nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Tue, 13 Feb 2024 15:25:25 GMT

Hi Schmid,

I am trying to jog my memory and source changes on this. I definitely had some copy/paste issues with the peripheral settings being used for the central settings for the 8 dBm power setting. Hung had pointed those out a few replies back.

For the Wireshark issue, I had tried a number of versions of Wireshark but version 4.0.8 seemed to work. I still found it somewhat temperamental as it did not always pickup all the packets but it was able to help resolve.

I have included the central and peripheral NUS, CodedS2 test programs I had used. I am not sure if they include everything but it should be from that timeframe.

Hope it helps.

devzone.nordicsemi.com/.../peripheral_5F00_uart_5F00_nus_5F00_codedS2_5F00_2024_5F00_09_5F00_06.zip anddevzone.nordicsemi.com/.../central_5F00_uart_5F00_nus_5F00_codedS2_5F00_2024_5F00_09_5F00_06.zip

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Schmid — Tue, 13 Feb 2024 09:50:55 GMT

Hello Trent, I have the same Error using the same configurations you described. Would you like to share how you resolved this issue?

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Fri, 22 Sep 2023 17:32:11 GMT

Hi Hung,

Finally back at this issue. We have resolved the issue with the code. The RF sniffer helped determine which side was the issue.

Thanks again for your help.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Thu, 07 Sep 2023 12:32:18 GMT

Hi Trent,
For the sniffer to be able to track the communication, you would need to select the advertiser in the device drop down list.

You can take a look at exercise 2 in this lesson in our Bluetooth Academy here: https://academy.nordicsemi.com/courses/bluetooth-low-energy-fundamentals/lessons/lesson-6-bluetooth-le-sniffer/

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Wed, 06 Sep 2023 15:57:49 GMT

Hi Hung,

Sorry - just one more thing.

With Wireshark, I have no issues capturing the advertising and anonymous data coming from the peripheral device.

The NUS communication is happening but I don't see any Wireshark packets. There are no filters and these are my settings:

I noticed this comment: https://devzone.nordicsemi.com/f/nordic-q-a/16674/no-data-to-wireshark-from-nrf-dongle where they indicate Wireshark version is important.

I have 4.0.8

Is there another setting I am missing or should I try an older version.

Thanks!

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Wed, 06 Sep 2023 15:39:11 GMT

Thanks Hung,

Thanks for letting me know about this setting. I now have the Wireshark system working with an older version of Python.

This confirmed that the peripheral device is sending advertisements properly and the central device is not responding correctly. I have communication working now. There was a coding bug on my part on the central devices scan_filter_match.

It has been resolved so you can close this ticket down.

Best regards,

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Wed, 06 Sep 2023 14:48:02 GMT

Hi Trent,
If you use the nRF52840 as the sniffer, it should be able to catch the CODED PHY advertising packet.

Make sure this box is ticked:

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Wed, 06 Sep 2023 14:04:09 GMT

Hi Hung,

The issue is related to the original question: "I am trying to create a working example of the central and peripheral NUS examples running under CODED PHY. I also am setting the TX power level to 8 dBm and am having some issues."

The 8 dBm issue is resolved and I am just trying to get CODED PHY working. I have successfully used the following example for testing CODED PHY since it allows you to adjust settings dynamically.

https://developer.nordicsemi.com/nRF_Connect_SDK/doc/latest/nrf/samples/bluetooth/throughput/README.html

The peripheral_hr_coded and central_hr_coded examples also work. I also have the sniffer running.

We get the following MAC address information:

PERIPHERAL
C1:97:BB:4C:E8:AF (193, 151, 187, 76, 232, 175)

CENTRAL
E0:CD:01:60:9E:3C (224, 205, 1, 96, 158, 60)

None of these address' show up in the sniffer output, unless I am misunderstanding the sniffer information.

So for the above examples and our NUS version, the sniffer cannot pick up the CODED PHY information. For the sniffer, I am using the example.py only as the Wireshark plugin does not install properly. I am sorting this out since it likely some python dependency however all the requirements are met as per the install instruction check.

If the Wireshark version can pickup the CODED PHY connections, then I will continue to pursue this. Any other insight is welcome.

