nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Fri, 13 Feb 2026 13:48:32 GMT

Hi Håkon,

Could you please try using the BME688 and use Thingy53's Qwicc connector with VCC, SDA/SCL and GND with the nRF54L15-DK. Thank you.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Thu, 12 Feb 2026 19:13:50 GMT

Hi Håkon,

Thanks. You’ve used the Thingy53’s BME688 whereas I’m using a BME688 directly connected to the nRF54L15 and external 10KOhm pull-ups.

It should not be such an issue to hookup a sensor to a nRF54L15-DK. The nRF5340-DK and nRF9151-DK have had no issues with the same sensors and leads. Also, no issues with using the same setup with other MCUs.

I’ll try using the Thingy53 BME688 via TP4 and TP5, as per your test.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

Håkon Alseth — Thu, 12 Feb 2026 14:24:59 GMT

Hi,

[quote user="zpm1066"]How did you connect the Thingy53 to the nRF54L15-DK? [/quote]

TP4 and TP5, ie. SDA/SCL of the thingy53.

And GND between the two boards.

[quote user="zpm1066"]What VDD voltage setting did you have for the nRF54L15-DK when using the Board Configurator? Thanks.[/quote]

I setup the kit with 3.3V.

Kind regards,

Håkon

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Thu, 12 Feb 2026 13:24:20 GMT

HI Håkon,

How did you connect the Thingy53 to the nRF54L15-DK? I used the 3.3V from the Qwicc connector with 10KOhm external pull-ups.

What VDD voltage setting did you have for the nRF54L15-DK when using the Board Configurator? Thanks.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Thu, 12 Feb 2026 13:21:29 GMT

Thanks. Agree regarding the need to debug the issue with a oscilloscope. Unfortunately, I don't have access to store one yet.

I've been using the VDDIO on the nRF54L15. with 0K external pull-ups. Also, tried using the Thingy53's 3.3V from the Quick connector.

The nRF54L155-DK does seem to be more sensitive than the other devices. I've had no issues at all when using the BME688/BME280 with nRF5340-DK, nRF9151-DK, various nRF52840 boards or the ESP32/Raspberry Picos.

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

PaKa — Thu, 12 Feb 2026 08:06:50 GMT

My setup is direct connection so only one Power rail. Håkon used the Thingy:53. Key thing is bypass the level shifters and have same VDD on the SoC and the sensor. This does work so if it fails on your side it is a connection issue so check that you have the external pull-ups and no additional load on the lines. And not too long lines on the I2C as that increases the capacitive load.

An oscilloscope is the tool to debug this.

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Tue, 10 Feb 2026 17:57:36 GMT

Hi PaKa

Thanks. What VDD voltage setting do you have for the nRF54L15-DK when using the Board Configurator?

I tried taking the 3.3V from the Thingy53's Qwicc connector with common ground to nRF54L15-DK and used the Board Configurator to set VDD to 3.3V on the nRF54L15. 10kOhm pull-ups used and "bias-disable" set in the pinctrl of the nrf54l15dk_nrf54l15_cpuapp_ns.overlay. The BME688 still failed unless VDD was reduced to 2.4V.

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

Håkon Alseth — Tue, 10 Feb 2026 15:22:39 GMT

Hi,

Here's a quick setup, with i2c shell, with nRF54L15-dk connected to the SDA/SCL of a thingy53 (3.3V):

uart:~$ i2c scan i2c@c8000
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:             -- -- -- -- -- -- -- -- -- -- -- -- 
10: 10 -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
60: 60 -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
70: -- -- -- -- -- -- 76 --                         
3 devices found on i2c@c8000

It was not a problem to detect the address of the bme-sensor.

[quote user="zpm1066"]A scope access is unlikely for a while.[/quote]

I would strongly recommend that you obtain these tools for development, such as logic analyzer and/or oscilloscope.

Kind regards,

Håkon

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

PaKa — Tue, 10 Feb 2026 15:19:27 GMT

Can confirm the nRF54L15 works without problems with the BME688 at 3.0V when directly connected and with 10K external pull-ups on the datalines. Most likely issue is the supply/signals on your setup.

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Tue, 10 Feb 2026 13:36:41 GMT

Hi Håkon,

I tried connecting the BME688 directly to the nRF54L15-DK. Same result, only works up to 2.4V. I originally started with a direct connection of the sensor to the board.

A scope access is unlikely for a while. Is there anyway that you could try reproducing this issue with a BME280 or a BME688 internally? Thanks.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

Håkon Alseth — Tue, 10 Feb 2026 09:13:18 GMT

Hi,

Try connecting the BME sensor directly to the DK, and omit the interconnect board and ssd-display, to isolate the issue.

[quote user="zpm1066"]I do have the Power Profiler II, so could try using the logic connections to SDA/SCL.
[/quote]

You should get a oscilloscope to debug this.

Kind regards,

Håkon

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Mon, 09 Feb 2026 16:09:49 GMT

Hi Håkon,

Unfortunately, I don't have a scope to review the I2C signals.

I think we can safely ignore the Pimoroni BME688 / BME280 sensors. I've tested the same sensors with a nRF5340-DK. The BME688 sensor and SSD1306 work at both I2C_BITRATE_FAST and I2C_BITRATE_STANDARD. The default "nrf5340_cpuapp_common-pinctrl.dtsi" was used for the nRF5340-DK.

The exact the same hookup as the nRF54L15-DK was used with the exception of using the nRF5340-DK's I2C1 at P1.0.3 (SCL) & P1.02 (SDA).

I suspect that there may be an issue with the nRF541L15-DK board. I don't have another nRF54L15-DK but may order one.

I do have the Power Profiler II, so could try using the logic connections to SDA/SCL.

