Tilt preamp

[samhay]

Here's something incorporating a couple of circuit snippets I have been meaning to build for a while:

It uses a bootstrapped BJT buffer (think 'Cornish buffer' or some of Gus' designs), but I have made the input impedance adjustable by playing with the bootstrapping. The Zin control might be useful to some, but is quite subtle in many cases and could be set with an internal trimmer, or the pot replaced with a 10k resistor and C2 taken from Q1 emitter.
There is a fair bit of hype surrounding the Cornish buffer. I can't say that it has changed my world, but it works nicely, has high enough impedance for our needs (perhaps excluding a piezo) and is fairly bulletproof.

The active 'tilt' control is similar to that used in some '70s Quad preamps and the Diamond compressors. The version I started with came from Rod Elliot's (ESP) page here (Fig 20):
http://sound.westhost.com/articles/eq.htm
I added 2 additional caps to roll off extreme bass and treble (C9 and C12, respectively) - compare top and bottom frequency response below. The treble roll-off is definitely recommended, as it gets quite hissy otherwise when turned all the way up (as one would expect with a 20 dB treble boost). For less roll-off, reduce C12 to 1n5 or 1n.
I am not sure if this control would be very useful for subtle Hi-Fi EQ'ing, but I like the tilt control as a guitar and bass effect. The crossing (0 dB) frequency makes more of a difference than I was anticipating, so is worth experimenting with and/or making switchable, as is the case in the bass version of the Diamond compressor.

I will try and get sound clips up this week and will be building one of these as a bass preamp with switchable bass cut (C5) and crossing frequency - probably 3 options each via on-off-on switches.


'view' for larger version

Simulated frequency response with Cx = 10, 15, 22, 33, 47, 68n.
Top: Tilt section with C9 and C12 removed.
Bottom: Circuit with C5 =  1u.

'view' for larger version

[Kipper4]

Very nice Sam.
Reminds me of the fa1 without the F.
I'll try this as I really like the fa1.
I like the tilt too.
What's the reason for the bias 22k 27k.? Guess its running at about 6v

Why the 5523 and not tl072?
Thanks for sharing

[garcho]

I like the tilt control as a guitar and bass effect.

me three

in some '70s Quad preamps

those Quad amps sound sweet!

[samhay]

Thanks Rich.
Had to look up the FA1, and yes the same general arrangement - 2 inverting op-amp stages with a transistor buffer up front for sensible input impedance.
Putting the level control before (rather than after) an active EQ makes sense as you can turn the gain down to prevent the EQ stage from clipping.
If you don't want a level control that can go below unity gain, and you don't mind an effect that inverts the signal polarity, then you could replace Q1 and IC2A with a non-inverting op-amp with variable gain.

>What's the reason for the bias 22k 27k.? Guess its running at about 6v
Gives Q1 a bit more headroom - collector should sit somewhere near half supply this way. IIRC, it might help with the op-amp headroom too, as some of the usual suspects can swing closer to the + rail than the - rail.

>Why the 5523 and not tl072?
TL072 is a fine choice too, but I like the NE5532 and the tilt control can work the op-amps pretty hard, so the extra grunt from the 5532 won't hurt.

[samhay]

>me three
Cool. Am not too sure how well know this control is, so good to hear that I am not alone.

>those Quad amps sound sweet!
Never heard one in the flesh, but they are certainly a very nice piece of kit.

[deadastronaut]

another cool samhay project...looking forward to hearing clips..

[electrosonic]

I have used that circuit in a flatline type compressor (vactrol based) and put the tilt control after the compressor part. Pretty happy with how it works.

I got the idea from a commercial compressor  (diamond ) - a quick google search shows they use it in a EQ pedal too

Andrew

[diydave]

Found this/your 'tilt'-circuit last week on the interwebs (somewhere) and breadboarded it for a Tone-knob on a lm386 amp.
Fiddeled with C10 and C11 to chang the 'tilt'-point from 1Khz to 640 Hz.
These changes were very very subtle for guitar. It didn't alter the frequencyresponse as much as I had hoped (or not as dramatically as I had anticipated). Opamp was a 4558.
Thought it would have a better use in a TS-circuit  .
But it is an interesting and simple one-knob tonecontrol without 'gain-loss' like you would have with a Big Muff toneknob.

[alfafalfa]

Seems like a good idea to use this tonecontrol Samhay.

those Quad amps sound sweet!

And they last a lifetime if you take care of them.
I still use a 303 , 405 and a Quad 33 preamp. In the poweramp I put new powersupply caps.
Had to make some adjustments fot the smaller size of those electrolytics nowadays.