Best regards,

Trent.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Wed, 06 Sep 2023 09:19:56 GMT

Hi Trent,
Please clarify the issue you described here is related to the original question or not ?
Have you managed to do a connection before ?
Please create a new ticket if it's not related to the issue before.
You can take a look at the peripheral_hr_coded and central_hr_coded to see how they work if you plan to do CODED PHY advertising and establishing connection when doing CODED.

Please use the sniffer to track the communication. It will provide insight on why the connection can't be established.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Tue, 05 Sep 2023 19:43:19 GMT

Hi Hung,

I have been away from this for a few weeks but am back at trying to resolve the connection issue.

I adjusted the filter settings in the Central code to allow for a specific BT name. This at least allows the filter to acknowledge it sees the Peripheral device as follows:

[00:18:31.528,594] <inf> cent_uart_CS2: scan_filter_match: Filters matched. Address: C1:97:BB:4C:E8:AF (random) connectable: 1
[00:18:34.530,975] <inf> cent_uart_CS2: connected: Failed to connect to C1:97:BB:4C:E8:AF (random) (2)

The error code is a 2. Not confident of the meaning of this when I look into the error listings - is this ENOENT or no such file or directory?

The device shows as being connectable. The listed device MAC is correctly referencing the Peripheral device.

Is there anything that is preventing this from finishing the connection properly?

I have attached updated code for both central/peripheral.

devzone.nordicsemi.com/.../0624.central_5F00_uart_5F00_nus_5F00_codedS2.zip

devzone.nordicsemi.com/.../2843.peripheral_5F00_uart_5F00_nus_5F00_codedS2.zip

Thanks for your assistance.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Mon, 31 Jul 2023 13:56:43 GMT

Hi Hung,

Thanks for this - it is setting the power now properly on connection (without CODED PHY).

I still have some issue when CODED PHY is enabled as it does not make the connection. I will review the code again and see if there is anything obvious and let you know.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Mon, 31 Jul 2023 09:26:41 GMT

Hi Trent,
You should call set_tx_power(BT_HCI_VS_LL_HANDLE_TYPE_CONN, 0, 8); when you are in a connection, not when you do scanning and don't have any active connection.
For example I called it inside connected() function and I got this in the log:

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Fri, 28 Jul 2023 14:35:22 GMT

Thanks Hung,

I missed the handle definition being for an advertiser; makes sense.

I have removed the child_image folder as this is a leftover from the example code as there is no intent to use RPC.

I have set the code as follows:

	set_tx_power(BT_HCI_VS_LL_HANDLE_TYPE_CONN, 0, 8); // should be 8dBm now.
	get_tx_power(BT_HCI_VS_LL_HANDLE_TYPE_CONN, 0, &txp_get);

From this, I still get:

Set Tx power err: -5 reason 0x00
Read Tx power err: -5 reason 0x00

I have moved the set and get calls around in the code so it is before and after scanning has started but no change to the result.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Fri, 28 Jul 2023 08:14:10 GMT

Hi Trent,
I can see that you called :

set_tx_power(BT_HCI_VS_LL_HANDLE_TYPE_ADV, 0, 8);

Could you explain why you used BT_HCI_VS_LL_HANDLE_TYPE_ADV ? It's for advertising. The central wouldn't do any advertising.

In your prj.conf you only configured your device as CONFIG_BT_CENTRAL=y and no CONFIG_BT_PERIPHERAL=y. This explains why it crashed when you call set_tx_power(BT_HCI_VS_LL_HANDLE_TYPE_ADV, 0, 8);

If you want to set TXPower when the device is in connection you should call:

set_tx_power(BT_HCI_VS_LL_HANDLE_TYPE_CONN, 0, 8); // should be 8dBm now.

instead.

In addition, I notice that you have rpc_host.conf in your child_image folder, do you plan to use the RPC ? I'm not so sure if any change needed for the TXPower setting if RPC is used.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Thu, 27 Jul 2023 14:33:01 GMT

Hi Hung,

I have tested the example and it works. I don't get any error or incorrect reading of the power reading. This code is behaving the same as my above code labelled in peripheral_uart_nus_codedS2.zip for the Peripheral configuration - it also works with the power reading. (no fault and the power sets and reads fine.)

