nRF52833 + external 32 kHz TCXO: soft reset hangs

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Tue, 09 Sep 2025 17:23:41 GMT

Thank you for re-testing this again also without the WDT active. I will add this to my bug report.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Tue, 09 Sep 2025 15:38:52 GMT

I did have the watchdog peripheral enabled in both mcuboot and the main app. I've tried disabling it in both and seems like the behaviour is the same from what I can tell. I still require the start_lfclk sys_init for it to work properly.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Tue, 09 Sep 2025 14:53:58 GMT

Thank you. One more follow up question, are you using the Watchdog peripheral in your application? The watchdog will automatically request the RCOSC if active on startup and it may potentially be relevant to the issue.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Tue, 09 Sep 2025 12:59:44 GMT

Great thanks a lot. If you need me to provide more info for debugging, then just let me know.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Tue, 09 Sep 2025 12:57:42 GMT

Thank you for the update. I’m happy to hear that you’ve found a temporary workaround. I think this issue should be investigated further on our end, so I’ll report it as a bug internally. I’ll update the thread as soon as I know more.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Tue, 09 Sep 2025 12:43:06 GMT

I have tried enabling that config and it had no effect unfortunately.

The only way I have gotten it to work reliably is by adding this:

int start_lfclk() {
	nrfx_clock_lfclk_stop();
	NRF_CLOCK->LFCLKSRC = (CLOCK_LFCLKSRC_SRC_Xtal << CLOCK_LFCLKSRC_SRC_Pos) | (CLOCK_LFCLKSRC_BYPASS_Enabled << CLOCK_LFCLKSRC_BYPASS_Pos) | (CLOCK_LFCLKSRC_EXTERNAL_Enabled << CLOCK_LFCLKSRC_EXTERNAL_Pos);
	nrfx_clock_lfclk_start();
	return 0;
}
SYS_INIT(start_lfclk, POST_KERNEL, 0);

to one of our custom drivers that is included in both the main image and mcuboot. This is good enough to allow us to continue development but it would be good to find a better solution.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Tue, 09 Sep 2025 11:49:38 GMT

I have asked around internally but I haven’t found anyone who has used a TCXO and experienced this problem. Could you please try enabling CONFIG_SYSTEM_CLOCK_WAIT_FOR_AVAILABILITY in both bootloader and app to see if that might help? From the clock_control_nrf.c implementation it seems like this should ensure that the LF clock start task is triggered. I will continue checking internally in the meantime.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 22:04:22 GMT

So i managed to make some progress.

From what I can tell the LFCLK is not turned off when it resets or when mcuboot passes execution off to the primary image (can you confirm?), and if it's already running from an external source, it does not get configured correctly. So it's able to go from RC->TCXO, but not TCXO->RC or TCXO->TCXO. This would explain the behaviour in the OP.

When it's in this state where it's not configured correctly, i can attach with a debugger and read out the LFCLKSTAT and LFCLKSRC registers:

(gdb) x/wx 0x40000518
0x40000518:     0x00000000
(gdb) x/wx 0x4000041c
0x4000041c:     0x00000000
(gdb) x/wx 0x40000418
0x40000418:     0x00010001

so LFCLKSRC and LFCLKSRCCOPY is configured to use the internal RC while LFCLKSTAT is configured to use the xtal.

I can set LFCLKSRC manually, but this then switches LFCLKSTAT to use the RC oscillator:

(gdb) set {int}0x40000518 = 0x00030001
(gdb) x/wx 0x40000518
0x40000518:     0x00030001
(gdb) x/wx 0x40000418
0x40000518:     0x00010000

BUT now if I let execution continue, my app boots normally, but with the RC clock.

Please advise on what I should do. Is this a bug in clock_control_nrf.c?

I can likely stop the LFCLK in my application and configure it from scratch but this doesn't solve the case where it reboots and enters MCUBOOT while running off the TCXO.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 17:45:01 GMT

Never mind.. it actually appears like it wasn't using the TCXO at all when it was booting successfully. It must have fallen back to the internal RC since the drift I measured was way way too high to be the TCXO, even though it was configured with EXT_FULL_SWING.

EDIT: confirmed, i printed the LFCLKSRC register. if mcuboot is configured with CONFIG_CLOCK_CONTROL_NRF_K32SRC_EXT_FULL_SWING=y, this register is 0, meaning it's using the internal RC.

