TPA2016D2: THD+N specs on the datasheet of TPA2016D2

Thank you very much for your reply.

Have you used AUX-0025(Switching Amplifier Measurement Filter) with AP2700 when you measured THD+N characteristics?

Yes, AUX-0025 is used as well.

Best regards,
-Ivan Salazar
Applications Engineer

Actually, I measured THD+N under the following conditions, however the result was 0.25% in minimum. Could you point out what is wrong with my measurements?

Measuring equipment : APx585 + AUX-0025

Generator frequency : 1kHz

Measurement object : TPA2016D2EVM

Load (resister) impedance : 8ohm

Power supply : 5VDC

Fixed gain setting : 6dB

Output power : 0.73W

The datasheet says that I should get less than 0.1% under the above conditions.

I'll look for an EVM and try on my side. As mentioned, you may need to remove the LC filter at the output.
I'll get further comments by end of the week, or let you know otherwise.

Best regards,
-Ivan Salazar
Applications Engineer

I've been doing some tests on my side and got some relevant information:

• TPA2016D2 includes some features such as DRC and AGC. When measuring THD performance special care must be taken such that the device does not enter any of these non-linear control features ie. if the signal is being affected by compression, limiter or AGC, the THD reading will be higher than expected.

Please consider the linear regions displayed along data sheet figures 5 through 9.

Best regards,
-Ivan Salazar
Applications Engineer

Thank you for the relevant information.

I retried the measurement and I got the smaller THD-N than previous measurement, 0.15% at 0.23W Pout. However, this result is still larger than the datasheet.

I checked the output LC filter on my TPA2016D2EVM, no capacitor has been mounted (C24, C29, C30 and C31 are empty)

Inductors (L4, L5, L6 and L7) are mounted, but I guess that I don't need to short them because the measurement condition on the datasheet seems to include them.

The I2C setting parameters are as follows.

Attack Time : 6.402 msec/6dB

Release Time : 164.4 msec/6dB

Hold Time : 0msec

Compression Ratio : 1:1

Noise Gate Threshold : 1mV

Fixed Gain : 6dB

Max Gain : 30dB

Limiter : Disable

What is your PVDD level?
Is the gain 6dB as expected? 0.23W should be within the linear region but just want to make sure.
I'll try to replicate on my side, data sheet suggests ~0.8% so you're pretty close to the expected value.

Best regards,
-Ivan Salazar
Applications Engineer

I apologize that my previous information was not enough and included wrong value.

My PVDD level was 5VDC from stable DC power supply.

The output power when I got 0.15% THD+N was NOT 0.23W BUT 0.92W (2.71Vrms on 8ohm resistor)

At the time, input level was 1.36Vrms, therefore 6dB gain was as expected (20*log10(2.71/1.36)=5.99)

Under the above condition, the datasheet suggests ~0.09% and my result (0.15%) is still larger.

Thanks for adding more details, I'll do the same test on my side, however I think the value you're measuring may be correct already as it's close to the plot.
I'll update later today or tomorrow after testing.

Best regards,
-Ivan Salazar
Applications Engineer

I found the reason why my result was worse than the datasheet. I replaced the 8ohm load (resistor) with the other one, and I got the result similar to yours or the datasheet. I guess that the previous resistor includes helix wire, therefore it has unexpected inductance.
I'm sorry to have troubled you.

I'm sorry to visit here again.

Could you measure THD+N under the following conditions and give me the result?

Measuring equipment : APx585 + AUX-0025

Generator frequency : 1kHz

Measurement object : TPA2016D2EVM

Load (resister) impedance : 4ohm

Power supply : 5VDC

[I2C settings]

Attack Time : 6.402 msec/6dB

Release Time : 164.4 msec/6dB

Hold Time : 0msec

Compression Ratio : 1:1

Noise Gate Threshold : 1mV

Fixed Gain : +12dB

Max Gain : 30dB

Limiter : Disable

And could you refer to the attachment (my result under the above conditions)?