Alf

[samhay]

Andrew - It was the various internet chatter about the Diamond compressor's EQ control that got me interested in this design again.
I thought I had forgotten about their EQ pedal, but did think about adding a mid range boost/cut, so it was probably lurking somewhere in my subconscious.

diydave - it have been playing with setting the crossover frequency / pivot (just looked at the Diamond Boost-EQ demo) frequency quite low - 200-300 Hz - for bass. I found that this makes some difference, but it may be in conjunction with the bass and treble roll-off filtering I have added.
One reason I would like to build this is to get to know it better - if switching the crossover frequency doesn't make much useful difference, then I will know for next time - and post my thoughts.

Also, the version of the tilt control I am using is slightly different to the version that pops up in most searches - it same generally topology and principle, but different implementation and component values - and I have only seen it on the ESP site. Which version did you try?

Alf - thanks.

[diydave]

Not quite sure (since I'm at work for now), but I think the filtercomponents were 2k2 and 22n. The other resistors and pot were the same. Where I got it from... no clue at all. It wasn't from ESP, and I can't seem to find the site.
I just copied the image to my harddrive.
To change the pivotfrequency, I switched between 22n and 33n filtercaps. Before breadboarding, I simulated it in LTspice (which I saved, so I can look this up).
Circuit was simpel: a non-inverting buffer/boost followed by the tilt going into the lm386.

[Kipper4]

I found this while searching for tilt eq

http://www.edn.com/Pdf/ViewPdf?contentItemId=4368935

im half way through my build straight to perf just got the tilt to do.

[samhay]

wow - you don't muck about Rich.

I have seen the linked article - it's a nice read.

Have you decided which crossing frequency to use for the tilt? If not, you could socket the caps and play...

[Kipper4]

The plan is to socket some caps it's an experiment build I want to know what it sounds like. 😃

[ashcat_lt]

If imagine you could mess with component values a but to get something like a Big Muff's mid scoop in the center position.  Could probably get a bit if mid boost there if you prefer.  I'm sure there's also a way to get more (or less) overal gain from it, but it makes me dizzy looking at the schematic so I couldn't tell you how. 

I had a stupid idea a while back for a nasty distortion thing based on the tilt control.  I never saw this in my "research", but I did find that a single pole filter like this wasn't quite steep enough to get what I was shooting for.

[Freppo]

Hey Sam. Cool project, as always

I did something similar a couple of years ago, but without the input buffer.
http://www.parasitstudio.se/uploads/2/4/4/9/2449159/active_tilt_eq_pcb.pdf

I found the most use of the tilt EQ with the center frequency as low as 125hz.
I have about 50 fabbed PCB's for mine laying around. Was going to sell them in the webshop but I never bothered..

[samhay]

ashcat - You can certainly unbalance the tilt control so that it is not flat at mid position, but I'm sure that it would be amenable to adding a mid scoop - will have a play.
Conceptually, the tilt circuit is not so hard to understand if you think about how it works at each extreme of the tilt pot and either side of the crossing frequency - i.e. when 2 of the 4 'legs' of the filter have much higher impedance than the other two.

Freppo - cool circuit, but isn't yours a mid boost/cut? It looks like the mid band from the old TL082 data sheet 3 band active tone control example.

[PRR]

> isn't yours a mid boost/cut?

Yes, but "mid" may be defined by capacitor choice to be 64Hz to 4KHz. For guitar, that is "anywhere". A 64Hz choice is clearly a deep bass bump; 4KHz will boost-cut the string-zing and ice-pick.

[Kipper4]

All done and it sounds good too. Thanks Sam
I dont have any C10k in stock so i used a B10 for Zin. might just be me but i cant see that it adds anything to the circuit when turning it. Maybe I'm not listening to what i need to listen to, hence why i cant see why its there. Unless its for active/ passive pups?
Next time i'm thinking just make a voltage divider with 2x 4k7s

I prefered the bigger caps sizes for C10,C11
I tried 82nf range to 10nf.
my favorites are the 68nf or 47nf
having said that all testing done at low volume (its the middle of the night right now) with a small noisy cricket amp. Not ideal but needs must.
And my test measuring stick (ears) are not what they used to be.

This is a really effective tone stack. I really like it. One knob, keeps it simple. less box drilling.
i will bread it with a TL072 and see how it fairs. for future use in other circuits.

Brilliant thanks Sam
I think Dead Astro will like this stack.

[PRR]

> i used a B10 for Zin. might just be me but i cant see that it adds anything to the circuit

Sam Hay's plan combines *two* features:

1) variable Z buffer
2) tilt EQ

Either could be used alone.

There may be some musical synergy, with Tilt to shift highs/lows balance and Zin to zing-up/down the topmost half-octave of naked guitar pickup, giving two actions and many more overall shapes.

It is possible that even right off a guitar, the lowest input Z (100K) has hardly any effect.

Just for jollies, tack 5K across R3. Now Zin goes from ~~500K to 5K. 5K should suck all the zing and half the mids/lows out of a naked guitar. (May do "nothing" if there's already a buffer or other active stage before; may just suck signal with a hi-Z source such as many pedals' output volume pot.)