchange of copies between individuals for
purposes of private study, while still not without legal sanction (the
copyright holder is entitled to a fee), it is not very likely that
action would be taken over such vintage material.
John Woodgate
<j...@jmwa.demon.co.uk> inimitably wrote:
> This is especially
>significant for the feedback type. It is not easy to get a low enough
>source impedance (at the collector), a high
For 'high' read 'low': the base is a virtual earth point, of course, or
should be, but normally the achievable stage gain is too low to make it
a good virtual earth. The gain needs to be large compared with 10, since
that much closed-loop gain is required for 20 dB boost at the extremes
of the control settings and frequency ranges.
>enough load impedance (at
>the base) and enough stage gain simultaneously from a single bipolar
>transistor.
--
James Meyer
<j...@jmwa.demon.co.uk> wrote:
>It is unwise from the legal point of view to post copyright material on
>a web site, but for the exchange of copies between individuals for
>purposes of private study, while still not without legal sanction (the
>copyright holder is entitled to a fee), it is not very likely that
>action would be taken over such vintage material.
The RCA Radiotron book (the big red 4th edition) has a tone
control circuit on page 1483 in the appendix along with a reference to
the WW Baxandall article and a correction published a little later.
The diagram in the Radiotron doesn't actually say it's a
Baxandall, but it looks like all the other circuits that I've seen
that have claimed to be.
P.S. The Radiotron book is available in CD-ROM form.
Jim
John Woodgate
or the blue-covered Radio Designer's Handbook, Classic Edition, recently
reprinted.
> has a tone
>control circuit on page 1483 in the appendix along with a reference to
>the WW Baxandall article and a correction published a little later.
The correction clarifies that the pots are linear, and recommends 330 K
resistors from each end of the treble pot to earth if the centre-tap is
not used, to give the grid of the SP61 a path to earth.
>
> The diagram in the Radiotron doesn't actually say it's a
>Baxandall, but it looks like all the other circuits that I've seen
>that have claimed to be.
Yes. SP61 - those were the days.! Anyone got any Mazda Octal
valveholders? I've got some large-size top-cap connectors somewhere.
Allan Herriman
<j...@jmwa.demon.co.uk> wrote:
><7edjrg$n1j$1...@wrqnews.wrq.com>, Walter Harley <wal...@wrq.com>
>inimitably wrote:
>>Hi. I'm trying to design a Baxandall tone section for a bass guitar preamp.
>>I've found many schematics for the Baxandall circuit, but no description of
>>how to choose component values to obtain particular filter parameters.
>>Trial and error in PSpice is not helpful here, and I'm finding that the
>>circuit is a bit more complex than my mediocre attempts at algebraic
>>analysis can help me with (out of school for too long, I guess).
>>
>>Can anyone help me understand how to select component values, and
>>potentiometer tapers, for this circuit?
>>
>You have found the problem that has perplexed audio engineers since the
>original publication. The thing is **difficult* to analyse, especially
>if you wnat to see what happens at intermediate positions of the
>controls. Also, be careful: there are at least four different
>configurations, only two of which are genuine Baxandall. I would venture
>a modest bet that the one you are using is not actually one of
>Baxandall's.
John, you're not wrong about the difficulty. I thought I'd be a hero
and analyse the thing, but gave up. It's not that it's difficult,
just tedious. I may try with Mathcad later.
In any case, I did come up with some results that may be of interest.
1. Some of the tone controls are "passive". Some are "active" and
use an opamp (or triode, etc) with an identical RC network.
There is a simple transformation between the transfer functions, for
the case of a 3 terminal RC network. (This doesn't cover what I think
might be the genuine Baxandall (I haven't read the article), with the
grounded tap on the treble pot which makes it a 4 terminal network.
Dang.)
If the transfer function for the active case is G1(s) and the transfer
function for the passive case is G2(s), then
1
G1 = --------
1 - 1/G2
and
1
G2 = --------
1 - 1/G1
G1 and G2 are rational polynomials in s. We can express them as
N1 N2
G1 = ---- , G2 = ----
D1 D2
where N1, D1, N2, D2 are polynomials in s.
We then get N1 = N2, D2 = N1 - D1, D1 = N2 - D2
This means that for the same component values and pot settings, the
frequency response of the passive and active tone controls will be
different.
