Just FYI
That stuff about a Baxandall type circuit never being used for guitar amps is baloney.
Ampeg used a variation on the Baxandall tone controls that sounds really nice to my ears. Look at old Gemini and some Reverberocket models. It's a different 'sound' than the Fender/Marshall style, but the Baxandall type circuit can be perfectly good for guitar or bass amps.
Ampeg G12 schematic
Ampeg G15-R schematic
As was mentioned previously, just tweak the RC and pot values to get the filter slopes you enjoy best.
--
That stuff about a Baxandall type circuit never being used for guitar amps is baloney.
Ampeg used a variation on the Baxandall tone controls that sounds really nice to my ears. Look at old Gemini and some Reverberocket models. It's a different 'sound' than the Fender/Marshall style, but the Baxandall type circuit can be perfectly good for guitar or bass amps.
Ampeg G12 schematic
Ampeg G15-R schematic
As was mentioned previously, just tweak the RC and pot values to get the filter slopes you enjoy best.
--
Big differences between Baxandall/James and the Fender/Marshall/Vox (or whatever) passive stack arrangement are...
The James/Bax circuit is most effective when tweaked to provide linear response with controls in mid position. From that point on controls either boost or cut.
Oh, passive controls can "boost" too. It just means more "insertion losses" to achieve that 0 dB point. That is, passive controls tend to have greater insertion losses (signal attenuation) but in other regards their overall functionality may not differ at all from similar active controls.
The "F/M/V" tonestack is most effective when it provides a notable band-stop filter, usually around mid-range frequencies. This is due to circuit being largely derivative of a plain T-notch band-stop filter. Controls of this circuit will merely just cut. Linear response is traditionally acquired by cranking most of the controls.
Not only that, the controls are very interactive with each other: Adjustments to bass and treble controls affect the mid-range notch frequency, adjustment of the mid-range affects the mid-range notch frequency (along with other frequencies) and basically all controls interact with each other to some extent.
The Bax/James stack has much less of such control interaction and the effect of the dial on frequency response is much consistently centered around specific frequency ranges. Even if an additional "branch" implementing a mid-range control is added in.
So those two circuits work quite differently and also provide very different initial frequency responses.
Ok... So we can add a mid-range control to Bax/James circuit as well so that isn't a limit. We can't really eliminate or "add" control interactivity to either circuit so to great extent this characteristic is a user preference. Personally I hate the randomness of the FMV stack, some other people love it. The Bax/James stack works more "consistently" at the frequency ranges the controls are supposed to work at, and those controls also work in a logical way you expect them to work.
So this brings us to mid-range notch. Well, this is another key difference between these two tone controls, really, and in practice we are talking about nothing but a mid-range notch.
Do we want mid-range notch? Vintage Fenders certainly didn't need it, neither vintage Magnatone or Orange amps, or many other "legendary" amps to think of it. I don't know where this "Baxandall stack is unpopular in guitar amps / doesn't work in guitar amps" - myth comes from because it's absolutely hogwash in all regards.
It works just fine. Just differently.
Ampeg was mentioned. Many of their amps actually had a pretty effective resonant mid-range control (Boost/Cut) that mops floors with mid-range control of the traditional FMV stack. Couple this to their effective and logical bass/treble controls.
But anyway, if you need mid-range notch (yes, it certainly helps with those higher-gain tones) that's just a minor additional circuit implementation in practice. If you look at some guitar amps that happen to have Baxandall stacks (e.g. Gibson) they simply implement the mid-range notch as an individual notch-T filter. In certain Fender, Peavey and Crate amps (and many others I can't remember) the filter is actually even converted to a lousy mid-range control (a pot replaces one resistor), which works pretty similarly to mid-range control of the FMV stack in all its randomness...
So except for akward tone control interactivity the FMV stack offers nothing that couldn't be achieved by other means if the designer has such goals. Often these additional practices far surpass the FMV stack in overall functionality.
The James/Bax circuit is most effective when tweaked to provide linear response with controls in mid position. From that point on controls either boost or cut.
Oh, passive controls can "boost" too. It just means more "insertion losses" to achieve that 0 dB point. That is, passive controls tend to have greater insertion losses (signal attenuation) but in other regards their overall functionality may not differ at all from similar active controls.
The "F/M/V" tonestack is most effective when it provides a notable band-stop filter, usually around mid-range frequencies. This is due to circuit being largely derivative of a plain T-notch band-stop filter. Controls of this circuit will merely just cut. Linear response is traditionally acquired by cranking most of the controls.
