• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 345 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

PCM6240Q1EVM-PDK: Ac-coupled rext calc

user6199457
Prodigy 120 points
Part Number: PCM6240Q1EVM-PDK

Hi,

I have some doubts about ac-coupled rext calc, please help to solve, thanks.

1)     What’s the attenuate factor at first stage? And I couldn’t find the specific value in the datasheet.

2)     PGA amp INCM is 1.35V in the differential table.

Why it is the same as the single-ended table? Whether using 1.35V/2 is be more reasonable in the differential application? If you have more detailed data, please share it for me, thanks.

3)     R tolerance (in %) is 0.8.

Why this value is 0.8? The accuracy of commonly used resistor is 1% or 5%, is the R tolerance 0.95 more reasonable?

4)     swing margin (in V) is 0.2V.

What is the definition of swing margin? 10Vrms or the value we expected(for example 5Vrms) between the max differential input value and 14V or 0V?

Why the swing margin select 0.2V? Is this value little small?

In our practical design, the max differential input is 5Vrms, and the micbias is setup 9V, but I couldn’t know how to select the pull up resistors and how to setup the PGA gain. Thanks very much.

  • 0
    TI__Mastermind 23050 points

    Hi,

    The calculator was designed with a conservative approach in mind to ensure it gives you values that will guarantee you stay within the operating range of the device. Most of the parameters you are describing are not actually included in the calculator that you can find in the PCM6260 product folder as they are not generally necessary for the user to consider when choosing the bias resistors. 

    I can't share all of the details of the input stage as the specific implementation is proprietary, but I can explain why some of the parameters were chosen. 

    1) The input attenuation is 1/5 which is how 10Vrms is achieved with a 2Vrms ADC.

    2) The resistor tolerance mentioned in the spreadsheet is not for external resistors you might buy, but the internal resistors as part of the silicon process. In devices that are not laser trimmed for high precision, it is extremely common for resistance values to vary +/-20% with process. This is not generally an issue because while the absolute value of resistors requires fine trimming, it is much easier to match ratios of resistors. 

    3) If you look at the recommended input range for a microphone recording, it is 0.1V to MICBIAS-0.1V. Designing for 0.2V then ensures some margin from the recommended limits.

    Since you are AC coupled and there is no substantial DC offset that will be amplified by the PGA, you should be able to apply up to 6dB of channel gain without saturating the ADC. You could also use the digital gain instead.

    Best,

    Zak

  • 0 in reply to Zak Kaye
    Prodigy 120 points

    Hi,

    Thanks for your information.

    But I still confused about the calc table.

    Take the EVM for example, the max differential input is setup to 10Vrms, I find the different result from calculated and measured.

    And if I only change the max differential input value based on the other condition keep no change, the VCM will have changed also. 

    The calculate value and actually measured value is displayed as below table.

    I'm very confused how to select the pull up resistor when the max differential input is 5Vrms.

    Please help me find where the problem is？Thanks.

    the max differential input pull up resistor micbias voltage VCM value from the calc table actually measure at EVM
    10Vrms 4k7 8V 6.4V 6.75V
    10Vrms 4k7 9V 7.4V 7.57V
    10Vrms 22k 8V 0.7V 4.41V
    10Vrms 22k 9V 1.7V 4.91V
    5Vrms 22k 9V 5.6V 4.91V

  • 0 in reply to user6199457
    TI__Mastermind 23050 points

    Hi,

    One thing I will point out is that in each of your examples you have violated the condition of the max Rext allowed by the calculator (except for in the 5Vrms case which should work fine). This isn't to say the configuration won't necessarily work, but remember the calculator takes a conservative approach to guaranteeing you won't violate the input conditions. 

    Furthermore, Vcm in is given as a worst case with full scale input assuming the input impedance of the device is at the extreme of -20% with process variation. The value the calculator gives for this and what you actually measure may not match because your device's input impedance is much more likely to be closer to the 50kOhm typical than the 40kOhm worst case. With the bias conditions you have described, it's not likely you would be able to support your desired full swing input, so I would encourage you to keep the value of your bias resistors below the Max Rext allowed given by the calculator and you should not have any issues! 

    Best,

    Zak 

  • 0 in reply to Zak Kaye
    Prodigy 120 points

    Hi, 

    Thanks for your reply.

    But I have other 2 questions in the user's guide(June 2020) of PCM6xx0Q1EVM-PDK Evaluation Module.

    1)AC-Coupled Line Input Biasing

    There are some recommend resistor for 10Vrms(high swing) and 2Vrms(low swing) mode.

    But in the datasheet the CHx_MIC_RANGE bit is only just setup for microphone input mode not for line input,

    so I couldn't understand why you separate high swing mode from low swing mode in line input?

    2) about AC-coupled capacitor

    I find that it is changed from 1uF 0603 25V to 22uF 1210 25V the latest user's guide.

    Could you tell me why you change?

    What factor should we consider about this capacitor? For example, package, rating voltage and so on, thanks.

  • 0 in reply to user6199457
    TI__Mastermind 23050 points

    Hi,

    The line input mode always operates in high swing mode, it is only the mic input setting that allows low swing/high swing settings and we typically recommend operating in high swing mode.

    Yes this capacitor was changed to a larger footprint and larger value to decrease the low frequency distortion contributed by AC coupling and give more flexibility in capacitor selection. If you are interested in how the AC coupling capacitors can impact distortion I wrote an article on the subject that you can find here: 

    Best,

    Zak