OPA1678: OPA1678 single-ended input to differential output with total 32dB

Hello Nick,

One of our Applications Engineers in our Precision Amplifiers group specializes in audio applications. He resides in another time zone much different from the US and likely hasn't seen your e2e post just yet. I am sure he is going to want to assist you with your questions.

Meanwhile, here is a TI Designs, Electret Microphone Pre-Amplifier circuit, that provides information about modeling the electret microphone element in TINA:

www.ti.com/.../tidu765.pdf

Regards, Thomas

Precision Amplifiers Applications Engineering

Hi Thomas,

Yap, I had seen the ECM microphone circuit and the design of the document, but here I still need to know how to design a pre-amp with OPA and 32dB differential output.

It is fine to waiting for your AE.

thanks,

Nick.

Hi Nick,

I have informed our audio products Applications Engineer about this e2e post. He should be getting back to you soon.

Regards, Thomas

Precision Amplifiers Applications Engineering

Hi Nick, Thom,

Thank you for your patience.

Can I ask you about the details of your project?
- Are you preparing phantom power supply for ECM microphone?
- Can we consider JFET is integrated into ECM microphone?

If your project was dedicated to a specific microphone (know characteristics, especially SPL), I believe TI-design; TIDU-765 could be a smart solution to you and might be modified with FDAs such as OPA1637 for the differential output.

However, if you must consider versatile ECM microphone input, I would like to recommend the front end circuit with the high input impedance. Because, as you know, the output impedance of ECM microphones is typically high.

Now I have prepared a simulation with 26dB + 32dB front end circuit in as below.

INA1678-FDA.TSC

Could you please check the attached file?
If it looks OK, I can also modify this with BPF and some external components such as the input common mode voltage generator, etc. to support 3.3V supply voltage condition. 

Thank you and Best regards,
Iwata Etsuji

Hi Iwata san,

Thanks for the reply, my fault, our application is ECM including JFET connect to preamplifier, the detail datasheet is in attached slides. The mic bias (LDO) with LPF is the microphone power which did not shown in previous slides. Besides, the purpose is to replace MAX4063 preamplifier for cost reduction activity, and check the diagram again, seems using only one FDA is the better and simple way to implement?

Actually, I have no idea about ECM microphone output impedance, because it is JFET output. The datasheet shows it is 2.2k ohm but I think it means the pull high 2.2 resistor.

As for your simulation, I think it is not 32dB, right?

OPA1678_MICAMP_diffOUT_1028.pptx

Hi Nick,

Thank you so much for sharing your project details.

I will share the concept of the ECM microphone front end modification from the TI-design with FDAs.

OPA1637_ECM-AFE01.TSC

As you can see in the attached simulation, lager ac-coupling caps are required due to 3.3V supply.
Is it possible to use larger cap value than your original schematic?

Best regards,
Iwata Etsuji

Hi Iwata,

Thanks for the design. 22uF AC-couple capacitor is too large for me.

After studying your circuit, the "R1" for the gain setting is the 330 ohm inside ECM microphone, right? If so, the design must be changed with different ECM... How about using actual R1 at the front of input? I tried to add R1=10k, Rf=200k in your simulation, but the gain is not 20dB, it confused me...

Hi Nick,

The nominal AC gain of my proposed circuit is equal to 20log(R3/R1)+6dB=20log(10k/330) + 6dB = 35.6dB.
If you could change the resistor value R1=10k, R3=200k, the Ac gain is equal to 32dB.

While, the cut-off frequency of low frequency side is defined with R2 and C2.
In order to bias the ECM microphone properly, it is difficult to choose higher resistor value than 2.2kohm, so ac-coupling cap C2 should be increased for the lower cut-off frequency unfortunately.

This is the same situation in another circuit I'm preparing in as below.

OPA1678_ECM-AFE01.TSC

Is it difficult to use higher bias voltage for ECM microphone than 5V (e.g. with boost convertor such as LM27762?), or 9V (same as the TI-design)?

Best regards,
Iwata Etsuji

Hi Iwata,

Yes, it is my concern about the R1=330 ohm, because the circuit is focus on the specific ECM microphone, it is not a common design. Means, when we change another ECM, the circuit have to be changed accordingly.

And, it is difficult to use higher supply voltage, because our system 5V is from switching regulator, it must use an LDO to clean up the switching noise. Considering the LDO working drop output area, Vin=5V, and LDO output 3.3V is a reasonable choice.

Back to my purpose, to replace MAX4063 preapmlifier. I am curious about why it only require set proper Rgain, not consider the constrain we talk about these days. 

Regards,

Nick.

Hi Iwata san,

Thanks, the simulation looks good for me. I am going to rework for evaluation. thanks again!!