Not only that, the controls are very interactive with each other: Adjustments to bass and treble controls affect the mid-range notch frequency, adjustment of the mid-range affects the mid-range notch frequency (along with other frequencies) and basically all controls interact with each other to some extent.
The Bax/James stack has much less of such control interaction and the effect of the dial on frequency response is much consistently centered around specific frequency ranges. Even if an additional "branch" implementing a mid-range control is added in.
So those two circuits work quite differently and also provide very different initial frequency responses.
Ok... So we can add a mid-range control to Bax/James circuit as well so that isn't a limit. We can't really eliminate or "add" control interactivity to either circuit so to great extent this characteristic is a user preference. Personally I hate the randomness of the FMV stack, some other people love it. The Bax/James stack works more "consistently" at the frequency ranges the controls are supposed to work at, and those controls also work in a logical way you expect them to work.
So this brings us to mid-range notch. Well, this is another key difference between these two tone controls, really, and in practice we are talking about nothing but a mid-range notch.
Do we want mid-range notch? Vintage Fenders certainly didn't need it, neither vintage Magnatone or Orange amps, or many other "legendary" amps to think of it. I don't know where this "Baxandall stack is unpopular in guitar amps / doesn't work in guitar amps" - myth comes from because it's absolutely hogwash in all regards.
It works just fine. Just differently.
Ampeg was mentioned. Many of their amps actually had a pretty effective resonant mid-range control (Boost/Cut) that mops floors with mid-range control of the traditional FMV stack. Couple this to their effective and logical bass/treble controls.
But anyway, if you need mid-range notch (yes, it certainly helps with those higher-gain tones) that's just a minor additional circuit implementation in practice. If you look at some guitar amps that happen to have Baxandall stacks (e.g. Gibson) they simply implement the mid-range notch as an individual notch-T filter. In certain Fender, Peavey and Crate amps (and many others I can't remember) the filter is actually even converted to a lousy mid-range control (a pot replaces one resistor), which works pretty similarly to mid-range control of the FMV stack in all its randomness...
So except for akward tone control interactivity the FMV stack offers nothing that couldn't be achieved by other means if the designer has such goals. Often these additional practices far surpass the FMV stack in overall functionality.
Last edited:
I haven't measured it, but the Duncan Tone Stack Calculator suggests that the only way to get flat response from a Fender style tone stack is to turn both Bass and Treble controls down almost to zero. Guitar amp tone controls are not required to provide flat frequency.response. They are only.required to 'sound good.' The trick is to define what that means to guitar players.
Also according to the Duncan TSC, turning the Fender tone controls to 10 yields a nice midrange 'suck out.'
Also according to the Duncan TSC, turning the Fender tone controls to 10 yields a nice midrange 'suck out.'
It is not the chassis. It should be circuit ground, actually the signal ground, and generally that is joined to a single point ground, where power supply ground,and the Power ground of the Power amp also is connected.
Ampeg - NOT BAXANDALL!!
I know this is almost trivia BUT Ampeg's tone controls are not Baxandall and never were. James Baxandall invented the negative feedback type equalization. The tone controls used by Ampeg are lossy and require a make up gain stage.
The Baxandall circuit was designed for hi-fi (before there was stereo) in 1952.
Ampeg claims their tone controls are Baxandall controls but they aren't. This is some slick marketing since probably 99% of musicians don't know the difference but hey, Baxandall!, must be special!! But it is just calling 87 octane Hi test. Just false advertising but less engine damage from the Ampeg deception.
The James - Bax ??? I don't know who thought this up but there never was one. Baxandall did not invent this.
The 2 types look similar on paper but have different eq curves. Both have an amplifier stage BUT:
The lossy circuit is made from parts values using logarithmic curves. - The pot has log taper, the upper caps are 10 X smaller and resistors 10X higher in the lossy control. There is a make-up gain stage following the controls. There is no negative feedback involved, except cathode resistor derived, independent of the tone controls.
The Bax uses linear pots, equal values for caps and resistors top and bottom. The controls are inside a negative feedback loop. In the center position the gain of the whole circuit is 1 - no gain no cut. In the boost positions there is less NFB at the shelving freq so the whole circuit has gain. In the cut position the NFB is greater and thus there is a cut.
In the Bax circuit the frequency point where the cut and boost kick in moves. The slope remains the same at all settings. The lossy (not Bax) has a fixed center point, usually 1khz and the slope changes.
One isn't better than the other, just different and it becomes a matter of personal taste.
Except for one thing. The amplifier stage used in each contributes noise. In the lossy (Ampeg) type the makeup gain stage noise stays relatively constant. In the Bax circuit the noise from the amplifier inside the Bax feedback loop varies with the settings, most noticeable in the high frequencies. In the cut position the noise of the feedback amp is reduced.
