Dual EQ filter with a single pot?

[fryingpan]

I'd like to make some sort of Pultec-in-a-box (or actually an onboard bass preamp), I'd basically like to have a bass boost and a mid cut in a single pot, and I'd also like the preamp not to ever be flat; furthermore, it should boost more than cut (let's say, in a 3:2 ratio). Is there any way to do so? One idea is with a dual pot (and with one side wired in reverse) and some added resistance to one side of each filter so that the control is "skewed" to one side.

Furthermore, can you point to some design that allows Q to be set (once and for all) in the circuit? Many peaking or shelving filter circuits come with a preset Q.

[ElectricDruid]

Ok, sounds possible.

Do you want the Bass boost and the Mid cut to happen at the same time? so "more bass" means "less mids"?

And you said you want 3:2, so what?...the Bass goes up by 15dB and the mid only goes down by 10dB say?

Or do you want +/-15dB on the Bass control and +/-10dB on the Mid control? That's possible, with a bit of choosing of values. The obvious way would be a Baxandall-type control, which *would* give you a completely flat response in the middle, but as you say, you could skew it a bit with added resistances.

I think it's doable.

[Rob Strand]

The bass contour/enhance controls often only use a single-gang pot,





You can get mid cut but some boost the bass more than the treble, and some boost the treble more than the bass.

The basic idea is to mix two signal paths.  You change the shape of the response depending on what is mixed.

[fryingpan]

Ok, sounds possible.

Do you want the Bass boost and the Mid cut to happen at the same time? so "more bass" means "less mids"?

And you said you want 3:2, so what?...the Bass goes up by 15dB and the mid only goes down by 10dB say?

Or do you want +/-15dB on the Bass control and +/-10dB on the Mid control? That's possible, with a bit of choosing of values. The obvious way would be a Baxandall-type control, which *would* give you a completely flat response in the middle, but as you say, you could skew it a bit with added resistances.

I think it's doable.

Your first guess. The control should be boost only (for the bass, so cut only for the mids). Also gain range would be much lower (there is no point in having more than 6dB boost for the bass, therefore 4dB cut for the mids). That's quite excessive as it is.

The bass contour/enhance controls often only use a single-gang pot,





You can get mid cut but some boost the bass more than the treble, and some boost the treble more than the bass.

The basic idea is to mix two signal paths.  You change the shape of the response depending on what is mixed.

Hmm. What kind of filter is this? How does it work? Also, I'd like to be able to tinker with the values (and bandwidth of the mid cut, for instance).

[Rob Strand]

Hmm. What kind of filter is this? How does it work? Also, I'd like to be able to tinker with the values (and bandwidth of the mid cut, for instance

This circuit boosts the treble and makes a notch.  So to get bass boost and make a notch it will need to be tweaked or maybe re-jigged.

It's a bit tricky to understand.   Look at the second circuit.
- R25 just tweaks the response at 12 O'Clock
- R26 and R30 provide the base-line flat response, gain of -1 amp.
- C20, R31, R32, C22 for a band-*pass* filter.   This is a Wein-bridge circuit.
  It sets the notch frequency = 1/(2*pi*sqrt(C20 R31 R32 C22)
- The peak of the band-pass arm is less than one.
- C20 is the input to the band-pass/non-inverting arm.
- R26 and C18 and R27 are the input to the inverting arm.
- Ignoring C18, R27 + R28 appear in parallel with R26 and that makes the inverting gain more than 1
- For the band-pass/non-inverting arm R30, and R27 + R28 appear in parallel with R26 form the
   gain resistors for the non-inverting arm.
- With out C18 you have a non-inverting boosted band-pass filter which adds to the boosted "clean" signal
  for the inverting path.   With very judicial choice of parts/gains the band-pass filter partially cancels the clean
  inverted signal.   That changes the band-pass filter to a notch filter.
- C18 just forms a treble boost to the "clean" inverted path.
- When the Enhance is on minimum the treble boost and notch are removed and you just
  have an inverting amp.
-  As the Enhance is advanced the treble boost and band-pass filter gradually add to the inverting part.

All these ideas are used a lot in filters but it's pretty darn convoluted if you haven't dealt with this type of thing before.
You can do some calculations for frequencies but at the end of the day a spice simulation will get you there in no
time provided you understand how the band-pass can cancel the though signal.

[fryingpan]

C27? Do you mean C18?

