About the DW3120 External PA

[usb]

Can you provide the external PA application note documents mentioned in the DW3000 manual?
Thank you！

[Emre_Ozbas_Qorvo]

Hi @usb ,

The application note mentioned in the DW3000 user manual hasn’t been published yet, but if you have any questions regarding external PA usage, we are happy to help you from here.

Kind regards,
Emre

[usb]

Hi Emre，

We are using DW3120+external PA to implement PDOA function,
In terms of hardware, GPIO4 and GPIO6 of DW3120 are configured to drive external PA,

The control logic we envision is:

1. When GPIO4 and GPIO6 are HIGH and LOW, DW3120 emits and gains through external PA;
2. When GPIO4 and GPIO6 are LOW and LOW, DW3120 is in the receiving state, and the external PA is bypassed directly.

For this purpose, our firmware is set as follows(DwT_PA_ENABLE=0x02) and emits data in a loop:

if (dwt_initialise(DwT_DW_IDLE) != DwT_SUCCESS)
{
return -1;
}
dwt_Setleds(DWT_LEDS_ENABLE | DwT_LEDS_ INIT_BLINK);
dwt_setlnapamode(DwT_PA_ENABLE);
if(dwt_configure(&config) != DwT_SuccEsS)
{
return -1;
}

while(1)
{
dwt_writetxdata(sizeof(tx_poll_msg), tx_poll_msg, 0);
dwt_writetxfctrl(sizeof(tx_poll_msg), O, 1);
dwt_starttx(DWT_START_TX_ IMMEDIATE);
HALGPIO_TogglePin(LED_1_GPIO_Port, LED_1_Pin);
HAL_Delay(2000);
}

At present, we can see the following waveforms on the oscilloscope (channel 1 is GPIO4, channel 2 is GPIO6).

We want to confirm if the waveforms are correct, especially since the falling edge of GPO4 has a large number of nanosecond level pulse width groups, which cannot drive the external PA because it takes at least 500ns for the PA to switch from the sending state to the bypass state.

Thank you!

[Emre_Ozbas_Qorvo]

Hi @usb ,

Thanks for the explanation of your issue.

I think the main source of your issue is not disabling Fine-grain Tx sequencing. In case if it is enabled, the GPIO pin that toggles the PA control pin would try to disable the PA in between pulses. Could you try to set the command below into your code and try again?

dwt_setfinegraintxseq(0);

Kind regards,
Emre

[usb]

Hi Emre，

As you said, after adding this line of code, the waveform of GPO4 no longer has short pulses.

I noticed that the duration of GPO4 high level is 570us.I want to know, after GPIO4 is pulled high, how long is the interval between the RF envelope output to the TX port of DW3120? Can this time be set?

There is another question. Our current firmware configuration is as follows. How can we adjust it to increase the distance further?

static dwt_txconfig_t txconfig_options = {
0x34, /* PG delay. /
0xfdfdfdfd, / TX power./
0x0 / PG count */
}；

static dwt_config_t config = {
5， // Channel number//
DWT_PLEN 128, // Rceabkle length. Used
DWT_PAC8, // Rceable acquisition
9， // TX ereable code. Usec
9， // RX eceable code.
1， // 0 to use standard 8 s]
DWT_BR 6M8, // Data rate
DWT_PHRMODE_EXT, // PHY header mode
DWT_PHRRATE_STD, // PHY header rate
(129 + 8 - 8), // SFD timeout (pceamble
(DWT_STS_MODE 1 | DWT_STS_MODE_SDC), // S
DST_STS_LEN_256, // STS length see allowe
DWT_PDOA_M3 // PD0A mode 3
}；

Thank you！

[Emre_Ozbas_Qorvo]

Hi @usb ,

The duration of GPIO4 to be pulled high depends on your transmitted frame duration. So having a 540us duration indicates that your frame duration is approximately half a millisecond. The duration of the frame depends on the radio configurations that you shared and the number of bytes that you want to transmit in a frame.

To optimize the radio configurations to improve max range, it is possible to tweak some of the settings, but there is no significant change in the maximum range performance because there is a trade-off between the sensitivity improvement and the frame duration that you need to calibrate the Tx power to be -41.3 dBm/MHz. I can see that you are using the 0xfdfdfdfd setting, which is likely exceeding the regulatory limit.

Kind regards,
Emre

[usb]

Hi Emre，

Ok， At present, PA can be controlled and data transmission and reception are normal. Our RF front-end also integrates LNA, and we have enabled the LNA function. The control logic looks correct on the oscilloscope, but the received data is quite strange. The red and blue colors in the picture are PDOA1 and PDOA2 collected after one rotation of the turntable. Normally, it should be a single wavelength sine wave. What is wrong with this graph now?

Thank you！