Here are the results with a nRF5340-DK, BME688 and SSD1306.

&i2c1 {
	compatible = "nordic,nrf-twim";
	status = "okay";
	clock-frequency = <I2C_BITRATE_FAST>;
	zephyr,concat-buf-size = <4096>;
	pinctrl-0 = <&i2c1_default>;
	pinctrl-1 = <&i2c1_sleep>;
    pinctrl-names = "default", "sleep";
	ssd1306: ssd1306@3c {
		compatible = "solomon,ssd1306fb";
		reg = <0x3c>;
		width = <128>;
		height = <64>;
		segment-offset = <0>;
		page-offset = <0>;
		display-offset = <0>;
		multiplex-ratio = <63>;
		segment-remap;
		com-invdir;
		prechargep = <0x22>;
	};
	
	bme688: bme688@76 {
		compatible = "bosch,bme680";
		reg = <0x76>;
	};
};

nRF5340-DK hookup to BME688 & SSD1306

Console output

*** Booting nRF Connect SDK v3.2.1-d8887f6f32df ***
*** Using Zephyr OS v4.2.99-ec78104f1569 ***
[00:00:00.267,303] <inf> app: Board target: nrf5340dk/nrf5340/cpuapp

[00:00:00.267,303] <inf> app: bme688x_iaq_ssd1306 - App started
[00:00:00.267,303] <inf> app: Init ssd1306
[00:00:00.267,486] <dbg> cfb: cfb_framebuffer_init: number of fonts 2
[00:00:00.295,043] <inf> oled: Index[0] font dimensions  5x8
[00:00:00.295,074] <inf> oled: Index[1] font dimensions  8x8
[00:00:00.295,104] <inf> oled: Selected font: index[0]
[00:00:00.295,104] <inf> oled: x_res 128, y_res 64, ppt 8, rows 8, cols 128
uart:~$ 
uart:~$ i2c scan i2c1
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:             -- -- -- -- -- -- -- -- -- -- -- -- 
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
30: -- -- -- -- -- -- -- -- -- -- -- -- 3c -- -- -- 
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
70: -- -- -- -- -- -- 76 --                         
2 devices found on i2c1

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

Håkon Alseth — Mon, 09 Feb 2026 10:12:56 GMT

Hi,

Can you share the logic trace when trying to communicate?

Please share both working and non-working scenario.

Kind regards,

Håkon

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Fri, 06 Feb 2026 19:11:33 GMT

Hi Håkon,

Actually the BME280 and BME688 sensors we're using are from Pimoroni.

We've been using these for several years on a number of devices, including nRF9151-DK, nRF5340-DK, nRF52840, ESP32, and Raspberry Pi Picos without any issues at voltages up to 3.3V and at both I2C_BITRATE_STANDARD / I2C_BITRATE_FAST settings.

It's only the nRF54L15-DK that seems to be problematic.

The Pimoroni BME280 and BME688 are rated at 2V-5V range, have I2C level shifting and 10KOhm pull-ups. So, don't quite understand why they don't work with the nRF54L15-DK at 3.3V. Problems start with voltages above 2.4V.

The Adafruit BME688 in your link above also provides the same 10KOhm pull-ups with level-shifting.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

Håkon Alseth — Fri, 06 Feb 2026 08:28:18 GMT

Hi,

Your sensor seems to be a module, more specifically this one:

https://shop.pimoroni.com/products/adafruit-bme688-temperature-humidity-pressure-and-gas-sensor-stemma-qt?variant=39358272798803

Which states:

For your convenience, Adafruit have pick-and-placed the sensor on a PCB with a 3.3V regulator and some level shifting, so it can be easily used with your favorite 3.3V or 5V microcontroller.

Although the IC accepts lower voltage, your module does not seem to do this.

Kind regards,

Håkon

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Fri, 06 Feb 2026 03:31:50 GMT

Hi Håkon,

Thank you for the quick response. Very much appreciated!

I tried using 100 KHz I2C (I2C_BITRATE_STANDARD) with the internal pull-up on the nRF GPIO. This behaves the same as previously. The BME688 sensors again work only at 2.4V or below but the SSD1306 works up to 3.3V.

Update:

With I2C_BITRATE_STANDARD and bias-pull-up disabled, I tried using 4.7kOhm on the SDA/SCL GPIOS to see if it makes any difference. Unfortunately, not much of a difference. The BME sensors were not stable at anything much higher than 1.8V. SSD1306 display was fine at 3.3V, same as with internal pull-up.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

zpm1066 — Thu, 05 Feb 2026 19:21:16 GMT

Hi Håkon,

Thank you for the quick response. Very much appreciated!

I tried using 100 KHz I2C (I2C_BITRATE_STANDARD) with the internal pull on the nRF GPIO. This behaves the same as previously. The BME688 sensors again work only at 2.4V or below but the SSD1306 works up to 3.3V.

It is somewhat strange that at I2C_BITRATE_FAST, the SSD1306 has no issue on I2C bus with the internal pull resistors.

As per your suggestion, I'll try 5kOhm resistors on the SDA/SCL GPIOS to see if it makes any difference.

Best Regards,
Ravi

RE: nRF54L15-DK I2C voltage issues with sensors failing to be detected as "ready" unless voltage reduced to below 3.3V

Håkon Alseth — Thu, 05 Feb 2026 10:14:18 GMT

Hi,

In order to run 400 kHz I2C, you will need a stronger pull resistor than the internal pull on the nRF gpio, which is nom. 13 kOhm.

You want 5kOhm or lower.

You can look at the SDA/SCL lines with an oscilloscope to confirm how the waveform looks.

Can you please try to add stronger pull-resistors, and see if the system starts to behave more stable?

Kind regards,

Håkon