How to measure the THD+N of TAS5706A

Hello Jian,

I'd be happy to look into this for you. A few quick follow up questions:

What are you using to measure the thd?

What voltage are you putting into the device? (PVCC)

Could you tell me a little more about the load you are using? What is the resistance of the speaker?

Could you include some oscilloscope screenshots of the input signal and output signal? This could be very useful to see what is causing the THD discrepancy.

I look forward to the response,

Nate

Hi，Nathaniel 

   1. The instrument that is used to measure THD is harmonic distortion analyzer（not  Audio Precision）.It is Manufactured by factory of radio instrument of beijing （not existing）

2. PVCC is18V.It is produced by Transistor power.

3. The resistance of the speaker(5W) is 8ohm.

4.About oscilloscope screenshots of the input signal and output signal，I‘m sorry that I have come off work.But I can do it next monday.

Hmmm... yes for those settings your thd does seem quite high. I'll take a look at the screenshots on monday and if I can't determine the cause I'll see if I can recreate it in the lab and find out what could be going on.

Thank you for the info!

Nate

Hi,Nathaniel

  The following graphs are oscilloscope screenshots of the input signal and output signal.The parameters of the amplifier is configured as follows。By the way，The interfaces of C and D is not connected to speaker.

frequency(KHZ)  Vin（mVrms） volume（dB） Vout（Vrms）

6                         419.6                -12                   2.12

graph 1. oscilloscope screenshot of the input signal

graph 2. oscilloscope screenshot of the output signal

Jian,

I'm actually hearing from another person about getting an inflated thd+n level with this EVM. I'll try to create the issue and see if I can find some possible solutions.

Nate

Jian,

I just set up the THD+N test using your same settings and got a thd+N of 0.15%. I used a single ended analog input, 18 V pvdd, 5 volts for VIn. .66 watts, and 6khz sine input to an 8 ohm load. I'm using the audio precision to measure the values. The values I got here line up with the results found in the datasheet.

From this, I would suspect there is one of two reasons for seeing an increased THD+N:

1) There is some error in the total harmonic distortion meter (you would have to refer to the manufacturer to see if it has any limitations of resolution or certain thresholds), or

2) There is some setup issue that we are overlooking.

One thing that stood out is how your speaker is connected to your evm. I'm not seeing it connect to the A and B jack near JP1 and JP2. How is your speaker attached to the output of the tas5706evm2?

Nate

Hi,Nate

From oscilloscope screenshot of the output signal,I think that it may be not problem of  harmonic distortion meter. I also suspect that there is some setup issue.Can you send the picture of measuring connection of TAS5706EVM2 to me？Also，please give me a quick-start setup of GUI and file of BTL .ini （tested）.thanks a lot

Hello Jian,

Here is a document I put together to describe the exact steps I took to run the test. It includes pictures of the setup along with screen shots of how I set up the GUI. 3324.How to measure THD+N TAS5706EVM.pdf

I'm not familiar with the .ini file type you are looking for. Which program outputs the .ini?

Hope this helps! Please let me know if you need anything else!

Nate

Hi,Nate

I have seen the document about measuring THD+N of the TAS5706EVM2.It seem to be similar with method provided by the document，besides AP.Unfortunately ，I could not own the analyzer and therefore not follow the document. Did you measure output signal of TAS5706EVM2 with oscilloscope ?Can you provide connection to measure output signal with a oscilloscope for you？

Hello Jian,

When I run the tests, I usually do use an oscilloscope on the outputs (and inputs) to make sure all of the signals look correct. When connecting it to the TAS5706EVM2.

When measuring the output I connect the scope probe to the positive side of the load.

You will have to be careful with attaching the ground clip of the oscilliscope prob to the negative side of the load, as it's a differential output. So if you have another probe that is grounded it could short the negative signal.

Did this answer your question? I'd like to confirm this is what you were asking.

Nate

Hi,Nate

I mean that distortion of output signal was large when I used a oscilliscope to measure output of TAS5706EVM.Did you experience it？How should I resolve it ？