Parametric Eq

This is one band of a parametric Eq (30Hz To 1k Range) design I've had for a few years now. I never got around to building it. I'm currently trying to tighten up the Boost/Cut parameter. All suggestions are welcomed.

Analag

At a quick glance, it looks like a reasonably standard state-variable based setup.

As shown, I think that the bandwidth control will affect the gain of the pass band/center frequency. U2 looks like it's drawn upside-down... surely as shown it will just lock itself against one or other of the rails. I assume that the inverting input should be the feedback and the summed input introduced at the non-inverting input, instead of the other way round.

For that design, I think that Graham Langley's variation on that sort of bi-quad arrangement is very neat. He controlled the bandwidth by varying the ground shunt leg and feeding everything back to one pole of an op-amp, thus controlling the bandwidth feedback and the gain by the same amounts... elegant and simple. Also, the signal path with no cut & boost usually only goes through the cut boost op-amp (IC2). In your version, it passes through IC1 also. Ordinarily when this topology is used, a version of theinput is also fed to IC1, but not made part of the straight-through chain... I think that Rafafredd might have posted the M2500 topology which was the second Amek console from Graham's pen which used this design.

An interesting approach. -I wonder if having a glance at Graham's M2500 schematic might give you some ideas?

Keith

I will check it out and try some revisions. Thanks for the input.

Analag

Amek M2500 EQ:

http://www.gyraf.dk/schematics/schematics.html

Jakob E.

Maybe I should abandon my schematic and try this one. What are the freq ranges. 50k rev log ganged pots, gotta check my sources for that one.

Analag

Omeg will make them for you..

Excellent link Jakob, will see how reasonable they are for a few one offs.


Analag

Hey Analag,
Have you consider using 2 x 2180 VCA replacing the dual gang freq control pot and just use 1 rev log pot to control both VCAs??? I think it would work out cheaper, also eliminate pots in the signal path.

Do you really think replacing pots with VCAs is an improvement?

Samuel

[quote author="Samuel Groner"]Do you really think replacing pots with VCAs is an improvement?

Samuel[/quote]

I do not know if there will be other problems that might branch from using VCAs but personaly do not like noisy pots and all pots will eventually become noisy......

Check out THAT corps design notes on VCFs, if you are interested. I am looking to try it out when I get time...... may be I can give you a more definite answer then.

Do the research I'm all eyes and ears, so far you are sounding good. It adds a level of complexity I think, but it should offer a tighter tolerance than pots, where one side might measure say 10.14k while the other side is 9.67k.


Analag

For decent tracking with relatively inexpensive pots, it can be an advantage to use voltage dividers to ground in stead of floating series resistances for state-variable frequency setting. This is e.g. what FocusRite does in the ISA eq's

Jakob E.

[quote author="analag"]Do the research I'm all eyes and ears, [/quote]

Yes, it would help a lot if I know someone that is expert in this field. However, I have not been that fortunate so I must warn ya it could be a while before I have made any progress in this area.
Still struggling trying to bias a single tube circuit and kill the noise, I was more like hoping you would do the hard work and I can harvest the fruit of your success???? :green:

so far you are sounding good. It adds a level of complexity I think,

Click to expand...

Well, you don't get something for nothing. The "complexity" might come useful some where else later, not to mention it might help to keep the budget down. I hope......

Anyway, all I did was pinch it off the synth geeks. May be JH can shed some light in this subject, he is one hell synth freak~

I mean that in a good way JH! :green: :thumb:

LOVE your work on the TD circuit, have to lay it out on protel soon.... another holiday project.

but it should offer a tighter tolerance than pots, where one side might measure say 10.14k while the other side is 9.67k.

Analag

Click to expand...

I think so, however my problem againest it is the friggin NOIZE. Can't stand it, just drive me insane.



I hate replacing parts, to me it just means it was not done right the first time imho.

Also Neve in the 'Formant' EQs used in the 55, 81, 82 and V-series consoles.

The VCAs are difficult beast so use linearly in this application. i've seen it done using transconductance amplifiers and the result was functional, but definately semi-pro lo-fi; The distortion was not pleasant sounding. If you were to use a VCA you can forget the reverse log thing entirely. A linear pot, 'bent' with some tugging resistors would do the job for you and remove the requirement for the rare pot law.

Basically, everyone's tried a bunch of ways to avoid having to use the reverse-log law, and the manufacturers still use it, because -guess why- it works the best. If in doubt, get some reverse log pots from SSL or whoever... -Failing that, just do what Rebis did and "train your fingers to work backwards"... using straight log pots!

Keith

[quote author="SSLtech"]
The VCAs are difficult beast so use linearly in this application.

Keith[/quote]

What type of VCAs are these? I am also looking to try out some TI multipliers to see how it would work..... Sounds like a great deal of work yet to be learnt, you game Analag???

:green:

I think we need some more synth freaks in the house. :grin: :thumb:

what about using the digital ic pots ; anyone try this yet

iv had some crazy ideas of mixing digitally controlled analog parametrics and a digitally controlled analog summing bus with a midibox cubase controller.