I know this is almost trivia BUT Ampeg's tone controls are not Baxandall and never were. James Baxandall invented the negative feedback type equalization. The tone controls used by Ampeg are lossy and require a make up gain stage.
The Baxandall circuit was designed for hi-fi (before there was stereo) in 1952.
Ampeg claims their tone controls are Baxandall controls but they aren't. This is some slick marketing since probably 99% of musicians don't know the difference but hey, Baxandall!, must be special!! But it is just calling 87 octane Hi test. Just false advertising but less engine damage from the Ampeg deception.
The James - Bax ??? I don't know who thought this up but there never was one. Baxandall did not invent this.
The 2 types look similar on paper but have different eq curves. Both have an amplifier stage BUT:
The lossy circuit is made from parts values using logarithmic curves. - The pot has log taper, the upper caps are 10 X smaller and resistors 10X higher in the lossy control. There is a make-up gain stage following the controls. There is no negative feedback involved, except cathode resistor derived, independent of the tone controls.
The Bax uses linear pots, equal values for caps and resistors top and bottom. The controls are inside a negative feedback loop. In the center position the gain of the whole circuit is 1 - no gain no cut. In the boost positions there is less NFB at the shelving freq so the whole circuit has gain. In the cut position the NFB is greater and thus there is a cut.
In the Bax circuit the frequency point where the cut and boost kick in moves. The slope remains the same at all settings. The lossy (not Bax) has a fixed center point, usually 1khz and the slope changes.
One isn't better than the other, just different and it becomes a matter of personal taste.
Except for one thing. The amplifier stage used in each contributes noise. In the lossy (Ampeg) type the makeup gain stage noise stays relatively constant. In the Bax circuit the noise from the amplifier inside the Bax feedback loop varies with the settings, most noticeable in the high frequencies. In the cut position the noise of the feedback amp is reduced.
Ouch
I innocently looked up guitar amplifier topologies, and was greeted by such a mix-up of terminology that to try correct that at this stage will be fruitless. As said by another, because the topology of the James and Baxandall looks vaguely similar, folks without electronic background care little regarding terminology.
The difference has been well explained in previous posts. I have no hope of seeing this confusion sorted out by a few posts on this forum. The one vague similarity is that the progression of frequency response with controls' positions is roughly similar; mirror-image boost/cut either side of mid-setting within the stage's capability, flat frequency/phase respone with controls in mid-position.
I innocently looked up guitar amplifier topologies, and was greeted by such a mix-up of terminology that to try correct that at this stage will be fruitless. As said by another, because the topology of the James and Baxandall looks vaguely similar, folks without electronic background care little regarding terminology.
The difference has been well explained in previous posts. I have no hope of seeing this confusion sorted out by a few posts on this forum. The one vague similarity is that the progression of frequency response with controls' positions is roughly similar; mirror-image boost/cut either side of mid-setting within the stage's capability, flat frequency/phase respone with controls in mid-position.
Mr Moscode,
It is Peter Baxandall, not James. But I do not blame you for getting mixed up - as said above.
Triva: Peter Baxandall published his circuit in Wireless World, October 1952. E.S. James published his in W.W., February 1949, both for use in hi-fi amplifiers. But the passive topology was apparently first used by one Michael Volkoff in 1939.
It is Peter Baxandall, not James. But I do not blame you for getting mixed up - as said above.
Triva: Peter Baxandall published his circuit in Wireless World, October 1952. E.S. James published his in W.W., February 1949, both for use in hi-fi amplifiers. But the passive topology was apparently first used by one Michael Volkoff in 1939.
In any case none of the Ampeg schematics of the 1960s series of amps, B15, etc. show a Baxandall tone circuit. So if the reissue amps claim to be true to their heritage they don't use it either. Their advertising is probably making a fib.
Yes. One of the late Fenders. Probably opamp.
Put a make-up amplifier after the loss. Basic audio planning. Same cost; sometimes less if tone-gain can be combined with other gain in one stage.
I notice this is still going strong!
Just to mix matters up further (!) and as it is festive season: There is a further variation, as used in the Quad tube control units!
Actually it is in essence the same as the Baxandall circuit, but with any amount of gain. Here the amplifier circuit can be any degree of say two tubes, with the original tone control configuration fed from the output anode (as per Baxandall) to the input tube's cathode. The controls are still linear, but a suitable resistor network gives feedback to the input stage cathode.
Perhaps attractive for those who do not want to 'waste' a full tube's gain in a gain = 1 topology.
