• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 398 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.

Original question:

Compiler/PCM1863: About PCM1863 input connection and SNR

PCM1803A: VINL is having problem near 0V input, VINR has range issue, with DC measurements

Richard Burnett
Prodigy 220 points
Part Number: PCM1803A

Hello!

I have a circuit with a PCM1803A setup to take a 0-5V input signal.  I've scoped both signals going into VINL and VINR and looked at the recorded values.

I have verified that the input is in the proper range (no negative voltage)

I have verified my 5V and 3.3V supplies

I have verified my connections of VREF1 and VREF2 to ground with the recommended cap values.

Both channels seem to have a pretty big dead zone up to like 1V where no input is detected.  Other ADC using the same input topology show full range no problem.

VINL at like 0.5V or so shows a reading coming out.  It's a VERY narrow range that this happens, but it's a blip I can see as I rotate a voltage coming in.

Driving into ADC is a R2R opamp -> 100R -> VIN, with a 10nF cap to ground.  

This is DC measurements.  

With largest oversampling, there is just a blip
With smallest oversampling, as it goes to zero, it rises up.  

Vref2 shows 5V
Vref1 shows 1V

I notice that I have the wrong caps for the larger cap on vref (non-electrolytic) could this cause the problem?

  • 0
    Prodigy 220 points
    I swapped out the 10uF for electrolytic on the VREF lines and no difference. Whatever is causing that blip to occur at very low voltage on VINL is just bizzare.
  • 0 in reply to Richard Burnett
    TI__Guru 57800 points

    Hi, RIchard,

    Can you please provide more details on the input used?, if possible any scope capture will help. You mention that the data you are sharing is in DC; this ADC actually is used for audio applications, where DC is removed. The device has an integrated HPF, which will remove the DC signal provided to the device.

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Diego Melendez
    Prodigy 220 points

    Sure.  First, for some historical reference, this ADC has been used in a lot of cases where DC voltages are being used.  Specifically, in the audio interface world, where you want to capture CV signals along with audio signals.

    For this reason, I selected this ADC.  I am using it in an environment where I can control the DC offset and scale to fit the waveform properly, and I need that from low to high frequency, and to maintain the DC level.

    What I am confused about is how in the Datasheet I would know that the signal even in DC is expected to sit around 2.5 volts and that it should extend up to 4.0 volts and down to 1.5V but not be outside that range?

    Empirically I have found that if I am outside that range, strange things happen on VINL (but not VINR for whatever reason).

    This is where I saw the 3Vpp range (Look at the thread titled "PCM1803A: Vref2 for PCM1803A"

    Then if you look at the thread "PCM1803A - full scale voltag" (sp) you will see this response from TI (from Justin):

    "Furthermore, this input should be centered around 2.5 V or sent through an AC coupling capacitor. As the ADC will have a range from 1.5 V to 4.5 V if a signal is passed to the input without an AC coupling cap, then 1.5 V from ground will produce a negative full scale output."

    This is pretty coincident from what I am seeing, but I just don't understand where I should have seen this in the Datasheet to know this is what I should have been expecting.

    I think the output circuitry that I am driving with is irrelevant at this point.  It's just an OPA342NA op amp, feeding in a signal from 0V to 4V at the moment that I am now modifying to do a range of 1.5V to 4.5V.

    I am just looking for clarity that I understand this correct.





  • 0 in reply to Richard Burnett
    TI__Guru 57800 points

    Hi, Richard,

    Thanks for the clarification. The input level behavior can be explained as follows:

    In the Electrical Characteristics table of the datasheet, you can find the parameter Vi (Input voltage), which corresponds to the  full-scale input level accepted by the device. This level is specified as 0.6×VCC (Vpp), where VCC is 5V, so the maximum input voltage supported is 3Vpp. Then, you can find that the center voltage (or common mode voltage) is 0.5×VCC, which is 2.5V. This sets an operating range from 1V to 4V, with a common mode of 2.5V.

    Actually it seems there is a typo in the comment you added from the other thread, as it seems that a common mode of 0.6×VCC is assumed.

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Diego Melendez
    Prodigy 220 points

    Excellent!  I see now where this is shown in the table, it was not obvious at first.