another active Summing Mixer designed for my needs

[matthias]

Hello,

since I'm not really satisfied about the way all current DAW's (I use Logic at the moment) are integrating external effects
I'm at the point where I decided to build my own summing mixer.

inspired by the outstanding mixer from Roland (radiance) and the great assistance of you guys here, I'd like to show you my first concept.

what I need is:

- 12 channels summed to a master bus
- switchable attenuator (0 / -12 / -24 /-36 ) in front of every channel ***
- master insert (with hard bypass switch)
- output for monitoring with stepped volume control and mono switch
- output for headphone
- output for external peak meter

*** the reason for the attenuator is that I like drive the output transformers of the previous unit (e.g. urei 1176, adr compex, ...) a bit into saturation.
then I need to attenuate the level after the output to not overdrive the following input stage.

before I begin to draw a complete schematic, I want to ask you some things:

- do you have any improvements in mind?
- do you think the 1200p is the best ic for this application or should I use the 1240 instead?
- what type of attenuator would you choose in front of the input?
- can I simplify the master insert switch by leaving the 1646 (master insert send) always connected?

mm-summing-box.v0.3.pdf

thanks in advance,
matthias

[JR.]

If all you are doing before the summing bus is simple pad attenuation, and ASSuming all of your input sources are balanced, I'd be tempted to drive directly into a balanced (differential) summing bus. Further I would configure the input pads to increase the value of the resistance feeding the bus to generate the cut. This way of padding the inputs, reduces the noise gain of the sum bus for best performance.

So I'd be tempted to just eliminate all those 1240 line receivers. If you expect to see unbalanced inputs perhaps add a few if needed.

JR

[radiance]

- do you think the 1200p is the best ic for this application or should I use the 1240 instead?

If you're going to put the attenuators in front of the line receivers I think you better go for the 1200.
The 1240's 90dB common mode rejection relies on precise matched resistors. I don't think you'll get the resistors from your attenuator as closely matched (0,005%) as the ones inside the 1240.
The 1200 OTOH does not care about impedance imbalances...

- can I simplify the master insert switch by leaving the 1646 (master insert send) always connected?

I can't think of anything that would cause a problem when doing this...

[matthias]

thanks for your reply.

I have some balanced and unbalanced units and I interchange them from time to time (rack space is limited here).
I decided to build an active design to factor out all the problems you can have with a passive summing circut. I also
want to use the bunch of that 1200 and 1646 ics that I bought ome time ago

for the attenuation I found a stepped u-pad circuit from JLM ...
it seems that this is exactly what I need.

do I have to alter the shown resistor values regarding the input impedance of the that1200p line receiver ?

[JR.]

Whatever... (the customer is always right).

If you are placing a pad in front of high CMR circuits, use precision resistors. Especially for the 470 ohm (those look like 5% values).

The impedance of that pad may be marginal for unbalanced (prosumer?) gear, but that is the tradeoff between loading and noise.. I might scale it up a factor of two for other than professional sources. Some consumer gear use opamps on outputs that can barely drive 2k loads. (I know because I've done it in the distant past).

JR

[mediatechnology]

What John said WRT the differential impedance values being on the low side with the values shown.

I hadn't really thought about it before but radiance makes a very good point about why a 1200-series (InGenius) should be used.

If you're going to use a U-pad with InGenius, like you've shown in the drawing, I don't think you're going to blow CMR significantly by external resistance mis-match. With a 1240-series having low CM impedance I think it would be a significant issue.

do I have to alter the shown resistor values regarding the input impedance of the that1200p line receiver ?

If you're concerned about the 48k (+/-20%) differential impedance of the 1200/1206 loading the pad it's not going to be a problem with the pad values you've shown.