The issue only occurs when I apply the same code and configuration to the Central setup. The code previously attached above named: central_uart_nus_codedS2.zip

Hope this helps clarify things.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Thu, 27 Jul 2023 14:15:18 GMT

Hi,
Are you testing with my example ?
I just did a test with it this morning and got no error at all ?
Did you make any modification except for removing CONFIG_BT_CTLR_ADVANCED_FEATURES=y ?
Please also try testing with the hci_pwr_ctrl sample, it's almost the same.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Thu, 27 Jul 2023 13:55:00 GMT

Hi Hung,

I have turned both off and the code hard faults in this code:

static void set_tx_power(uint8_t handle_type, uint16_t handle, int8_t tx_pwr_lvl)
{
  struct bt_hci_cp_vs_write_tx_power_level *cp;
  struct bt_hci_rp_vs_write_tx_power_level *rp;
  struct net_buf *buf, *rsp = NULL;
  int err;

  buf = bt_hci_cmd_create(BT_HCI_OP_VS_WRITE_TX_POWER_LEVEL,
        sizeof(*cp));
  if (!buf) {
    printk("Unable to allocate command buffer\n");
    return;
  }

  cp = net_buf_add(buf, sizeof(*cp));
  cp->handle = sys_cpu_to_le16(handle);
  cp->handle_type = handle_type;
  cp->tx_power_level = tx_pwr_lvl;

  err = bt_hci_cmd_send_sync(BT_HCI_OP_VS_WRITE_TX_POWER_LEVEL,
           buf, &rsp);
  if (err) {
    uint8_t reason = rsp ?
      ((struct bt_hci_rp_vs_write_tx_power_level *)
        rsp->data)->status : 0;
    printk("Set Tx power err: %d reason 0x%02x\n", err, reason);
    return;
  }

  rp = (void *)rsp->data;
  printk("TX Power: %d\n", rp->selected_tx_power);

  net_buf_unref(rsp);
}

The code bt_hci_cmd_send_sync is where the hard fault occurs. The hard fault is as follows:

Full Description: Indicates that a fault with configurable priority has been escalated to a HardFault exception, because it could not be made active, because of priority or because it was disabled

I don't understand why this is not working since it has no problem being set and read on the peripheral device.

Any other areas to look at?

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Thu, 27 Jul 2023 08:14:24 GMT

Hi Trent,

Could you try again with CONFIG_BT_CTLR_ADVANCED_FEATURES=y removed ? It's not needed when using our controller.

It's important that BT_LL_SW_SPLIT should not be set =y because then you will use Zephyr's controller, not our controller.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

TrentS — Wed, 26 Jul 2023 22:33:53 GMT

Hi Hung,

Thanks for the reply. Sorry for the delay in getting back to you.

I have look at the link in the ticket and I had already implemented these functions in both the central and peripheral devices.

The reason for the CONFIG_BT_LL_SW_SPLIT=y is as follows:

Your example zip file (from above link) found here: /cfs-file/__key/communityserver-discussions-components-files/4/peripheral_5F00_lbs_5F00_txpower.zip

Has the settings enabled as follows:

CONFIG_BT_CTLR_ADVANCED_FEATURES=y
CONFIG_BT_CTLR_TX_PWR_DYNAMIC_CONTROL=y

I have the same CONFIG settings however, I get the squiggles under the Advanced features item:

If I highlight it, I see:

So the compiler is asking me to enable the BT_LL_SW_SPLIT in order for the device to work properly. If I disable it, the processor resets when it tries to set or read the power function:

In terms of testing with Coded PHY, I did start this with the code that I had attached. It had worked without coded PHY.

RE: nrf52840 - CODED PHY and NUS Central and Peripheral not communicating. Transmit power setting issue.

Hung Bui — Fri, 21 Jul 2023 11:48:10 GMT

Hi Trent,

I would suggest to take a look at this ticket: How to set BLE tx power

There is also a link to a sample that I created that worked for me.

I'm not so sure why you use CONFIG_BT_LL_SW_SPLIT=y ? By using this you are not using Nordic Softdevice Controller, but Zephyr controller instead.

Please try to test without CODED PHY first and move to CODED PHY after you verify that it works.