The zeros will be in the same locations, but the poles will be in
different locations, except when the controls are set to flat, in
which case they are in the same location. (Did that make sense?)
2. It is not possible to get complete isolation between the bass and
treble controls in a single stage with "reasonable" component values.
(By isolation, I mean lack of interaction.)
(You can get complete isolation by making some of the ratios between
resistors approach infinity, but the noise and/or distortion
performance would be shot.) If you must have complete isolation
between the controls, use an extra opamp, and have one tone control on
each opamp.
(You don't really need that much isolation.)
3. S plane representation of ideal tone controls.
There is a real pole and zero pair to control treble, and a real pole
and zero pair to control bass.
When the pot is in the middle, the pole and zero are at the same
location, and cancel, giving a flat response.
Tone controls flat ^ jw
Poles and zeros |
in same location |
x x |
-------o-----------o-+-----> sigma
|
treble bass|
|
|
Treble boost ^ jw
|
|
pole zero x |
-----x---o---------o-+-----> sigma
|
treble bass|
|
|
Treble cut ^ jw
|
|
zero pole x |
-----o---x---------o-+-----> sigma
|
treble bass|
|
|
Bass boost ^ jw
|
|
x |
-------o----------o-x+-----> sigma
|
treble bass|
|
|
Bass cut ^ jw
|
|
x |
-------o----------x-o+-----> sigma
|
treble bass|
|
|
Notice that the pole and zero for bass control move the opposite way
to the ones for treble. The DC gain (at s = 0) is also changed by the
bass control, and the pole and zero move this way to keep the mid band
gain constant.
4. I did manage to analyse the tone controls used in the LM4832.
(See the national semiconductor web site for a datasheet, and typical
plots of tone control performance.) This analysis is quite simple,
and only takes a few lines of algebra. (It's not a true baxandall,
but it does the right things in the S plane, except for a small
feedthrough from the treble control to the bass pole/zero locations.)
In C1 (1-T)R1 T.R1 C2
------+----||-----=====--+---=====-----||-----+
| | |
| | |
| (1-B)R2 | B.R2 |
+-----------=====--+-+-=====------------+
| |
| |
| |\ |
+---|- \ | Out
| \---------+----
| /
+---|+ /
| |/
gnd
T is the treble pot position. R1 is the treble pot.
B is the bass pot position. R2 is the bass pot.
T and B would normally vary from 0 to 1 (min to max), with 0.5 in the
12 o'clock postion, but here I've lumped in some other resistors (see
the data sheet for details), so
0.1357 < T < 0.8643
0.1520 < B < 0.8480
In both cases, the 12 o'clock position is still 0.5
Transfer function:
OUT(s)
G(s) = -------
IN(s)
treble zero bass zero
-B (1+sC2TR1) (1+sC1((1-T)R1+(1-B)R2)
G(s) = --------------------------------------------
(1-B) (1+sC1(1-T)R1) (1+sC2(TR1+BR2))
DC gain treble pole bass pole
Notice how C1 and C2 have to be matched to get a flat frequency
response when the controls are centered.
Notice also how the treble control feeds through to the bass control
(the TR1 term in the bass pole and zero).
From the datasheet of the LM4832, R1 = 14kohm, R2 = 200kohm, so the
feedthrough is rather small, because the R2 term swamps the R1 term.
Also, it is now obvious that a linear pot taper will give a linear
control over the time constants associated with each pole or zero,
rather than their frequencies. (This is not a bad thing...)
With the recommended value of C1, C2 = 8.2nF, we get the following
time constants:
Treble centre
zero at s = -1/57.4us, pole at s = -1/57.4us
Treble min
zero at s = -1/15.6us, pole at s = -1/99.2us
Treble max
zero at s = -1/99.2us, pole at s = -1/15.6us
Bass centre (treble in centre as well)
zero at s = -1/877us, pole at s = -1/877us
DC gain = -1
Bass min
zero at s = -1/307us, pole at s = -1/1.45ms
DC gain = -0.179
Bass max
zero at s = -1/1.45ms, pole at s = -1/307us
DC gain = -5.58
5. On just about every variation of this circuit I've seen (do a web
search, there's lots out there), the only DC path to the opamp input
is through the wiper on the bass pot. Now I know why the things get
so crackly when the pots get old. A better design wouldn't do this.
