Craig Anderton Super Tone Control

[a soBer Newt]

So as to practice my skills at copying schematics and doing board layouts I thought that it would be fun to work through Craig Anderton's Electronics Projects for Musicians and copy the schematic and do the layouts so here is the first in my series. 

Here is the schematic:


Please feel free to verify it.

The board layout:


Top 3D display:


And a link to the gerber files:
http://dl.dropbox.com/u/7870536/EPM_17.zip

If anyone wants the KiCAD files just let me know and I can upload that as well.

[Rylan]

Nice clean schematic, I did one tonight using an LM324 but it's not quite as easy to read haha... PS this is untested, I am building it this weekend. I THINK I got the schematic right but am not %100 till I build it and it works. In retrospect I should have just scanned CA's and changed the pin numbers but this was good practice learning Eagle. I've been an LTSpice and TinyCAD user till this week.

NOTE: the pots with lugs labeled 1 2 3 and 4 5 6 are dual ganged and should be 10k linear. CA recommended 10k log pots for lpf/bpf/hpf controls, and 1m linear for resonance.

[Rylan]

One thing I would like to point out is the two sets of grounds, digital and analog grounds (board and chassis).
I am thinking this design calls for the 2 grounds to connect via some brass standoffs or something? Does anyone know for sure? I am guessing it should be buss or star grounded together?

[Ronan]

Where did you get the idea of analog and digital grounds? This circuit is all analog. Check the original schematic, which shows all grounds as chassis ground...

[Rylan]

Actually no, the original EPFM schematic had indeed 2 sets of grounds which he describes on page 25 as "chassis ground" and "board ground" I referred to the IC chips ground as 'digital' ground as that seems to be a common term used to describe this situation. Also I was under the impression that the LM324 is a digital IC? My source for this is: http://www.geocities.ws/njbibin/robo11a.html

[Ronan]

Ahh, OK, I see. In the link, they are using the LM324 as a comparator, which is a digital application of this IC, but the LM324 is an op amp, which is an analogue device. Chassis ground is the earth connection to the metal case of the project, usually there is only one connection between board ground and chassis ground. Board ground I guess is the common ground on the PCB between all the components.

[a soBer Newt]

Hey just wanted to let people know the layout is verified and works I will have frequency sweeps posted soon.

[Mark Hammer]

I put one of these in a wah shell with a foot-controlled dual-ganged pot years ago.  Makes a helluva wah.

If anyone does attempt such a thing, I would recommend use of a linear dual-ganged pot, rather than log.  This makes it possible to use a toggle to reverse the direction of sweep (by flipping the connection to the outside leads of the pot around).  That may not seem particularly useful for bandpass or lowpass modes, but makes the highpass notcieably more usable.

Thanks for the layout.

[a soBer Newt]

Here is THD+N at and arbitrary point.  It will get much higher than this if you are boosting at the frequency your boosting at.



Here are just a few of the response cures that I was able to get

[sbcssbcs]

I've designed a dual-filter version of this circuit because I want independent cutoff and resonance for the lowpass and highpass (and the respective bandpass) bands.

I'm not an expert so here goes my schematic. Anything look odd? Can anything be improved? Any parts redundant?

[Rob Strand]

Anything look odd?

Joining  U2A.pin2 and U3A.pin2 looks odd.   If U3A.pin2 had it's own 100k wired  to the 220nF cap, or, 100k + 220nF cap wired back the input then at least it would look more "normal".   If you have simulated it (I haven't) and you are happy with the results then that's a different matter.

As a whole it forms a kind of parallel EQ,   is that what you want?

You aren't using the HP output from U2A, is that what you want?

[sbcssbcs]

Joining  U2A.pin2 and U3A.pin2 looks odd.

It does look odd now! I was thinking that since I have the filters in parallel, the input signal could be picked up after the 220n cap and 100k resistor as input for the second filter block. Will joining U2A.pin2 and U3A.pin2 lead to any side effects?

As a whole it forms a kind of parallel EQ,   is that what you want?

Yes, I believe a parallel EQ is sort of what I want. I want the Anderton Super Tone control but with separate controls for cutoff and resonance for low pass and high pass.

You aren't using the HP output from U2A, is that what you want?

Yes, I believe that is what I want, haha. I need one low pass and one high pass, which meant that I got rid of one in each circuit. This way the low pass comes from U2 and the high pass from U3, which are then mixed back together.

[Rob Strand]

Will joining U2A.pin2 and U3A.pin2 lead to any side effects?

It's not a normal connection so I don't know.   I'm fairly sure it won't be the same as separating the two circuits.  You would need to simulate the circuit to see what that actually does!    It might work close enough it or it might be totally whacked!

[ElectricDruid]

Will joining U2A.pin2 and U3A.pin2 lead to any side effects?

It would be better to have separate 100K input resistors for both mixers. That way there *won't* be any side-effects and it's much more likely to do what you want. You can get away with one input cap for both. You should consider that now you have two 100Ks in parallel they'll act like 50K, and any HPF effect of the 220n cap will have gone up by an octave, but by my calculation you're still down at 15Hz, so no problem.

Doing it the way you've done it, both op-amps have their virtual earth nodes connected together. Since the op-amp feedback acts to try and make sure that point is held at ground, they'll fight. It's not great.

[sbcssbcs]

Thank you @Rob and @Druid! I went ahead and added both cap and resistor to the second input, thinking about future mods. I also used the remaining opamp to cram in a little overdrive circuit, will see how useful that will be.

I simulated the circuit and it all seems to check out, does exactly what I need it to do.

One more thing I tested in simulation was using a variable resistor for the respective Q controls (originally a 100k with a switchable 12k in parallel) and I noticed that for values in between the original 100k and ~10k there seems to be a substantial gain loss until about halfway then gain recovers approaching the other original value. (in other words, gain both at 100k and ~10k is reasonable, but e. g. gain at ~50k is really low. Is this due to the opam feedback path? Is there no easy way to implement a reasonable variable Q pot with this circuit?