# What's the rpoblem of this MIC circuits?

#### tony_lth

Hi, Gurus,
I found a MIC circuit, which convert the single ended MIC signal to diff-MIC signals.
I simulated it in ADS, and found the results are weird.
The following pictures are the TD response & FD response,
who can comment the problem of this circuits?
Thanks.
Tony Liu

#### Attachments

• 41.4 KB Views: 11
• 26.5 KB Views: 14

#### FvM

##### Super Moderator
Staff member
Forgot to show the circuit?

#### tony_lth

PS:
I wonder if it has some self-oscillation from the FD response?

- - - Updated - - -

Hi,
The following is the circuit. I guess it should be enough to estimate from the response.

#### Attachments

• 71.7 KB Views: 24

#### volker@muehlhaus

I wonder if it has some self-oscillation from the FD response?
If you FFT your square signal, you will see the related spectrum, and that frequency response seems to amplify some harmonics a lot.

tony_lth

### tony_lth

points: 2

#### FvM

##### Super Moderator
Staff member
How did you arrive at this circuit? It doesn't seem right.

Can you give a gain specification?

tony_lth

### tony_lth

points: 2

#### KlausST

##### Super Moderator
Staff member
Hi,

I think the circuit could work.

But why do you use a pulse input ... of that high magnitude.?
--> Use a sine shape waveform of 1kHz and 30mV amplitude.

The circuit as is amplifies high frequency signals higher than low frequency signals.
I expect this is not what you want unless it´s to compensate for the used microphone behaviour.

--> I´d add a series input resistor. Maybe 10k, then this reduces max gain to about 5. You can play around with the value: The lower the resistor the higher the gain.

Klaus

tony_lth

### tony_lth

points: 2

#### FvM

##### Super Moderator
Staff member
I think the circuit could work.
Only with the necessary corrections, e.g. the suggested gain setting resistor.

tony_lth

### tony_lth

points: 2

#### KlausST

##### Super Moderator
Staff member
Hi,

A2 has a defined gain of -1, thus it should work.

A1 has an input impedance of 1600 Ohms @ 1kHz (C7). This causes a gain of -32 @ 1kHz. But there is a good chance that it oscillates ... with the simulation.
The whole circuit rather operates as a "differentiator" from DC up to 9kHz. (limited by GBW)

But in a real circuit the mic will have source impedance. Thus I see a better chance for it to be stable. But the gain depends on the mic parameters.
No nice solution, though.

Klaus

tony_lth

### tony_lth

points: 2

#### tony_lth

Hi, Gurus,
This circuits is another company designed for MIC, then this MIC+/- links to my designed module.
Then weird thing happens, the MIC volume is not stable with strong noise.
Attachment is my audio circuits, please note in real there is a 604 ohms resistor between the left ends of C400 and C402.
I am afraid that the other company's A1 & A2 just pass 120R then linked to 10uF, which may cause A1 and A2 unstable.
So, I want to know, how to change my design with the least change to solve this problem？
Many thanks.
Best,
Tony Liu

- - - Updated - - -

Hi, @KlausST,
You mean "series a 10K resistor", is series a 10K resistor on the negative input of A1? I just want to confirm it.
Thanks.

#### Attachments

• 76 KB Views: 8

#### KlausST

##### Super Moderator
Staff member
Hi,

You mean "series a 10K resistor", is series a 10K resistor on the negative input of A1? I just want to confirm it.
With "input" I meant the signal in your circuit that is marked "Vin".
And yes it goes to the inverting input of the amp.

Note: In your circuit of post#9 there also is a 10k series resistor at the input node.

Btw: I don't like (undamped) L-C input in audio circuits, they tend to create resonances. But in your case it's not that bad, because it's damped by the 10k resistor and the 120 Ohms source resistor.

Klaus

#### tony_lth

HI, KlausST,
Thanks.
How to estimate the possible resonance freq?
In post #9, L400 is a bead with 1000ohms, say at 1MHz, then can estimate the L from 2*pi*f*L=1000, i.e. L=159uH.
So then possible resonance freq is f=4KHz, right?
Thanks.
Tony Liu

#### Audioguru

Maybe you did not see the missing 51k input resistor for the first opamp because your schematic was as large as my neighbourhood. I cropped your schematic here:

#### Attachments

• 30.7 KB Views: 6
tony_lth

points: 2