post fader amplifier questions

[Dimitree]

I was looking at the schematics for the Allen & Heath GS3000 console, to understand how Mute/Solo-in-place was achieved, and add that to my DIY console. This is the schematic of the post fader amplifier.



I don't understand the purpose of the components highlighted in yellow. Are those needed because of R132 and J111?
I'm wondering if I could just add R132 and J111 to my post fader amplifier (schematic below). Does R132 need to be the same value of R142, in the schematic above?

 

[zamproject]

I guess it's a small differential injection, that null the opamp output at very low level.
The fet don't cut 100% as J111 Rds on is about 30Ω

Quick math give 5mV residual for 1V input considering R132/Ron voltage divider, this is only 45dB attenuation.
R129/130 is about 26kΩ, considering the voltage divider with R131 you get aprox the same residual voltage.

An improvement is probably to put a trim in place of R129/130, to be able to balance the best possible null at output.

 

[abbey road d enfer]

I was looking at the schematics for the Allen & Heath GS3000 console, to understand how Mute/Solo-in-place was achieved, and add that to my DIY console. This is the schematic of the post fader amplifier.

View attachment 117353

I don't understand the purpose of the components highlighted in yellow. Are those needed because of R132 and J111?

The FET is not a perfect switch.
It has a residual value of 30 ohms, it cannot fully mute the signal, so the mysterious components feed a portion of the signal anti-phase (see how they feed the inverting input).
With 5.6k + the fader output resistance, the attenuation is about 45dB.
The other circuit provides a similar attenuation, so the reult is cancellation of the residual.

I'm wondering if I could just add R132 and J111 to my post fader amplifier (schematic below).

If you think 45dB off is good enough, why not?

Does R132 need to be the same value of R142, in the schematic above?

No. But beware you're going to alter the gain of the output stage by a fraction of dB.

 

[zamproject]

Seem we had simultaneous post Abbey

For the OP, have a look at some other mute circuit, like Studer 961, which involve 2 fet, series/shunt divider.
This involve little more logic as we need NO and NC switching.

 

[Dimitree]

thanks for the infos!
I saw the solution used by Studer 961, it's indeed more parts needed for the logic.
I'm wondering if this solution using half of DG403 could be ok. It may be overkill for what I'm after (I don't need single channel mute button, but solo-in-place.. i.e. I need to mute all other channels when I press Solo, and a traditional AFL system won't let me hear AUX properly).
But after all, it's not that much expensive than 2x FET and associated parts, considering that a single DG403 can be used for 2 channels.

 

[Analog_Fan]

You can use a (A)DG419 for mute or (A)DG442 for multiple channels.

 

[Analog_Fan]

thanks for the infos!
I saw the solution used by Studer 961, it's indeed more parts needed for the logic.
I'm wondering if this solution using half of DG403 could be ok. It may be overkill for what I'm after (I don't need single channel mute button, but solo-in-place.. i.e. I need to mute all other channels when I press Solo, and a traditional AFL system won't let me hear AUX properly).
But after all, it's not that much expensive than 2x FET and associated parts, considering that a single DG403 can be used for 2 channels.

Use a PIC18Fxxx, like PIC18F1320, PIC18F252 or so, CD4017 using diodes and the "push button" on the clock input pin.

 

[abbey road d enfer]

I've used the ADG419 and can testify to its excellent performance in a Mute cicuit.
However I can't figure out how to use a CD4017 (decade counter) in this position. Neither a PIC.

 

[Dimitree]

You can use a (A)DG419 for mute or (A)DG442 for multiple channels.

DG419 (about 2.5€) is more pricey than a DG403 (about 1.5€). Did you choose DG419 for its really low RDS(on) compared to thee DG403?
Also, the latter has two normally open and two normally closed switches, so I could do 2 channels in shunt + series configurations (as shown in my proposed schematic), so higher RDS(on) shouldn't be much of an issue I guess.
Am I missing something?

I've used the ADG419 and can testify to its excellent performance in a Mute cicuit.
However I can't figure out how to use a CD4017 (decade counter) in this position. Neither a PIC.

I think he means to use 4017 and/or a PIC to drive the DG419, not for the mute itself.

 

[zamproject]

Ok... a DG 403 or 413 is good option, 2NO and 2NC in a sigle package, if series/shunt mute is preferred

I need to mute all other channels when I press Solo

This is different then, you need a solo bus, and not just a mute switch on each channel.
Then have a look at the Studer 169 schemo, it offer a simple and clever mute/solo logic with a solo logic bus and single ON-OFF-(ON) switches per channel.
Solo is at the momentary ON side of the switch, but you can use a simple 2 position ON-OFF for solo function only.

 

[Analog_Fan]

I've used the ADG419 and can testify to its excellent performance in a Mute cicuit.
However I can't figure out how to use a CD4017 (decade counter) in this position. Neither a PIC.

I made this to control a CD4052.

