mixer circuit gain problem

[JAY X]

Hi!

First I wish you all the best for this 2025!!

Atached is a schematic of a summing mixer. Despite showing TL072, I built it with NE5532. That said let's get into the details:

The circuit is already made in a PCB. It mixes the outputs of 55 oscillators, through 120K value mix resistors, in one stage, no bucket summing. Total paralell value is 2181 ohms. In the feedback arm I have 20K and 100K potentiometer as a variable resistor. Ok so far...

When i measure the value of the 20K + 100k potentiomer IN CIRCUIT, it measures HALF the value!!! around 60K! I don't know why..
And when I disconnect the output of the first stage (the mixer) from the second inverting stage... the 20K + 100k potentiometer measures 120K!!

I suspect a driving capability problem of the NE5532... maybe i have to replace it for an LM4562, but I don't have any right now..

Any help is wellcome!

Jay x

 

[JohnRoberts]

How much gain do you get?

JR

PS: Not a great idea to put a pot in the feedback loop. Wiper bounce can result in spiky behavior.

 

[ruffrecords]

This is not really surprising. probably the dc voltage from your meter biasing up part of the NE5532

Also, the right side of the pot is connected to ground via the two 22K resistors and the output pot - probably 54K in total. It is possible there is a dc path from the inverting input to the non-inverting input which is in turn connected to ground via 22K. It is possible this little lot looks close to 120K i total which results in your 60K reading.

The best way to be cerain is to remove the 5532 and in future us a DIP socket for it.

Cheers

Ian

 

[JAY X]

How much gain do you get?

JR

PS: Not a great idea to put a pot in the feedback loop. Wiper bounce can result in spiky behavior.

Hi!

The gain I get goes From 0,18 to Unity. When the signals reach the input of the audio interface, they read about -30dbfs.

 

[JAY X]

To Ian: the NE5532 is socketed.
I Will check all your points!!

 

[ruffrecords]

Hi!

The gain I get goes From 0,18 to Unity. When the signals reach the input of the audio interface, they read about -30dbfs.

That is pretty much the gain range I would expect the posted circuit to give.

Unfortunately dBfs readings depend very much on the interface. You really need to read the actual signal voltage at the output on a scope or meter but overall it looks as though the circuit is functioning as expected.

Cheers

Ian

 

[JohnRoberts]

So what is the problem?

JR

 

[JAY X]

So what is the problem?

JR

 

[JAY X]

So what is the problem?

JR

Too low signal level reaching the audio interface...and me overthinking...
Well, tomorrow i Will measure levels again and get things clear..