TLV320ADC6140: TLV320ADC6140 & ADC6140EVM-PDK Setup & Comparison to PCM1865

Hi Collin :)

Thanks so much for the insight — very helpful. I will give DRE a try at the lower sample rates.

A couple more questions comparing the two chips:

1.) Is there a similar chart of ADC6140 Figure 13 (Input-Referred Noise vs Channel Gain) for the PCM1865 ? Can you comment on noise for the two chips? Along with the built-in PGA, I am using OPA1664AIPWR + digipot for programmable front end mic gain, sensitivity matching and differential signal drive before PCM1865. In this way, noise performance is optimized for my electret mic capsules. Should I expect to use a similar front-end again with ADC6140 after taking into account changes for the lower full-scale input?

2.) If so, out of curiosity, are there any other front-end opamps you would recommend for this purpose (The OPA1692IDGKR looks good, but I don't believe it comes in a quad package)?

Thanks again for your help!

Colin

Hello,

1.)  We don't have the requested data available for the PCM186x, but since the input range is higher yet the performance is the same, we can infer that the input referred noise of the ADC6140 is lower.  

Are you operating with very low-level signals from the microphones?  It's not very common that an external gain stage is required with these devices, but it may be doing a SE-to-DE (or vice-versa) conversion before it reaches the device.  

Additionally, the ADC6140 features fine gain and phase adjustment capabilities which may also eliminate the need for the external adjustment requirements.

2.)  We can certainly help recommend an external amplifier, but would still like to confirm that one is required.  Is it for a receiver or signal conversion stage, or is extra gain really required?

Capsule noise floor is 16 dB-A and there are plenty of folks trying to capture nature sounds where that can become a problem, as they apply lots and lots of gain at capture and in post. Since I also want to capture sounds up to 136 dB-SPL, I use the front end to accommodate large gain changes, while the PGA in the PCM1865 provides finer control. Yes, this is doing SE to DE conversion before PCM1865. The front end also stores sensitivity calibration.

The gain and phase adjustments got me excited, but in practice a gain calibration of -0.8 to 0.7 for analog capsules is not enough. I know it can be for MEMS and some PDM mics, but not the higher SNR analog capsules I've seen, where sensitivity tolerance is commonly +/-3 dB at 1 kHz. The fine 0.1 dB resolution could be helpful, however!

Collin, Thanks so much for your time — I appreciate the review of my approach and will look at the opamps you mention.

I have the ADC6140EVM-PDK here now and am evaluating. I missed the part in the datasheet where it says you cannot change the analog PGA gain once the device is powered up — something I suppose I took for granted with the PCM1865.

So, am I correct that you cannot monitor signal level while changing analog gain or when using the phase and gain calibration? If so, how does one go about using the calibration features without monitoring signal?

To follow up my questions from yesterday:

If I can't use the internal analog PGA when the ADC6140 is in active mode, is there a TI part you can recommend that could potentially do this as an integrated digitally controlled front end? Noise from zero-crossing on gain change is a concern here and the typical PGA2500, PGA2505 seems like overkill, since I would need 8 of them ($$$) and don't need the higher voltage rails. These parts look to be aimed at ultra-sound, but would any be worth looking at for audio in my application?

Zac,

Thanks so much for the additional information about gain staging and control — super helpful. Any comment on Texas Instruments PGA/VGA parts for audio use? For example, would they have audible artifacts on gain change?

Thanks,
Colin