Tilt EQ

[gnd]

I'm thinking of putting tilting EQ in my console channels. Basically it should be able to tilt freq response to some 10 or 12 dB+- on extremes, and possibly avoid boosting too much low freq like 20Hz. Probably tilt should be centered arround 600Hz to 1kHz. And it should be controlled with one pot only - left for darker, right for brighter.

Something like this: http://www.tonelux.com/images/tilt%20eq.pdf or lie this http://www.diamondpedals.com/products/compressor.html

Any suggestions? Are any schematics like this arround?

 

[clintrubber]

Sorry for stating the obvious, but I guess the graphs in the pdf you showed can be realized very simple: just use the common bass & treble topology (Baxandall(sp?)) but use only one stereo pot and mount the bass-section reversed.
I thought I've seen more nifty circuits around but last time I looked I couldn't find anything back.

Dunno how it would matter, but the resulting response doesn't resemble a 'rotated ruler' too well. The linked-to blockdiagram looks more promising, but might be doing the same in pratice - don't know.
EDIT: ah, further down those graphs look alike - they'll be doing the same.

OTOH, the easiest form of a dark-bright is the passive tone-control found in pedals like the EH Big Muff, Morley PWF etc.

Bye,

Peter

 

[Kit]

You could try this one...
Adjust values to suit your needs.

http://www.groupdiy.com/index.php?topic=15365

 

[gnd]

What about connecting integrator and differentiator in parallel, and mixing between them panpot style? If freq's are selected outside of audio spectrum (say at 20Hz and 20kHz), integrator and differentiator give linear response over all freq. Tilt is large, 20dB/decade, but when mixed together, it would be reduced, is it?

 

[Kit]

What about connecting integrator and differentiator in parallel, and mixing between them panpot style?

Might work, but it seems like a difficult solution to a simple problem.

OTOH, the easiest form of a dark-bright is the passive tone-control found in pedals like the EH Big Muff, Morley PWF etc.

Dont know about the morley, but the big muff wont be flat when centered.
There would be a mid dip somewhere.
We could make it flat by tweaking values, but that would/could mess up the response.

 

[Guest]

Tilt EQ +/- 6 dB centered at 900 Hz. Half way down the page. http://www.headwize.com/projects/showproj.php?file=equal_prj.htm

I'm pretty sure the Tonelux circuit is similar to this.

Cheers,

elco

 

[chrissugar]

When I was a kid in mid eighties I built the tilt EQ part of that Quad34 preamp, and liked very much the results. But don't take my word, cause I was a kid and since then never checked that EQ again.
The schematic is simple enough to build one on test board and do some listening tests.

chrissugar

 

[bcarso]

[quote author="gnd"]What about connecting integrator and differentiator in parallel, and mixing between them panpot style? If freq's are selected outside of audio spectrum (say at 20Hz and 20kHz), integrator and differentiator give linear response over all freq. Tilt is large, 20dB/decade, but when mixed together, it would be reduced, is it?[/quote]

Consider phase.

The integrator lags 90, the differentiator leads 90. Summed equally you have a 180 difference and hence a notch filter. Either side of the notch you get your indefinitely rising responses.

However, a single pole RC lowpass and single pole RC highpass variably summed does work---flat when in the middle, a shelf ~hinging at 1/(2pi*tau) on either side. But you get at most 6dB/octave slopes.

But you want tilt, although probably not indefinitely :green: So for a given desired amount you want a bandlimited integrator or differentiator, which can be realized in various ways---a single-pole RC lowpass with tau at some lowest frequency, and the gain adjusted to get some nominal gain well down into the stopband, is one example for the integrator.

 

[PRR]

> 12 dB+- on extremes

That's not "tilt", it is Sierra Nevada. We got Bass/Treb and stronger EQs for that kind of abuse. Tilt is a few dB.

> integrator and differentiator in parallel, and mixing

Slopes to infinity, which we don't want. And....

> big muff wont be flat when centered.... We could make it flat by tweaking values, but that would/could mess up the response.

