Internal Guitar Pre-Amp with Tone Controls

[Channelman]

Here's my version of an Active Internal Guitar Buffer with Bass and Treble tone Controls.
With the Volume Control set to full and the Tone Controls to 'flat' there is no overall gain.
This was necessary since with the tone controls set to boost some 'headroom' is needed to
allow for the extra voltage swing.
Total current consumption is 170 uA and I've kept the guitars original potentiometers.
Most guitars have at Volume Control and two Tone Controls which can be used (Sorry
Telecaster fans), without spoiling the look of the guitar. I've never liked Gibson's
'Volume Control for each pickup' style anyway and have rewired my pickups directly to
the selector switch and use one volume control. (I'll think of a use for the other, sometime). ;D
I like the idea of Pickups without any load so I have a high input impedance buffer similar
to JohnH circuit. This also provides a low impedance drive for the Tone Controls.
After the Tone Controls and Volume Control there is a high input impedance amplifier
with about 14 dB gain to compensate for the 'flat' attenuation of the Tone Controls.
This provides a low impedance drive (about 1 kohm) which will drive any cable/amplifier.

EDIT: Just to clarify a point raised by ChrisK in a PM. Whilst the output impedance is 1 kohm,
the circuit willl NOT drive loads this low. There is insufficient current available to do this.
It will drive loads as low as 20 kohm up to the required output swing of 1 volt peak.
I didn't see the point in accomodating unlikely low loads at the cost of supply current.
The advantage of this low output impedance, which is achieved by negative feedback, is
it's ability to drive the cable capacitance without any loss of high frequencies.



Here are the overall response curves, all with Volume at maximum
Apart from the overall gain the Volume makes no difference to the following curves.

1) Bass at 25%, Treble at 25 % (FLAT response)

2) Bass and Treble both at Max

3) Bass and Treble both at Min

4) Bass FLAT, swept Treble

5) Bass swept, Treble Flat

Channelman

[ChrisK]

Nice circuit.

I shall anti-smite you for this. ;D

[JohnH]

'Tis good indeed. +1 from me. We need more active circuits! A couple of thoughts:

The centre frequency of the tones, being at 1kHz is similar to some hifi designs. Would it be more useful to shift it down a few hundred Hz for guitar use? Do you think that with full treble boost, the output stage will clip? and if so how would it be? maybe it won't clip? Another version for the output stage would be a JFET common source, with source bypass cap for gain, for a musical smooth clipping.

John

[Channelman]

John, I did centre the Treble control on 500 Hz but when I did
the same with the Bass control it didn't look to have enough
lift and cut (to me) since the lowest guitar frequency is 82 Hz so
I put it at 1 kHz.
Here's what it looked like,

compared with


The output stage is capable of delivering a 6 volt pk-pk swing
which (I thought) should be enough to allow for full treble
boost, full volume, humbuckers and 'hard hitting'...all at once.
I'm open to advice here. If you think it's not enough then I
would need to use a higher supply voltage or, as you suggest,
change the output design to let it 'bend nicely'
CM

[JohnH]

I reckon it might just be enough. The biggest swings in the signal would probably feature mostly the low bass (me guessing). 10db gain is about 3x on voltage? so the 6V p2p range, being +/- 3V might do it, Are you going to breadboard it? The idea of the circuit is handy and it would be good to try it out and to tweak it by ear.

John

[Channelman]

Yes, I do intend to try it out but I need to buy some J201 FETs first.
Luckily, I've got an SG I can try it out on, so I don't need to
dismantle the whole thing....useful little cover on the back.
I have a Strat as well and its a much bigger job to try this on.
When I was 'a lad' I had a Gibson 335 which was an absolute
pig to do anything with. I had to 'fish' controls etc thru the
'F' holes in the front. Dunno if they've changed them. It was
an early 60's Gibson (what would that be worth now?). Sold
it to get married to wife #1...wish I'd kept the guitar instead
Sorry, I've started 'rambling on' now.
CM

[ccoleman]

CM Never sell your guitar in exchange for a woman! It's like selling your soul to the devil! I hope you learned your lesson!

+1 for the very low-current-draw musically useful circuit. I trust it is correctly designed or easily debuggable.. and I second JohnH's suggestion of trying it out and unk's idea to try swapping in a JFET final stage for smooth tube-like clipping in case it ever does clip.