If mcuboot is using the internal RC, then this register is 0x00030001 meaning it's using external TCXO.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 16:12:30 GMT

EDIT: this whole reply is inaccurate, see next reply

Okay I managed to get it to work reliably by swapping from k_sleep -> k_busy_wait in another driver init and by adding this function that explicitly starts the LF clock:

static int start_clock() {
	int err;
	struct onoff_manager *clk_mgr;
	struct onoff_client clk_cli;

	clk_mgr = z_nrf_clock_control_get_onoff(CLOCK_CONTROL_NRF_SUBSYS_LF);
	if (!clk_mgr) {
		LOG_ERR("Unable to get the Clock manager");
		return -ENXIO;
	}

	sys_notify_init_spinwait(&clk_cli.notify);

	err = onoff_request(clk_mgr, &clk_cli);
	if (err < 0) {
		LOG_ERR("Clock request failed: %d", err);
		return err;
	}

	int res;
	do {
		err = sys_notify_fetch_result(&clk_cli.notify, &res);
		// printk("err %d res %d\n", err, res);
		if (!err && res) {
			// LOG_ERR("Clock could not be started: %d", res);
			return res;
		}
	} while (err);

	return 0;
}
SYS_INIT(start_clock, POST_KERNEL, 0);

The strange part is that before updating that driver to use k_busy_wait, if I changed the log level in the module with the `start_clock` sys_init from LOG_LEVEL_WRN to LOG_LEVEL_DBG, it did boot normally.

I really don't like how unreliable this seems, especially sihce this is a device that we will definitely not be able to update in the field if they don't boot.

Is there a way to simply use the internal RC for both mcuboot and the primary image, but then switch to the external TCXO at runtime? that way if it doesn't work (or if the component breaks/falls off), we can always fall back to the internal RC.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Fri, 05 Sep 2025 15:06:26 GMT

Yes, the scheduler relies on the lfclk, so I agree, this indicates that the clock is not running. Not sure what could be causing this yet.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 14:37:15 GMT

I managed to find the sys init that caused the issue. It's in a custom driver init that hangs when it calls 'k_sleep'.

When I put `k_msleep` in my test_init function, it hangs there too. I believe that zephyr uses the LFCLK for its scheduling, so it would seem that the LFCLK is not actually started?

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Fri, 05 Sep 2025 14:24:46 GMT

I don't know how they are making this list, I would have to ask them. I don't immediately see any ways to check this manually. However, you can try to to search for this priority in the .config file for your app(build/<app name>/zephyr/.config)

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 14:17:44 GMT

I'm not using vscode unfortunately. do you what command line tool gives access to that info?

I can maybe try to get vscode working for this project.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Fri, 05 Sep 2025 13:54:34 GMT

Interesting. You can view the list of init functions here if you're using VS code:

[quote userid="124777" url="~/f/nordic-q-a/124183/nrf52833-external-32-khz-tcxo-soft-reset-hangs/547929"]CONFIG_SYSTEM_CLOCK_NO_WAIT=y had no effect, i'm assuming this is just skipping that spinwait function?[/quote]

Correct.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 13:48:35 GMT

I have confirmed that this breakpoint is hit:

static int test_init() {
	return 0;
}
SYS_INIT(test_init, POST_KERNEL, 79);

but this one never is:

static int test_init() {
	return 0;
}
SYS_INIT(test_init, POST_KERNEL, 80);

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 13:43:11 GMT

Small update:

I have confirmed that I do hit breakpoints in PRE_KERNEL_X SYS_INIT's and in POST_KERNEL SYS_INIT with priority 0, *but not* priority 99.

Currently narrowing down exactly which sys_init is causing the issue.

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

alexdr5398 — Fri, 05 Sep 2025 13:26:26 GMT

Hi Vidar,

My mistake. I forgot to mention in the original post that I did trying placing a breakpoint at the end of that function (clock_control_nrf.c:540, right after the 'clock_is_running' loop), and it does indeed hit that breakpoint, but then any time I break after that the kernel is just in 'k_cpu_idle' with the thread entry pointed at the idle thread:

CONFIG_SYSTEM_CLOCK_NO_WAIT=y had no effect, i'm assuming this is just skipping that spinwait function?

Thanks for the reply,
Alex

RE: nRF52833 + external 32 kHz TCXO: soft reset hangs

Vidar Berg — Fri, 05 Sep 2025 13:15:02 GMT

Hello,

I don't have a TCXO to test with here so I can't easily reproduce this, but I assume the reason it hangs is that the LF clock started event is not triggered causing the program to never exit clock_control_nrf.c->lfclk_spinwait() function. You may be able to confirm this by looking at the call stacks.

[quote user=""]If I switch mcuboot to also use EXT_FULL_SWING, then after reset I see the UART log:[/quote]

As a test, please try and see if you get the same result if you build the app with CONFIG_SYSTEM_CLOCK_NO_WAIT=y.

Best regards,

Vidar