Phew, I'm glad that's over,
Allan.
John Woodgate
fujitsu.com.au> inimitably wrote:
>This means that for the same component values and pot settings, the
>frequency response of the passive and active tone controls will be
>different.
>The zeros will be in the same locations, but the poles will be in
>different locations, except when the controls are set to flat, in
>which case they are in the same location. (Did that make sense?)
>
The passive circuits use log (audio taper) pots and the C values are
around 10:1 ratio as well. Other components appear in some cases, but
it's even more difficult to handle these variants analytically.
>2. It is not possible to get complete isolation between the bass and
>treble controls in a single stage with "reasonable" component values.
>(By isolation, I mean lack of interaction.)
Yes.
>(You can get complete isolation by making some of the ratios between
>resistors approach infinity, but the noise and/or distortion
>performance would be shot.)
Yes, although the low-impedance passive version used by Mullard in the
5-20 (IIRC) preamp used 250 kohm pots for both treble and bass and,
IIRC, for the immediately subsequent volume control as well. There was,
even so, acceptably low interaction.
Walter Harley
>The expression for the transfer function has a lot of terms and takes
>about a page to list, so I won't bother to post it unless someone
>explicitly asks for it.
Any hope that some of the terms vanish for realistic values of the
components?
Allan Herriman
I managed to analyse a tone control variant (I won't call it
baxandall) with three capacitors.
The expression for the transfer function has a lot of terms and takes
about a page to list, so I won't bother to post it unless someone
explicitly asks for it.
I might get around to factorising it (to give the pole and zero
locations). If they turn out to be simple enough I'll post them.
Regards,
Allan.
Allan Herriman
<wal...@cafewalter.com> wrote:
>Allan Herriman wrote in message <37114e1d.1761000@newshost>...
>>The expression for the transfer function has a lot of terms and takes
>>about a page to list, so I won't bother to post it unless someone
>>explicitly asks for it.
>
>Any hope that some of the terms vanish for realistic values of the
>components?
No unfortunately. Even when I make the components with nominally the
same values have exactly the same values, it's still huge. (Ok, I
guess some of the terms *do* vanish, but not enough to make a
difference.)
I could go on making approximations, but that defeats the purpose of
an exact analysis - you won't get to see the interesting second order
effects. Also, there are so many terms, it's difficult to work out
which ones are significant, and which ones aren't.
Current status:
I have a closed form expression for the transfer function. I can plug
in typical values of components, and observe the frequency response
changing in the expected manner when I change the treble and bass
parameters.
I can reduce this expression to a bicubic in s. (It has three poles
and three zeros.) It should be a simple matter to solve for the pole
and zero locations, except that Mathcad says it's too big. Dang. I
can't even print it (crashes). I can copy it to the clipboard, but
only in a binary format that no other program can read.
I'm not sure how to proceed. Does anyone have any ideas?
(Except for the obvious one - give up)
Here is the circuit I analysed.
In (1-T)R1 T.R1
------+-----------=====--+---=====------------+
| | |
| _ |
| _ C1 |
| | |
| +------------+ |
| | |
| R3 | |
| +--=====-----+ |
| C3 | | |
| +--||-----+ C4 | |
| | +-------||-+ | |
| R4 | (1-B)R2 | B.R2 | | R5 |
+-=====--+--=====--+---=====--+-|-=====-+
| |
+----------+ |
| |\ |
+---|- \ | Out
| \---------+----
| /
+---|+ /
| |/
gnd
My ascii art isn't getting any better with time.
This was adapted from
http://www.elec.qmw.ac.uk/staffinfo/eric/courses/eda/c_amp.pdf
R1 100k pot, with 3k3 pad resistors either side.
(0.03 < T < 0.97)
R2 22k pot (0 <= B <= 1)
R4,R5 2k7
R3 33k
C1 1nF
C3,C4 47nF
Regards,
Allan.
John Woodgate
It's not the genuine Baxandall. It's a feedback James or Volkoff,
depending on whether you accept the earlier claim or not. The Bax. has
no caps in series with the treble control, just one cap in series with
the slider of the pot, which is supposed to have an earthed centre tap.