 

[Analog_Fan]

DG419 (about 2.5€) is more pricey than a DG403 (about 1.5€). Did you choose DG419 for its really low RDS(on) compared to thee DG403?
Also, the latter has two normally open and two normally closed switches, so I could do 2 channels in shunt + series configurations (as shown in my proposed schematic), so higher RDS(on) shouldn't be much of an issue I guess.
Am I missing something?


I think he means to use 4017 and/or a PIC to drive the DG419, not for the mute itself.

(A)DG419 is DIP8, SOIC8, small package.

You can also choose CD4053 (something like 0.65€ / piece), but can only accept 20V, so you need to build a +10V/-10V power supply or +12V/-5V.

 

[Analog_Fan]

thanks for the infos!
I saw the solution used by Studer 961, it's indeed more parts needed for the logic.
I'm wondering if this solution using half of DG403 could be ok. It may be overkill for what I'm after (I don't need single channel mute button, but solo-in-place.. i.e. I need to mute all other channels when I press Solo, and a traditional AFL system won't let me hear AUX properly).
But after all, it's not that much expensive than 2x FET and associated parts, considering that a single DG403 can be used for 2 channels.

the Fader is already pulled low to GND, no need to bypass this using a analog switch.
R39 also provides a GND reference to the opamp, you could choose 1MOhm here.
You're probably best using a (A)DG442 or (A)DG441

 

[Dimitree]

Sorry, I don’t understand which part I could avoid, the switch that shorts to ground the fader or the switch that opens the connection to C1?

 

[Analog_Fan]

the switch that shorts to ground the fader?

Yeah, no need for that, redundant.
If the fader would be in it's top position, it's resistance would be reduced to 50 Ohm's (for DG series switches) if the switch mechanism is applied.

Vishay DG411LE, DG412LE, DG413LE is a 16 Ohm's analog switch, but requires +5V/-5V power supply.
You can build this kind of power supplies using LM4040 and a OPA2991 opamp (provides 75 mA) for the negative rail, a inverting opamp config sourcing from the LM4040.

 

[abbey road d enfer]

DG419 (about 2.5€) is more pricey than a DG403 (about 1.5€). Did you choose DG419 for its really low RDS(on) compared to thee DG403?

If I understand well, this is for a mixer. I would use a "modular" approach, by having one CMOS switch, one physical switch and one logic circuit for each channel.
But this is me, with too many years of habits.

Also, the latter has two normally open and two normally closed switches, so I could do 2 channels in shunt + series configurations (as shown in my proposed schematic), so higher RDS(on) shouldn't be much of an issue I guess.
Am I missing something?

I wouldn't do that. When the fader is fully up and the channel mutes, the full signal is applied to the shunt gate. Again it may be a matter of habits, but I wouldn't do that.
Offness with a Series CMOS switch is enough, as long as the layout is good and the impedances managed.

 

[abbey road d enfer]

View attachment 117378

I made this to control a CD4052.

If I understand well, that produces a 4-way switch from a momentary switch. Of course it could be set for a simpler toggle between Mute and Unmute.
How do you graft the Solo-In-Line on it.

 

[Analog_Fan]

If I understand well, that produces a 4-way switch from a momentary switch. Of course it could be set for a simpler toggle between Mute and Unmute.
How do you graft the Solo-In-Line on it.

the CD4052 has a 2 pin logic input and 4 options (00, 01, 10, 11), 1 to 4 switch x 2.
this requires considerable less space than a Alpha 12 pos rotary switch witch has a diameter of like 4.x cm, witch i got at home.

 

[Dimitree]

If I understand well, this is for a mixer. I would use a "modualar" approach, by having one CMOS switch, one physical switch and one logic circuit for each channel.
But this is me, with too many years of habits.

I wouldn't do that. When the fader is fully up and the channel mutes, the full signal is applied to the shunt gate. Again it may be a matter of habits, but I wouldn't do that.
Offness with a Series CMOS switch is enough, as long as the layout is good and the impedances managed.

Now that I know that I could use a single pole switch in series after the fader, I think I will use single DG468 for each channel. It’s super cheap (about 70 cents) and tiny (TSOP package). And it can be powered with +/-15V

 

[Analog_Fan]

Sorry, .....

This power supply was Designed by Don Buchla for the 259 complex wave generator.
I used it to power a part of the 259 circuit and another circuit that uses the CD4053 where it works great.
it powers 3 dual opamps the in the 259 circuit using a TL072 and it barely makes it, because the audio passing through it is like line level, it would not power 10 V.p.p. eurorack level, than just noise appears.

Recently some guy told me that the TL072 uses 3 mA @ 5V, just idle.

If you want more efficiency and precission change the zener for LM4040.
Need more current use OPA2991 opamp (75 mA) or another high power opamp.
A TL072 provides like 20 mA.

You need to measure how much a ADG442 consumes idle and loaded, how much of them ic's you wanna use.
You can replace the 1uF capacitors with 10 uF capacitors or bigger along with a 100nF at the outputs of these opamps in the above circuit.