[marcelomd]

I'm not on my computer now, so I can't draw anything.

One way to do it is to build a circuit that has a hard wired +6dB bass boost and -4dB mid cut, then blend it with a clean signal.

The EMG preamp that Gilmour uses on his strat works like that.

EDIT: Found the schematic. It's from the other forum, so I hope it's ok to post here. Anyway, this is just to illustrate the general idea of the hardwired boost/cut + blend.

[Rob Strand]

C27? Do you mean C18?

Yes, sorry about that.  (text has been fixed)

[fryingpan]

So, I came up with an idea for the bass boost, basically use a Wien bridge because it yields exactly the Q I'm looking for (a quite broad 0.32). The problem is that if I use it in a Baxandall-style configuration it doesn't work, probably due to its high insertion loss. How can I make use of it without placing it in a feedback loop (thus changing its Q) or strongly attenuating the clean signal (since mixing it with the clean signal would require around -15dB attenuation for the latter, and I want to minimise noise)? Other configurations require very large capacitors and I don't want to use electrolytics (I mean, with a guitar signal, I could get away with it, but electrolytics change over time and I don't want that).

[fryingpan]

And yes, using a low shelf makes the design much more convenient but I don't want a low shelf (I mean, I could sort of get a bell-style response from a low shelf but the curve is different in the low mids).

[marcelomd]

Have you considered a gyrator based configuration? I'm not too familiar with them, but I know you can set the Q.

[Rob Strand]

So, I came up with an idea for the bass boost, basically use a Wien bridge because it yields exactly the Q I'm looking for (a quite broad 0.32). The problem is that if I use it in a Baxandall-style configuration it doesn't work, probably due to its high insertion loss. How can I make use of it without placing it in a feedback loop (thus changing its Q) or strongly attenuating the clean signal (since mixing it with the clean signal would require around -15dB attenuation for the latter, and I want to minimise noise)? Other configurations require very large capacitors and I don't want to use electrolytics (I mean, with a guitar signal, I could get away with it, but electrolytics change over time and I don't want that).

You did well.

I played around with LTspice the other day and I moved the cap (C22) to ground on the opamp + input to the output of the opamp.  I might have removed the treble boost cap C18.    That got a bass-boost and instead of treble boost.    Due to lack of sleep, I was confident that was best solution and I haven't got back to it.

In the past I've done fixed bass boost + notch with a one of the bridge-T circuits and an added resistors.   That's a very simple circuit.   Variable versions are trickier.  I'm sure I've got something on my archived drive but it's not easy to get the stuff off.

For the sake of variations there's a GSR200 preamp on the freestomp forum.  That's an example of a circuit using a gyrator. I'm not 100% sure the circuit is correct (in fact I know it doesn't match some of the PCB's but I'm not sure if it matches one if the PCBs).  That circuit kind of boost the level a lot in order to get the max boost/max notch case.

PS cannot post a schematic for the enhance ckt mod because postimages is down.

[fryingpan]

Have you considered a gyrator based configuration? I'm not too familiar with them, but I know you can set the Q.

With a gyrator based filter you still basically need a 4u7 cap to achieve an 80Hz centred peaking filter. I suppose I could use a capacitance multiplier but that amounts to basically four opamps just for the bass filter.

[fryingpan]

So, I came up with an idea for the bass boost, basically use a Wien bridge because it yields exactly the Q I'm looking for (a quite broad 0.32). The problem is that if I use it in a Baxandall-style configuration it doesn't work, probably due to its high insertion loss. How can I make use of it without placing it in a feedback loop (thus changing its Q) or strongly attenuating the clean signal (since mixing it with the clean signal would require around -15dB attenuation for the latter, and I want to minimise noise)? Other configurations require very large capacitors and I don't want to use electrolytics (I mean, with a guitar signal, I could get away with it, but electrolytics change over time and I don't want that).

You did well.

I played around with LTspice the other day and I moved the cap (C22) to ground on the opamp + input to the output of the opamp.  I might have removed the treble boost cap C18.    That got a bass-boost and instead of treble boost.    Due to lack of sleep, I was confident that was best solution and I haven't got back to it.

In the past I've done fixed bass boost + notch with a one of the bridge-T circuits and an added resistors.   That's a very simple circuit.   Variable versions are trickier.  I'm sure I've got something on my archived drive but it's not easy to get the stuff off.