This gives a different treble control action: the +3 dB frequency varies
instead of the step height. On programme material which just has a roll-
off, it compensates for the roll-off without giving excessive mid-range
level.
Allan Herriman
(Allan awaits the answer patiently.)
John Woodgate
ujitsu.com.au> inimitably wrote:
>And the transfer function you calculated for it was...
>
>(Allan awaits the answer patiently.)
I've already posted that it's not easy, and I don't intend to attempt
it. Analysing the Bax. treble control by itself, however, is fairly
easy. But I'm still not going to do it.
Kevin White
The original article by Baxandall was in Wireless World magazine October
1952 issue, page 402-405.
kevin white
Walter Harley
John does is unwitting. ;-)
--Walter
Allan Herriman wrote in message <37141b40.8248348@newshost>...
>Thanks John,
> You unwittingly gave me a clue as to how to proceed with the
>analysis.
Allan Herriman
You unwittingly gave me a clue as to how to proceed with the
analysis.
analyse the effect of each control separately (with the other control
set to 'flat'), then I might have a chance.
I'll try again next weekend when I have some time. If I succeed, I'll
post the results.
Thanks again,
Allan.
robert_...@my-dejanews.com
allan.herrim...@fujitsu.com.au (Allan Herriman) wrote:
> Thanks John,
> You unwittingly gave me a clue as to how to proceed with the
> analysis.
> My problem is that the transfer function has too many terms. If I
> analyse the effect of each control separately (with the other control
> set to 'flat'), then I might have a chance.
>
> I'll try again next weekend when I have some time. If I succeed, I'll
> post the results.
>
> Thanks again,
> Allan.
>
> <j...@jmwa.demon.co.uk> wrote:
>
> ><3712bfd3.4833005@newshost>, Allan Herriman <allan.herriman.hates.spam@f
> >ujitsu.com.au> inimitably wrote:
> >>And the transfer function you calculated for it was...
> >>
> >>(Allan awaits the answer patiently.)
> >
> >I've already posted that it's not easy, and I don't intend to attempt
> >it. Analysing the Bax. treble control by itself, however, is fairly
> >easy. But I'm still not going to do it.
>
>
I'm new here. I will be happy to solve the tone control circuit for you, but
I'm not sure exactly what you want. Do you want the frequency response of the
circuit you posted? or do you want the component values to be optimized to
meet a desired curve?
If you want to take the time to learn how to run a circuit analysis program
you can download mine from http://www.nova.com Don't feel obliged to send in
the shareware charge, you may have the program free of charge.
Robert Stanton
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
robert_...@my-dejanews.com
allan.herrim...@fujitsu.com.au (Allan Herriman) wrote:
> Thanks John,
> You unwittingly gave me a clue as to how to proceed with the
> analysis.
> My problem is that the transfer function has too many terms. If I
> analyse the effect of each control separately (with the other control
> set to 'flat'), then I might have a chance.
>
> I'll try again next weekend when I have some time. If I succeed, I'll
> post the results.
>
> Thanks again,
> Allan.
>
> <j...@jmwa.demon.co.uk> wrote:
>
> ><3712bfd3.4833005@newshost>, Allan Herriman <allan.herriman.hates.spam@f
> >ujitsu.com.au> inimitably wrote:
> >>And the transfer function you calculated for it was...
> >>
> >>(Allan awaits the answer patiently.)
> >
> >I've already posted that it's not easy, and I don't intend to attempt
> >it. Analysing the Bax. treble control by itself, however, is fairly
> >easy. But I'm still not going to do it.
>
>
I'm new here. I will be happy to solve the tone control circuit for you, but
I'm not sure exactly what you want. Do you want the frequency response of the
circuit you posted? or do you want the component values to be optimized to
meet a desired curve?
If you want to take the time to learn how to run a circuit analysis program
you can download mine from http://www.nova.com Don't feel obliged to send in
the shareware charge, you may have the program free of charge.
Robert Stanton
PS My e-mail address is: http://www.rs...@wpe.com
James Meyer
wrote:
>If you want to take the time to learn how to run a circuit analysis program
>you can download mine from http://www.nova.com Don't feel obliged to send in
>the shareware charge, you may have the program free of charge.