Just to mix matters up further (!) and as it is festive season: There is a further variation, as used in the Quad tube control units!
Actually it is in essence the same as the Baxandall circuit, but with any amount of gain. Here the amplifier circuit can be any degree of say two tubes, with the original tone control configuration fed from the output anode (as per Baxandall) to the input tube's cathode. The controls are still linear, but a suitable resistor network gives feedback to the input stage cathode.
Perhaps attractive for those who do not want to 'waste' a full tube's gain in a gain = 1 topology.
This may not be the one I remembered but gives some idea.
U3a is a virtual-earth stage. The input side of the NFB also has shunt to ground paths so there is some passive loss; full analysis is left to someone who cares. "Insertion loss" is NOT "avoided": 410mV in makes 73mV out, which is not far from the classic loss-only plan; however changing 100k to 470k would make it about unity gain. (In this amp, they came-up to ~~1V to smack some diodes, but only needed 70mV for the power amp chip, so loss is fine.) I have no idea if this plan has the same taste and feel as the beloved 5F6a plan. The Frontman 15G is a <$100 starter-combo so I would not expect total refinement.
Attachments
Thanks PRR,
I was looking at the schematic on P224 of this book, and wondering if there was an actual implementation of it: https://www.jeanpierrepoulin.com/PDF/transistor.pdf
I was looking at the schematic on P224 of this book, and wondering if there was an actual implementation of it: https://www.jeanpierrepoulin.com/PDF/transistor.pdf
If you want to turn a passive stack into an active one you can try it yourself, see below. Note than an active stack doesn't eliminate insertion loss per se, really it adds gain after the stack to bring the signal back up again, the same as you can do with a passive stack. But the feedback approach does alter the I/O impedances and EQ curves, making them more symmetrical / resonant / wacky.
Attachments
There is! Fender 5F6 (not "a"). Leo connected the bottom of the tonestack to the NFB loop around the last stage, just as Kyttälä shows.
https://el34world.com/charts/Schematics/files/Fender/Fender_bassman_5a6_schematic.pdf
Difference: the Fender output stage is low gain and at the "R2 R1" junction there is about no spare gain to enforce the tonestack action.
Also I am not sure the 5F6 is a "real" amp. I've never seen it mentioned. I suspect it was a drawing mistake. It may be that all real amps were wired the 5F6a way, and the "a" drawing simply corrects the "no-a" drawing to reflect the as-built amps.
If you flip a James stack and wrap it inside an opamp, you get gain of 10 with +/- tone action BUT the pot rotation is "backward". Dynaco used reverse-audio pots. QUAD used audio pots but recessed them wired backward to hide the reverse rotation. I'm not sure the Kyttälä/5F6 plan works that way.
Any simple high gain system is prone to end up with too little or too much gain, because simple gain comes in fairly fixed values. So interstage loss is an important design tool. Taking loss in a tone control is one of the easiest.
Attachments
Apart from clearing up the meaningless 'James-Baxandall' confusion, which I would be glad to see the end of, one thing that hasn't come up here is that the passive circuit has 26dB insertion loss, where the Baxandall has zero. This means that practically every guitar amplifier on the planet could instantly be made 26dB quieter at the stroke of a soldering iron. And they are noisy beasts. This seems non-negligible to me.
> instantly be made 26dB quieter
No.
The 5F6a has potential 60dB gain ahead of the tonestack. The tone recovery stage has hiss-level hardly any higher than the input stage. A 60dB gain easily overwhelms a mere 26dB loss. (Also it is only -26dB at the dip, <-20dB in the highs where we hear hiss.)
The later SilverFaces might be more marginal, with only 33dB gain ahead of the stack, and the volume control hanging on the stack, and a mixer after that.
It is "clear" to an experienced amp diddler if tonestack loss is hurting the noise figure. Start VOL full down, listen. Now turn VOL to normal operating point. Does hiss rise? Indicating that input hiss dominates at working setting?
No.
The 5F6a has potential 60dB gain ahead of the tonestack. The tone recovery stage has hiss-level hardly any higher than the input stage. A 60dB gain easily overwhelms a mere 26dB loss. (Also it is only -26dB at the dip, <-20dB in the highs where we hear hiss.)
The later SilverFaces might be more marginal, with only 33dB gain ahead of the stack, and the volume control hanging on the stack, and a mixer after that.
It is "clear" to an experienced amp diddler if tonestack loss is hurting the noise figure. Start VOL full down, listen. Now turn VOL to normal operating point. Does hiss rise? Indicating that input hiss dominates at working setting?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.