This general class of panning between low-pass and high-pass is doomed to a dip. The phases go opposite directions, and when summed to equality they cancel. We reduce that by minimizing the EQ effect: the dip goes away when we have zero EQ action.

The two passive network scheme can work if you have an out of phase drive or summer. Since that will generally require another amplifier, it is comparable to the Headwize plan.

That Headwize plan is more clever, and more fussy, than it looks. It gives good range, no null, dead-flat in center and the most useful +/-1 +/2dB tilts can be resolved easily. If you change any value, it messes-up. In fact I assumed the "47K" were rounded to standard value, but it turns out 50K is flawed and 47K is near-perfect. The only change worth exploring is to shift both caps the same amount: 6,800pFd each gives 733Hz center-point.

Even with inverted drive, the two passive networks plan does not make curves as pretty as the Headwize plan.

 

[bcarso]

[quote author="PRR"]> .

That Headwize plan is more clever, and more fussy, than it looks. It gives good range, no null, dead-flat in center and the most useful +/-1 +/2dB tilts can be resolved easily. If you change any value, it messes-up. In fact I assumed the "47K" were rounded to standard value, but it turns out 50K is flawed and 47K is near-perfect. The only change worth exploring is to shift both caps the same amount: 6,800pFd each gives 733Hz center-point.

Even with inverted drive, the two passive networks plan does not make curves as pretty as the Headwize plan.[/quote]

Good old Quad. They did many sensible things in their heyday.

Interesting how many good ideas came out of Brits with last names starting with W: Williamson, Watling, Walker,...Watney's Red Barrel....

This tilt question is another example of how seemingly simple requests can entail quite subtle solutions. The note about absolute polarity prompts the question: how hard is it to do the job while preserving absolute polarity? Methinks pretty tough.

As to how important it is...

 

[pstamler]

[quote author="bcarso"]Interesting how many good ideas came out of Brits with last names starting with W: Williamson, Watling, Walker,...Watney's Red Barrel....[/quote]

Well, three out of four ain't bad. :sam:

This tilt question is another example of how seemingly simple requests can entail quite subtle solutions. The note about absolute polarity prompts the question: how hard is it to do the job while preserving absolute polarity? Methinks pretty tough. As to how important it is...

Actually not hard at all if you use two inverting amplifiers. You could make one of them the EQ and the other just an inverter, or make one of them the bass and one the treble, still with ganged pots.

Or, if this is to be part of a channel strip and always in-circuit, just switch the leads of the input preamp.

Or, heck, connect the inverted output to pin 3 of the output XLR, put in an inverter to feed pin 2, and you've got a balanced output while you're at it.

As to how important it is...well, if you're mixing a microphone signal with other microphone signals, it could be very important.

Peace,
Paul

 

[clintrubber]

[quote author="Kit"]What about connecting integrator and differentiator in parallel, and mixing between them panpot style?[/quote]
That's exactly what circuits like the Big Muff do as far as tone controls go, but they might haven't realized that they could have used these expensive words to describe the 2*R+2*C+pot-circuit. :wink:
(Since in essence: LPF = integrator, HPF = differentiator)

OTOH, the easiest form of a dark-bright is the passive tone-control found in pedals like the EH Big Muff, Morley PWF etc.

Dont know about the morley, but the big muff wont be flat when centered.
There would be a mid dip somewhere.[/quote]
Very true, these are only simple passive circuits. But who did care then - these were just only FX-pedals :wink:

Bye,

Peter

 

[clintrubber]

The nice site of Jack O. has more info:

http://www.muzique.com/lab/tone3.htm

 

[zilver]

In this link is a schematic of the quad tilt eq (active version).
It's a very informative paper about equalizers in general.

Good luck, Hans

http://headwize.com/projects/showfile.php?file=equal_prj.htm

 

[gnd]

here is a good one!

http://www.elby-designs.com/asm-2/noise/tilt.pdf

It is used in noise generation circuit, but still.... i think it is copy of Quad.

By changing C values a bit, you can even get adjustable central point, making it a bit different on dark then on bright.

