• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 898 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.

TAS5709: TAS5709 high THD+N @ low output power

Nicole Han
Intellectual 2230 points
Part Number: TAS5709

Hi, team,

My customer is using TAS5709, found THD+N is very large when the Master volume is small. As the red font shown in the following table.  For example, when the Master Volume= 0x72, THD+N reaches 80.1%. This should not be related to noise. Please help to analyze the issue, thank you!

 

Input Signal

Master Volume

L

R

Output Power (W)

THD+N (%)

Output Power (W)

THD+N (%)

1000Hz_-9dB.wav

0x24

8.21

0.423

8.11

0.341

1000Hz_-9dB.wav

0x27

5.887

0.306

5.797

0.237

1000Hz_-9dB.wav

0x2A

4.192

0.214

4.131

0.161

1000Hz_-9dB.wav

0x2D

2.984

0.155

2.93

0.117

1000Hz_-9dB.wav

0x30

2.116

0.12

2.081

0.09

1000Hz_-9dB.wav

0x33

1.502

0.103

1.473

0.075

1000Hz_-9dB.wav

0x36

1.064

0.08

1.045

0.057

1000Hz_-9dB.wav

0x39

0.756

0.052

0.739

0.041

1000Hz_-9dB.wav

0x3C

0.536

0.04

0.526

0.034

1000Hz_-9dB.wav

0x3F

0.3784

0.033

0.3715

0.031

1000Hz_-9dB.wav

0x42

0.2675

0.031

0.2631

0.03

1000Hz_-9dB.wav

0x45

0.1894

0.032

0.1859

0.032

1000Hz_-9dB.wav

0x48

0.134

0.0031

0.132

0.031

1000Hz_-9dB.wav

0x4B

0.0949

0.034

0.0931

0.034

1000Hz_-9dB.wav

0x4E

0.067

0.039

0.066

0.038

1000Hz_-9dB.wav

0x51

0.0475

0.04

0.0467

0.04

1000Hz_-9dB.wav

0x54

0.0337

0.06

0.0331

0.06

1000Hz_-9dB.wav

0x57

0.0239

0.07

0.0235

0.07

1000Hz_-9dB.wav

0x5A

0.017

0.08

0.0167

0.08

1000Hz_-9dB.wav

0x5D

0.012

0.094

0.012

0.096

1000Hz_-9dB.wav

0x60

0.0086

0.11

0.0085

0.11

1000Hz_-9dB.wav

0x63

0.0061

0.11

0.006

0.13

1000Hz_-9dB.wav

0x66

0.0044

0.13

0.0043

0.13

1000Hz_-9dB.wav

0x69

0.0032

0.15

0.0031

0.15

1000Hz_-9dB.wav

0x6C

0.0023

0.17

0.0023

0.17

1000Hz_-9dB.wav

0x6F

1.69 mW

0.8

1.67 mW

1.7

1000Hz_-9dB.wav

0x72

1.25 mW

20

1.25 mW

40

1000Hz_-9dB.wav

0x7B

580 uW

80.1

590 uW

78.7

1000Hz_-9dB.wav

0x7E

470 uW

74

483 uW

72

1000Hz_-9dB.wav

0x87

303 uW

55.3

318 uW

53.6

1000Hz_-9dB.wav

0x8D

256 uW

42

272 uW

40.6

1000Hz_-9dB.wav

0x96

225 uW

26.8

243 uW

25.7

1000Hz_-9dB.wav

0x99

220 uW

22.8

237 uW

21.8

1000Hz_-9dB.wav

0x9C

216 uW

19.3

235 uW

18.5

1000Hz_-9dB.wav

0xAE

212 uW

7

227 uW

6.7

1000Hz_-9dB.wav

0xBD

211 uW

2.9

227 uW

2.7

1000Hz_-9dB.wav

0xC0

210 uW

2.43

226 uW

2.3

1000Hz_-9dB.wav

0xC3

210 uW

2.03

226 uW

1.95

1000Hz_-9dB.wav

0xC6

210 uW

1.72

226 uW

1.65

1000Hz_-9dB.wav

0xC9

210 uW

1.48

226 uW

1.42

1000Hz_-9dB.wav

0xCC

209 uW

1.26

226 uW

1.21

 

  • 0
    TI__Mastermind 24835 points

    Hi Nicole,

    I think this data looks reasonable. When the output power is very low, the THD+N performance is dominated by noise actually. Since the noise floor is constant, the THD+N performance degrades with output power decreases. For the very low output power (e.g. uW level), it makes no sense for the mid-power class-d device evaluation. Please refer to this document(www.ti.com/.../slaa114.pdf), we have very detailed explanation on Page 11 and 12 in it.

    Best regards,

    Shawn Zheng 

  • 0 in reply to Shawn Zheng
    TI__Intellectual 2230 points
    Hi, Shawn,

    I couldn't find the document. could you share it again?

    I thought it was caused by noise at the beginning, but you can see from THD+N became 80% at580uW, then it started to become better. If this is caused by noise, THD+N can only become worse with the drop of output power.
    I also checked it on TPA3118 EVM, and found it doesn't get worse so fast.
  • 0 in reply to Nicole Han
    TI__Mastermind 24835 points
    Hi Nicole,
    The previous link is bad, please try this one. www.ti.com/.../slaa114.pdf
    You could find detailed info about THD measurement on Page 11 and 12. Again, it makes no sense for THD+N measurement with so low output power. In this case, AP even can't distinguish the valid signal from the noise floor. Can you read the measured frequency from AP? Is it expected and stable? If not, the measured THD+N number is not trust worth.
    Best regards,
    Shawn Zheng
  • 0 in reply to Shawn Zheng
    TI__Mastermind 24835 points

    Hi Nicole,

    As we talked about this on phone, I believe this question has already been resolved. Could you please help to close it? Thanks!

    Best regards,

    Shawn Zheng