[UnklMickey]

ccoleman said:

... and I second JohnH's suggestion of trying it out and unk's idea to try swapping in a JFET final stage for smooth tube-like clipping in case it ever does clip.

hey John, i guess you're channeling ME now.







CM,

i like the use of the Bax. IMHO, more versatile than a F,M, or V tonestack.

have you considered modeling a pickup's output as loaded by this buffer, and adding a switchable network after the first stage, to remove the peak in the response curve?


unk

[ChrisK]

Sold
it to get married to wife #1...wish I'd kept the guitar instead

Jeff Beck was forced to choose between a woman (?wife) and a guitar. He still has the guitar......

i like the use of the Bax. IMHO, more versatile than a F,M, or V tone stack

True, but bear in mind that this is still not comprised of active tone controls (as in having the tone control networks "wrapped around" an active gain stage, feedback is). This topology will have greater generated noise than such.

It IS possible to (sort of) add a middle control to this topology as well. Sort of in that interaction is interaction.

QED re: National's 1980 Audio Application Handbook (Floobydust and all.) ;D

have you considered modeling a pickup's output as loaded by this buffer

I was going to question this as well in my PM to CM, but realised that the input stage's capacitance was in the several pF range.

My pSpice models confirm, and IMHO, it's not an issue.

[Channelman]

True, but bear in mind that this is still not comprised of active tone controls (as in having the tone control networks "wrapped around" an active gain stage, feedback is). This topology will have greater generated noise than such -ChrisK

Yea, I did think about this Chris and I agree it is a compromise.
But if I had used an 'Active Baxandall' type tone control which
requires a low source impedance it would need to be fed
directly from the input buffer. Then there would be nowhere to
put the volume control except AFTER the tone control.
Since we need a low impedance, to drive the cable, the only
options left would be yet another buffer or change the Vol pot
to 50 kohm.
Personally, I don't like changing the Pots so I went with a passive
tone control.

It IS possible to (sort of) add a middle control to this topology as well - ChrisK

Yes, good idea....and a use for that 4th pot on my SG

I trust it is correctly designed or easily debuggable - ccoleman

I've simulated it over the full Vgs(off) range of the J201 and the
full hfe spread of the BC548B. The BC179 gain is not critical.
Also over temperature 0 degC to 40 degC

i like the use of the Bax. IMHO, more versatile than a F,M, or V tonestack - unk

Sorry to be a pain, unk, (but I am a Newbie) but what's a 'V tonestack'?
I've sussed out the Bax(andall), F(ender) and M(arshall)
CM

[JohnH]

Unk means V for Vox. (I know 'cos I am inside his head..Mwaa..ha...haa..!)

If you dont have it, check out the 'Tone stack calculator' - from www.duncanamps.com/tsc - real time sims of M. F. V, Bax and a couple of others.

John

[Channelman]

Now do I feel REEE.....ALY stupid, or what?
I've actually got a Vox AC30 with Base & Treble Boost
Never occurred to me. V...Vox....doh!

Thanks John, I'll check out that link
CM

EDIT Wow, Just checked out that Tone Stack Calculator.
What a nice little program it is, AND you can 'fiddle' with the values.

[UnklMickey]

channelman said:

Now do I feel REEE.....ALY stupid, or what?...

not even close!

you asked a question, who's answer later seemed obvious.

just wait 'til you make a statement, that later proves to be patently false.

(been there, done that!)



a friend once told me, "a closed mouth gathers no foot."

true enough, but ya gotta risk it, if you want to get involved in the conversations.

[UnklMickey]

chrisk said:

...

have you considered modeling a pickup's output as loaded by this buffer

I was going to question this as well in my PM to CM, but realised that the input stage's capacitance was in the several pF range.

My pSpice models confirm, and IMHO, it's not an issue.

i was actually thinking a bid differently on this, Chris.

since the impedance would be very high, the loading on the pickup would be minimal.

it should result in an extended response curve, but with a very prominant peak.

being able to notch out that peak, might be useful?


unk

[Channelman]

since the impedance would be very high, the loading on the pickup would be minimal.
It should result in an extended response curve, but with a very prominant peak.
Being able to notch out that peak, might be useful?
unk

And I thought differently again,
I was thinking that it would give the 'true sound' of the pickup,
which will have a natural resonance. If it didn't sound right or
you just didn't like it then putting a 'normal' load of 500 k across
the input would 'tame' it.
Guess I just like it 'bright'
CM

[UnklMickey]

i hope you do like it bright!

remember this?

channelman said:

...This is the response of the pickup ALONE…no volume control…no cable. It shows is the pickup's
self resonant frequency. We can't use this because we need a cable. It's not a practical usable peak
unless active electronics are used with a very high input impedance. I did try this once in a guitar and
it sounded reeee……eeealy 'bright'.