>
>Robert Stanton
I figured that everybody was invited since the massage was
posted to the group. But when I went to the site I couldn't see any
software or any links to any software.
Jim
Walter Harley
dejanews.com says...
> [...] Do you want the frequency response of the
> circuit you posted? or do you want the component values to be optimized to
> meet a desired curve?
>
Thanks, and thanks for your generous offer of the circuit analysis
program (although I too can't seem to find it at the URL you gave. I'm
using IE 4.01, and you mentioned some problems with IE.)
My interest (I started the thread) is twofold: first, I want to be able
to pick component values given a specified set of parameters; second, I
want to be able to understand the circuit analytically, that is, have a
clear sense of what's going on and why and how the parameters depend on
the component values. Help with either of those goals would be great.
--Walter Harley
c.c.t...@baxandallx.killspam.us.com
: In article <7gpgek$ec9$1...@nnrp1.dejanews.com>, robert_stanton@my-
: dejanews.com says...
: > [...] Do you want the frequency response of the
: > circuit you posted? or do you want the component values to be optimized to
: > meet a desired curve?
: >
: > If you want to take the time to learn how to run a circuit analysis program
: > you can download mine from http://www.nova.com [...]
: Thanks, and thanks for your generous offer of the circuit analysis
: program (although I too can't seem to find it at the URL you gave. I'm
: using IE 4.01, and you mentioned some problems with IE.)
I tried it with Netscape too, looks just the same as with IE. Oh well,
"no va" does mean "no go" in Spanish...
Stephen
: My interest (I started the thread) is twofold: first, I want to be able
Allan Herriman
I actually managed to analyse the circuit. I didn't bother to
post the results as I didn't see the point.
The reasons:
* The expressions (either for the transfer functions or for the pole
and zero locations) are so huge that it isn't really practical to use
them. The transfer function is only bicubic, but it's still pretty
big.
* It is unlikely that I could post them without making a typo. (I was
having trouble with cut and paste from mathcad.)
* It isn't really possible to obtain an understanding of the operation
of the tone control from these expressions either.
An intuitive understanding of the operation of the tone controls can
be had by simply looking at them for a few minutes.
(Ok, it's easy for me to say that after I spent a long time doing a
formal analysis. But they really aren't that difficult.)
My interest waned somewhat after I discovered what an "ideal" tone
control would do in the S plane. I didn't actually want to build one.
Regards,
Allan.
On Wed, 05 May 1999 13:18:15 GMT, robert_...@my-dejanews.com
wrote:
>In article <37141b40.8248348@newshost>,
> allan.herrim...@fujitsu.com.au (Allan Herriman) wrote:
>> Thanks John,
>> You unwittingly gave me a clue as to how to proceed with the
>> analysis.
>> My problem is that the transfer function has too many terms. If I
>> analyse the effect of each control separately (with the other control
>> set to 'flat'), then I might have a chance.
>>
>> I'll try again next weekend when I have some time. If I succeed, I'll
>> post the results.
>>
>> Thanks again,
>> Allan.
>>
>> On Tue, 13 Apr 1999 08:03:52 +0100, John Woodgate
>> <j...@jmwa.demon.co.uk> wrote:
>>
>> ><3712bfd3.4833005@newshost>, Allan Herriman <allan.herriman.hates.spam@f
>> >ujitsu.com.au> inimitably wrote:
>> >>And the transfer function you calculated for it was...
>> >>
>> >>(Allan awaits the answer patiently.)
>> >
>> >I've already posted that it's not easy, and I don't intend to attempt
>> >it. Analysing the Bax. treble control by itself, however, is fairly
>> >easy. But I'm still not going to do it.
>>
>>
>
>I'm new here. I will be happy to solve the tone control circuit for you, but
>I'm not sure exactly what you want. Do you want the frequency response of the
>circuit you posted? or do you want the component values to be optimized to
>meet a desired curve?
>
>If you want to take the time to learn how to run a circuit analysis program
>you can download mine from http://www.nova.com Don't feel obliged to send in
>the shareware charge, you may have the program free of charge.
>
>Robert Stanton
>
>