BOOSTXL-AFE031-DF1: the detailed feature of Power Amplifier Block

Hi Boru,

Boru Zhang said:

1. Regarding Gain: According to the AFE031 Powerline Communications Analog Front-End.pdf, it was mentioned that the PA is set up with an inverting gain of 6.5 V/V, however, the actual results did not reflect this.

Yes, AFE031 has a fixed gains of 6.5 V/V (typ.) inside of the die design. 

Boru Zhang said:

2. I do not quiet understand the initial negativity of the waveform within the red rectangle I drew. I had expected it to first become positive, similar to the waveform on TX_F_OUT.

Please provide me the power rails, MCU and AFE031 PCB and other related information. Did you provide powers to all the necessary pins in AFE031 IC.   

https://www.ti.com/lit/ug/sbou223/sbou223.pdf?ts=1718203581548&ref_url=https%253A%252F%252Fwww.google.com%252F

https://www.ti.com/lit/an/sprac94d/sprac94d.pdf?ts=1714621319149

Enclosed is an simulation example you requested. I would suggest to simulate each part first, before attempting to perform the entire simulation. 

00_idealOPA_130613_v25 Simulation 06122024.TSC

The PA_OUT may be AC coupled and that is why you see the waveforms that go below the GND. You have to tell me more about the setup. 

If you have other questions, please let me know. 

Best,

Raymond

Thank you, Raymond.

I successfully implemented the B-FSK example, and it worked well.

Currently, I am trying to use 16QAM modulation for a faster baud rate.

I have connected AFE031-DF1 with LAUNCHXL-F280049C, and applied a 10V DC voltage on TB2.

I modified the code based on the boostxl_afe031_f28004x_dacmode example (found in C2000Ware_5_02_00_00\device_support\f28004x\examples\boostxl_afe031_f28004x_dacmode) to send the desired waveform. 

Here is the waveform on pin TX_PGA_OUT pin.

the points sent by DMA to DA were generated by the following python code :

steps = np.linspace(0, 1, 20)
QAM_Table = [
    1.5 - np.cos(2 * np.pi * steps) - y * np.sin(2 * np.pi * steps) for y in (-1, -0.5, 1, 0.5)
]

['0.50', '0.88', '1.33', '1.79', '2.22', '2.58', '2.82', '2.91', '2.86', '2.65', '2.32', '1.90', '1.44', '0.99', '0.59', '0.29', '0.12', '0.10', '0.23', '0.50']
['0.50', '0.72', '1.02', '1.37', '1.74', '2.08', '2.36', '2.55', '2.62', '2.57', '2.40', '2.14', '1.81', '1.44', '1.08', '0.77', '0.53', '0.40', '0.39', '0.50']
['0.50', '0.23', '0.10', '0.12', '0.29', '0.59', '0.99', '1.44', '1.90', '2.32', '2.65', '2.86', '2.91', '2.82', '2.58', '2.22', '1.79', '1.33', '0.88', '0.50']
['0.50', '0.39', '0.40', '0.53', '0.77', '1.08', '1.44', '1.81', '2.14', '2.40', '2.57', '2.62', '2.55', '2.36', '2.08', '1.74', '1.37', '1.02', '0.72', '0.50']

These points are sent by DMA at 1MHz, resulting in a total duration of 80µs for 80 points.

The TX_PGA_OUT waveform appears to be correct.

The capacitor between the TX_F_OUT pin and PA_OUT acts as a high-pass filter, so the waveform in the first 20µs should resemble  -cos(2πf) + sin(2πf), which I have drawn above. (f = 50kHz) Consequently, the waveform initially rises.

Therefore, I still do not quite understand why the PA_OUT waveform initially drops. Could you please provide further clarification?

Please let me know if you need further information.

Best regards,

Boru Zhang