...

if you used a cable, to connect the pickup to the amp with no volume or tone controls, it still wouldn't have as sharp of a peak, because of the cable capacitance.

but, i think having the buffer connected will get you pretty close to what the graph looks like.

might be okay for a neck pickup, but i'd be real concerned about 'ice-pick-in-the-forehead' syndrome, with a bridge pickup.


unk

[ChrisK]

i was actually thinking a bid differently on this, Chris.

since the impedance would be very high, the loading on the pickup would be minimal.

it should result in an extended response curve, but with a very prominant peak.

being able to notch out that peak, might be useful?

Yeppers! Although I don't "notch out" the peak, I add capacitance to tune it downwards.

I had added such response peak "tuning" here:
guitarnuts2.proboards45.com/index.cgi?board=schem&action=display&thread=1140140143

[Channelman]

I certainly do remember it. I used it in a Burns guitar I owned
in the 70's....I liked it but not to everybody's taste.

Perhaps that should be the use of the '4th control'...'the tamer',
what do you think?

I always did like it bright, even 30 years ago.
Here's a link to a circuit I published in a Everyday Electronics Magazine
in 1978 still hanging around on an 'effects schematic site'
It uses a 'Bridged T' circuit in the feedback loop to give
a high peak at 5kHz.
www.montagar.com/~patj/tonebstr.gif
There is a mistake on his redrawing. The Cap marked
22000pF should be 2200pF. When it's working it gives a
peak of 23dB at 5 kHz.
CM

[UnklMickey]

i think it's your baby so, you'll have to be the one to decide what you want it to sound like.

i think a "tamer" might be a good idea.

i also think that adding capacitance to shift the peak downward and reduce it's amplitude is okay, but might be a missed opportunity. that would probably also introduce a strong rolloff above the peak.

it might be worth considering a simple LCR network to reduce, but not necessarily totally eliminate the the peak.

if the Q could be kept reasonably high, you could tame down the peak without serious consequences to the rest of the curve.

that would make it still seem 'bright', but not 'in your face bright'.


i think i very much like where the peak is, but the amplitude does scare me a bit.


unk

[ChrisK]

that would probably also introduce a strong rolloff above the peak.

That existing roll-off already looks like a HALO insertion jump.

(High Altitude [jump] Low [altitude 'chute] Opening)

[Channelman]

unk, hows about......and it's only a thought at the moment,
a 'variable' capacitor to 'tune' it, adjusted by a Pot using
the Miller effect and a variable voltage gain stage.
That way we could move the peak down in frequency similar
to the way it's done in some wah-wah pedals.
You said that you like 'where the peak is' but I don't think
it would be that high with Humbuckers.
CM

[UnklMickey]

and probably not that 'peaky' either.

might be good enough the way it is, unless you split the HB.

[JohnH]

Just catching up. On peaks and input impedances etc, for my buffers, I personally prefer the volume control before the buffer, using a 500k as on the guitar, to calm down the peak a bit. Theres enough added zing from just negating the cable capacitance. Similarly, I set the input resistance of these circuits to about 1M, instead of lots of M.

Im also interested in the varaible cap idea. I had such a thing working on a breadboard a little while back, using a source follower JFET buffer circuit to sample the signal from the pickups, via its own 'volume' pot (which becomes the 'variable cap pot'). Then the tone capacitor, instead of going from pickup hot to ground, went from hot to the output of this buffer. With the pot set low, the buffer output is zero and the cap thinks it is acting like a normal tone cap, with its full effect, between pup hot and a zero voltage. With the pot set high, both ends of the tone cap are getting almost the same voltage. The cap reactance is much less and the pups think they are connected to a very small cap. Hence the resonant peak and dip can be swept from very high to low. There was a second buffer JFET for output. I haven't built this properly, the principle seemed to work but it was picking up noise in my tangled breadboard version. But its on the list to go into my Tonemonster2.

John