• State
  • Replies 9 replies
  • Subscribers 30 subscribers
  • Views 201 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

TAS2563: THD+N measurement for Smart amplifier TAS2563

Rekha Menon
Prodigy 110 points
Part Number: TAS2563

Tool/software:

Hi Ivan,

  1. How is the THD+N measured in case of a smart amplifier like TAS2563? Is it any different from the normal Class-D amplifier?  
  2. We measured the electrical signal output distortion and notice that the peak captured is more dominant at 3K and 5K than the input frequency 1K sent. This was captured at a higher volume (Vol Level=9). The sine wave measure is attached below as well. Kindly let us know your inputs. Also, we observe that the THD+N captured at 1K is not in aligned with the datasheet value.

 

Fig 1: THD+N electrical output at 1K Sine input with dominant 3K and 5KHz at Vol level=9 (DVC value=0x0FEC9E0F)

 

 

 

Fig 2: Input signal Sine 1KHz signal at Vol level=9 (DVC value=0x0FEC9E0F)

 

3. However, with the volume level being lowered, we expected the THD+N value might have a decrease, but it still is dominant at 3K and 5KHz. This was captured at a lower volume (Vol Level=1). The input Sine wave is much linear in this case.

 

Fig 3: THD+N electrical output at 1K Sine input with dominant 3K and 5KHz at Vol level=1 (DVC value=0x0656EE3D)

 

 

Fig 4: Input Sine 1KHz signal at Vol level=1(DVC value=0x0656EE3D)

 

 

4. Also, what is the recommended distance from the speaker to mic for testing SPL and other parameters in case of miniature speakers.

BR,

Rekha

  • 0
    TI__Guru** 100965 points

    The odd harmonics are expected to be higher compared to even harmonics. Are you measuring higher THD than expected?

    SPL measurement distance I’ve seen is from 10cm up to 1m, depending on the speaker and the application. Usually, speaker data sheet includes SPL measurement including the distance used for the test.

    Best regards,
    -Ivan Salazar
    Applications Engineer

  • 0 in reply to Ivan Salazar
    Prodigy 110 points

    Hi Ivan,

    Yes, we are seeing higher THD+N (around 300%) with input of Sine 1KHz in anechoic chamber. What is the way to reduce this?

    Is there a user guide or recommendation for THD+N measurement for Smart amplifier? 

    BR,

    Rekha

  • 0 in reply to Rekha Menon
    TI__Guru** 100965 points

    Hi Rekha,

    Is this a microphone-based test? Do you have a recording you can share? Is the output is clipping severely?

    Note that electrical THD+N and acoustic THD testing are not same.

    Best regards,
    -Ivan Salazar
    Applications Engineer

  • 0 in reply to Ivan Salazar
    Prodigy 110 points

    Hi Ivan,

    No, this was electrical THD+N measurement and not a mic-based test. 

    Earlier, we did measure the acoustic THD+N using a mic and found it to be high. Thus, we decided to test the electrical THD+N and observed the above details.

    Do suggest what measures can be taken to reduce the electrical THD+N. 

    BR,

    Rekha

  • 0 in reply to Rekha Menon
    TI__Guru** 100965 points

    Hi Rekha,

    Can you share a time-domain capture with a couple cycles? Is the waveform clipping?
    What load are you using for this test?
    Also please make sure you're testing in ROM mode, in tuning mode the pilot tone at 60Hz will affect THD+N measurements.

    Best regards,
    -Ivan Salazar
    Applications Engineer

  • 0 in reply to Ivan Salazar
    Prodigy 110 points

    Hi Ivan,

    1. The waveform is not clipping as attached below.

    2. The time-domain capture is attached below. We still observe the dominant frequency to be 5 KHz.

    3. The load used for the test is a 8-ohm speaker.

    4. As suggested, we tried checking the THD+N in ROM mode. We observe the below measurement in ROM mode. 

    5. We are using the TAS2563 in tuning mode for our use case. Why is the THD+N to be measured in ROM mode instead of tuning mode? Also, how is the pilot tone of 60 Hz affecting the Sine 1 KHz signal?

    BR,

    Rekha

  • 0 in reply to Rekha Menon
    TI__Guru** 100965 points

    Hi Rekha,

    What is your test signal? Are you testing with sinewave at all?

    THD+N is measured by playing a given single-frequency tone, applying a notch filter on that specific frequency, and measuring the rest of the band after removing the test tone, usually with a bandwidth from 10Hz to 20kHz.
    If there is another tone not considered by the notch filter, this will appear as bad THD+N.
    The pilot tone (60Hz) is usually below the threshold where the micro-speakers are able to produce sound, so even if this affects electrical THD+N measurements, it should not have acoustic effects. However, if the pilot tone is hearable and cause problems, it can be fine-tuned down to 16Hz.

    Some literature about THD+N:

    Best regards,
    -Ivan Salazar
    Applications Engineer

  • 0 in reply to Ivan Salazar
    Prodigy 110 points

    Hi Ivan,

    Yes, we are testing with Sine wave 1 KHz signal. The input sine 1KHz signal used is attached below.

    1KHz_signal.zip

    The following is the electrical output of the samp, Without the input sine 1KHz being played, 

    The following is the electrical output of the Smap, when the Sine 1KHz signal is played.

    We can observe a mix of many frequencies in the output. Could you please explain this behaviour?

    Best regards,

    Rekha

  • 0 in reply to Rekha Menon
    TI__Guru** 100965 points

    Hi Rekha,

    That's the Class-D switching signal, you must use a filter such as AUX-0025 for THDN measurements.

    Best regards,
    -Ivan Salazar
    Applications Engineer