Passive Mastering EQ

[pvision]

Surely most, if not all, of these options can be achieved? It's just a matter of resistor values and the switches used

Nick Froome

 

[ruffrecords]

Surely most, if not all, of these options can be achieved? It's just a matter of resistor values and the switches used

Nick Froome

Almost anything is possible but I don't want to end up with a design that is an ergonomic nightmare because it tries to be all things to all men. Bottom line seems to be that 23/4 ways is better than 12 no matter what boost/cut range you want The only hing stopping  us doing this with a readily available ELMA is my reluctance to do the PCB layout. I can see two approaches. One is to use regular tab ELMA and hard wire it to a 24 way connector. That makes the PCB layout easy for me but is probably a PITA to consruct, especially if you have 5  to do. The other is for me to get off my ... and have a go at the PCB. Anyone got a ready made footprint for the ELMA 04?

Cheers

Ian

Edit: A third alternative would be to revisit the resistor ladder on every switch version of the design.  ELMAs with a PCB fitted ready to take the resistors are already available. Might need to do a special version because I think we will need a a fourth connection from the 0dB position to make it work but this is a lot simple PCB layout problem. The only thing I am not sure of is whether this screws up the Q. I need to do some sims.

Ian

 

[salomonander]

Anyone got a ready made footprint for the ELMA 04?

i think its here if you can import a 3d-model files. should work with most software:
https://www.elma.com/en-as/download-center/3d/rs/selector-switches-3d/

 

[Unit-8]

Anyone got a ready made footprint for the ELMA 04?

Hi Ian,

Here you go. It is the Elma (04 and more) library for eagle pcb.

http://www.ml-limited.co.uk/Ian/Elma.zip

Andrea

 

[ruffrecords]

Thanks guys, you saved me a lot of work.

Cheers

Ian

 

[ruffrecords]

Looks like Elma have already solved the problem for me;

https://www.elma.com/en-as/products/rotary-switches/rotary-switching-products/product-pages/uni-selector-detail/

Cheers

Ian

 

[pvision]

Ian, what did you make of my 1/2 db "nudge" switch idea?

Nick Froome

 

[ruffrecords]

Ian, what did you make of my 1/2 db "nudge" switch idea?

Nick Froome

A big question is how accurate do people want this EQ to be? With switches and fixed resistors it is easy to get repeatability but getting the boosts/cuts to an exact figure in dB is not straightforward because, apart from depending on the  resistance of the inductors, it depends on the source impedance of whatever is driving itand that variable is not under the control of the designer.

The reason I said all that is because an exact half dB nudge switch is not likely to be possible simply because the amount  of nudge you will get in a simple nudge circuit will depend on the amount of boost/cut already dialed in. The more serious problem though is that there is only one resistor ladder which means the nudge would act on all the bands at once. You could add a nudge resistor in series with each bands EQ components. In the bell shaped bands this would have the effect of dropping the gain a little but also lowering the Q a little. In the shelving bands it would nudge the gain but also slightly alter the turnover frequencies

Cheers

ian

 

[dagoose]

To me the smaller steps (0.5 db) are more important than exact values. Sound and recall are more important for mastering. I don't care if I boost or cut 52.1hz or 51.2hz, as long it does it's job and sounds great.  8)
Ears, not eyes is what I always say to students at SAE.

 

[ruffrecords]

To me the smaller steps (0.5 db) are more important than exact values. Sound and recall are more important for mastering. I don't care if I boost or cut 52.1hz or 51.2hz, as long it does it's job and sounds great.  8)
Ears, not eyes is what I always say to students at SAE.

I fully agree with that sentiment. Repeatability is the main thing. You see many step switches with very expensive funny valued resistors in them just so the steps are an exact number of dB. In most cases this is a completely unnecessary expense. For example I designed a 12 position switched pot with 3 dB steps using only standard E24 resistor values. At most positions the error was less than 0.1 dB and only in a couple was it as high as 0.16dB.

Cheers

Ian

 

[Gold]

I don't think boost and cut values need to be exactly 0.5dB or whatever. I do think it's helpful to have boost and cut be symmetrical so a boost can be undone by a cut.

 

[salomonander]

I do think it's helpful to have boost and cut be symmetrical so a boost can be undone by a cut.

wont happen with an inductor circuit... cut will be sharper than boost. which works out fine most of the time.

 

[ruffrecords]

wont happen with an inductor circuit... cut will be sharper than boost. which works out fine most of the time.

In the original 1971 circuit that inspired this, the cut and boost were both 6dB. In that one situation, the cut and boost bell shapes are identical. In a 10dB version the cut will definitely be sharper than the boost. You can make the cut the same shape as the boost  in a 10dB version but you have to use a separate set of LC for each frequency to do this which is basically another wafer on each of the boost/cut and frequency select switches. I did this in the REDDEQ. Not sure if it was worth it.

Cheers

Ian

 

[Recording Engineer]

I'm good with 6dB max. boost/cut even for tracking... I figure beyond, I need to do something else!

 

[ruffrecords]

I'm good with 6dB max. boost/cut even for tracking... I figure beyond, I need to do something else!

The beauty of this circuit is that a single resistor ladder determines the steps so, if we use 11 steps each way, we can set them up any way we like. So for example we could have:

0.0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0  ( or 4.0, 5.0, 6.0 at the end)

or

0.0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 4.0, 6.0, 8.0, 10,0

or

0.0, 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 5.0, 6.0, 8.0, 10.0  (or 7.0, 8.0 at the end)

or

0.0, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0

It is just a matter of calculating the values of the resistor ladder. All the other EQ components remain the same.

Cheers

Ian

 

[ruffrecords]

I have been working out all the resistor ladder values for a  plus or minus 10dB range in 0.5dB increments. With these values we can create a ladder of any sort within that range. I will put the values in a table and post it here.

Still w aiting for a quote fro ELma for the whizzy PCB mounting switches.

Cheers

ian

 

[thomasdf]

Great project! Following

 

[ilfungo]

Another great project Ian!!!
Waiting!!!

 

[ruffrecords]

I am trying to write this up so others can modify as necessary for their own purposes e.g. tracking. In theory it would be possible to make a plus or minus 10dB equaliser in 0.5dB steps using 41 way switches!!

Still waiting for a quote on the Elma switches.

Cheers

Ian

 

[dagoose]

Still following this with big interest.  8)