For the sake of variations there's a GSR200 preamp on the freestomp forum.  That's an example of a circuit using a gyrator. I'm not 100% sure the circuit is correct (in fact I know it doesn't match some of the PCB's but I'm not sure if it matches one if the PCBs).  That circuit kind of boost the level a lot in order to get the max boost/max notch case.

PS cannot post a schematic for the enhance ckt mod because postimages is down.

The bridged-T filter yields a narrow notch. By playing around with EQ and Reaper, I need a Q=0.6 band cut centred around  250-300Hz. Anyway the notch is not a problem, I can use a BJT based gyrator, which has the "advantage" of being "leaky" on the bass side, which might well be significant enough to provide some extra boost to the bass (I doubt so). It's the bass boost that's hardest actually.

[Rob Strand]

The bridged-T filter yields a narrow notch. By playing around with EQ and Reaper, I need a Q=0.6 band cut centred around  250-300Hz.

For the Bridge-T circuit where the "arms" of the T are the caps, the cap are normally equal valued.  That links the Q to the notch depth.    If you make the first cap larger than the second the Q goes up.  If you make the first cap small compared to the second the Q goes down.  The resistor in the leg of the T controls the notch depth.
(For the Bridge-T where the arms are the resistors you can get similar conclusions, bigger C in the previous ckt is like smaller R.)

Anyway the notch is not a problem, I can use a BJT based gyrator, which has the "advantage" of being "leaky" on the bass side, which might well be significant enough to provide some extra boost to the bass (I doubt so). It's the bass boost that's hardest actually.

I think the mods I made to the enhance ckt got about 9dB to 12dB bass boost.   It was very easy to get 8dB to 9dB but getting 12dB meant the notch parts needed to be adjusted carefully -  I didn't like it.    So maybe 8dB to 10dB was the limit.

In the Ibanez GSR200 circuit it's fairly easy to get more bass boost.   However, for circuits that (really) only use two capacitors the notch Q and the bass boost get coupled together.   The way to bail out of that is to separate the bass-boost and the notch completely.   That makes the circuit a bit more messy if you want to use a single ganged pot.

The Seymour Duncan/Cort preamps have a slap contour (circuit is on freestomp) .  It's a gyrator bass boost and a gyrator notch.  The hassle is the notch depth and bass boost are on separate trimpots.

[fryingpan]

The Ibanez onboard preamp is very noisy, I'm familiar with it. I'd like to do something better.

[ElectricDruid]

Here's my attempt at the problem. It's two gyrators, so you can easily tweak the freqs and Qs to taste. First one is in a feedback loop to give a bass boost, second one is not to give a mid cut. Changing the resistors above the gyrators changes the amount of boost or cut you get, so you can tweak that too. Final output is an active blend between the tweaked signal and the dry signal.





Treat LTspice's dB readings as relative. Unless you tell it not to, it takes the lowest reading as zero dB - in this case, the max cut.

HTH,
Tom

[iainpunk]

this is a cool bridge-tee based filter that boosts 84Hz and cuts 279Hz up to 15-ish dB



if someone else would simulate it to confirm the circuit sim in my head, that would be helpful

the pot should probably be log for the best 'feel'

cheers

[Rob Strand]

The Ibanez onboard preamp is very noisy, I'm familiar with it. I'd like to do something better.

High valued resistors in the gyrator circuit don't help with noise.
Ibanez gyator 2M2 on schem (1M on some PCB),  Seymour Duncan/Cort gyrators 100k.

[fryingpan]

So, I tried these circuits you kindly provided and, while they work, the Q is wrong (it's a bit too narrow, I'm looking for soft, broad boosts and cuts). I could reconsider the mid cut (make it narrower, even though, I must say, with a Q=0.6 it works very well, also because the bass boost-low mid cut and high mid boost-treble boost are voiced so as to actually be used and never left flat, and they compensate each other very well), but I'm quite happy with the bass boost. The high-mid boost and high shelf, piece of cake comparatively speaking, but one thing at a time.

I have this:



which does this (output taken from the lower right corner):



and I have this:



which does this (output taken from the emitter of Q1):



OK so it's a band pass and not a band reject, but that shouldn't be a problem. It's a bit lossy but I can always take the signal from the non-inverting opamp providing 20dB boost before the Wien bridge. How do I put these two together? Is there a better way to avoid these losses?