I like it. I may use this one. I goes almost +-10dB on extremes. Nice, soft curve... Sweet. :grin:

 

[bcarso]

[quote author="pstamler"]

This tilt question is another example of how seemingly simple requests can entail quite subtle solutions. The note about absolute polarity prompts the question: how hard is it to do the job while preserving absolute polarity? Methinks pretty tough. As to how important it is...

Actually not hard at all if you use two inverting amplifiers. You could make one of them the EQ and the other just an inverter, or make one of them the bass and one the treble, still with ganged pots.

Or, if this is to be part of a channel strip and always in-circuit, just switch the leads of the input preamp.

Or, heck, connect the inverted output to pin 3 of the output XLR, put in an inverter to feed pin 2, and you've got a balanced output while you're at it.

As to how important it is...well, if you're mixing a microphone signal with other microphone signals, it could be very important.

Peace,
Paul[/quote]

Just throwing an inversion in (easy) is a bit different I think. This is a situation where the polarity is a function of frequency and amount of tilt, I think.

 

[gyraf]

http://www.headwize.com/projects/showfile.php?file=equal_prj.htm

Jakob E.

 

[bcarso]

And from the notes just below that schematic:

The tilt EQ can be very beneficial for correcting tonal flaws in headphone sound, such as excessive brightness or darkness, without being too sonically obtrusive. The above schematic (figure 11) is from a preamplifier design by Reg Williamson and Alan Watling. It is a tilt control with a center frequency of 900Hz and a maximum boost/cut of 6dB. The circuit produces a shelving characteristic on either side on the center frequency (see the graph below). When the pot is in the center position, the EQ's response is flat (the bypass switch takes the EQ filter out of the audio path entirely). Turning the control to the left (counter-clockwise) lightens the sound; to the right (clockwise) darkens it. Note that at the extreme settings, the tilt EQ does result in a phase inversion.

 

[PRR]

> the polarity is a function of frequency and amount of tilt, I think.

For small tilts (which is what you want), the phase is zero (OK, 180) at both ends, and bumps up/down around 900Hz by an amount that is related to slope: approaching 90 degrees at steep slope but much less for gentle slopes.

And if the source or mix needs more than a hint of slope, we are fixing problems (or creating effects) that go beyond small phase shift issues.

I still think if you "need" anything like the +/-6dB of the Williamson/Watling knob, then you have a problem that needs more than "tilt". Tilt is useful for the 1dB/Octave slopes that sneak into the midrange on long lines, wrong-EQ tape, mellow mikes, sloppy disc-cutting, and other incidental losses. Most of these will need an additional narrow bump at the band-extreme, but that slant in the middle is hard to fix with narrow EQs.

 

[bcarso]

[quote author="PRR"]> the polarity is a function of frequency and amount of tilt, I think.

For small tilts (which is what you want), the phase is zero (OK, 180) at both ends, and bumps up/down around 900Hz by an amount that is related to slope: approaching 90 degrees at steep slope but much less for gentle slopes.

And if the source or mix needs more than a hint of slope, we are fixing problems (or creating effects) that go beyond small phase shift issues.

I still think if you "need" anything like the +/-6dB of the Williamson/Watling knob, then you have a problem that needs more than "tilt". Tilt is useful for the 1dB/Octave slopes that sneak into the midrange on long lines, wrong-EQ tape, mellow mikes, sloppy disc-cutting, and other incidental losses. Most of these will need an additional narrow bump at the band-extreme, but that slant in the middle is hard to fix with narrow EQs.[/quote]

Yeah, I took the website writer at his word but suspected that you couldn't get that much shift going away from center with that simple a topology. Considered as an end-to-end difference I suppose the statement of 180 is correct.

I don't know what the app would be, but the design challenge of a variable-slope tilt with phase correction is still intriguing. I know one can make a fixed-phase difference network with a bunch of stagger-tuned allpass sections, developed for use in SSB modulators. But the net delay through a given leg is substantial, which could create its own set of problems in a system.

Take one of those and use it with a network with the low-freq-limited integrator or band-limited differentiator, with the other material going through the other path of the shifter, maybe.