Part Number: TLV320AIC3212
HI :
I using an analog microphone as input to TLV320AIC3212IYZFR, I can record , but the volume is too low(hard to hear the sound except I knocking to the mic).
Here is the setting:


This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Part Number: TLV320AIC3212
HI :
I using an analog microphone as input to TLV320AIC3212IYZFR, I can record , but the volume is too low(hard to hear the sound except I knocking to the mic).
Here is the setting:


Hi,
I am a little confused by your schematic without the mic connections to the chip, but I see that MIC P and MIC N are coming from the same place... are these supposed to be differential inputs or two separate inputs? It is best to have one mic coming from 1 spot at a time. These will likely be the same signal so if the device is looking for the difference there will not be much. Can you disconnect MIC N? Are you having issues with the R input as well? And can you share your register configurations for the CM and input selection?
Best,
Mir
HI Jeffres:
Thanks for reply.
MIC N is not connected to audio code, I have rework MIC_N connected to GND, the result is same.R input is not tested yet.
BTW, if I connect lineout from PC instead of MIC, there is no volume problem.
For register:
codec work as slave mode , OSR=128
Book 0 / Page 1 / Register 8: Common Mode Register = 0x00
Book 0 / Page 1 / Register 51: Microphone Bias Control = 0x40
Book 0 / Page 1 / Register 52: Input Select 1: = 0x10
Book 0 / Page 1 / Register 54: Input Select 1: = 0x40
Book 0 / Page 1 / Register 55: Input Select := 0x10
Book 0 / Page 1 / Register 57: Input Select := 0x40
and one more thing, could share me a reference MIC schematic ?
Hi,
You can find the EVM schematic in the EVM user guide, this includes a typical microphone connection at the "MIC IN" jack. You can see there are mic bias connections and this is an AC-coupled input configuration.

Can you share the updated schematic? I am still confused here. Where are the inputs to the chip?
Connecting line out from PC having no volume problem leads me to believe the PGA gain is not enough, or the mic bias is set incorrectly. Again I am confused about the input connections so I am not sure I can recommend what to change yet. Can you start by checking the voltage at the pins once you connect the line out, vs voltage when you connect the microphone?
I see IN2L is connected to L PGA with 10k input impedance, and CM connected to M terminal. IN2R is connected to R PGA, also single ended. Not sure where these connect on your schematic.
Best,
Mir
HI Jeffres:
Maybe I have click "report as abusive" by mistake , so sorry for it.
I have rework the schematic by remove C314 as this:

And measure the MICIBASE(R260) = 1.61v
PGA gain: set :PGA Control - 0x00 / 0x01 / 0x3B (B0_P1_R59) set to 0x5f (47.5db)
Any other register for "PGA gain" ?
Hi,
Yes, this is the PGA gain register. Make sure you have the inputs set to the left PGA, these are set in page 1 register 52-54. If your MIC_P is connected to IN1L, for example, you will want to set register 52 to 0x40 (lowest input impedance will make it the loudest possible as well), and register 54 set to 0x40 as well, since this is a single ended microphone if I understand your schematic correctly.
There is also the ADC digital volume control, this would be page 0 register 83 for left ADC.
Can you check the voltage coming out of the microphone with a scope? This will help us figure out the source of the volume issue.
Best,
Mir
HI Jeffres:
As we use IN2L , I set register 52/54 to 0x20 or 0x30, result is the same.
I have checked the microphone output with a scope, I can almost not message the signal , but when I test the microphone by another project without volume problem, which use codec sgtl5000(mic bias=3.3v), same schematic of MIC, and also could not message the signal with scope unless I knock to the MIC.
So, does it possible that the codec filter the low input signal of MIC ? as the MIC signal is processed by PRB_Rx.
And an another question, as we know, the mic bias can set to 1.6v 2.4v 3.0v 3.3v , which voltage should be set ? according to what ?
Hi,
The mic bias level will depend on the microphone. Please check with your mic data sheet.
There is a chance that the low input signal is just due to the microphone. When you max out the PGA gain do you not see/hear the signal at all on the digital recording? Can you adjust the ADC volume control as well? This is registers 0x53 and 0x54 on page 0.
Can you share a recording of the signal you are getting - this may also help. If you have any scope shots to show what we are talking about that will be good too.
The ADC will automatically filter out DC and very low frequencies, but I am going to assume you want to pick up frequencies in the audible range so this will not be impacted.
Best,
Mir
HI:
I can here sound from recorded file.
I guess that there is the problem of too low input signal from microphone, but what kind of microphone can be used for aic3212 for effective voice record ?
I attached a file which record by set 0_0_0x53=0x28 and 0_1_0x3b= 0x5f
Hi,
Today and Friday is a holiday in the U.S. please expect a delayed response.
Thank you for your patience,
Jeff McPherson
Hi,

Typical full scale input to the ADC is 0.5Vrms (this is for 1 sine wave in the passband). So, any microphone that outputs around 0.5Vrms as the maximum output voltage, or you can add your own amplifier/attenuator circuit to get the signal in a reasonable range. There is a chance that your microphone will output more voltage if you give it a higher MICBIAS, this depends on the microphone again. There may be information in the microphone's datasheet about voltage ratings for the bias as well as sensitivity which is how you can calculate voltage out.
Best,
Mir