I2C Service not running

[ErikLundberg]

I had a BMP280 sensor, connected to a MacArthur HAT, and it was operating just fine for a few weeks, but now it's not.  

I replaced it with a BME280, and am having the same problem.   I also have an IMU module installed, and oddly, it continues to work fine (I can see heading data in a KIP dashboard, which reacts as expected when I rotate the PI.)

The Check System app reports this:

     Checking I2C sensors... | I2C enabled | Access to Signal K server validated
        ↳ service not running
            Error processing BMP280: No I2C device at address:  0x76
            Error processing BME280:  Failed to find BME280! Chip ID 0x0

I am running the latest version of everything.  (The Settings app shows no update candidates, after clicking the Refresh button, and I've been installing all Debian updates as they come along.)

I tried uninstalling the I2C app and reinstalling (4.2.0-stable).  The problem remains. 

I tried disabling the I2C interface in the Raspberry configuration, then re-enabled it (and rebooted), and still the problem remains.

I am running on a Pi 4, 4gb, with a MacArthur HAT, with both IMU & power modules installed
OpenPlotter 4.x.x, 64 bit

I tried getting a debug log, but could not figure out how to run the "Check system" app from a terminal window.  For instance...
     bash:  openplotter-check\ system
did not work.

When I tried running the I2C app from a terminal window, it just opened the app, but I saw no debug output anywhere.

I welcome any troubleshooting ideas!

Erik

[Sailoog]

This does not look like a bug, it looks like your sensors are damaged or misconfigured.

Why are you getting 2 errors? are both sensors connected and configured? try just one.

Make sure the sensor uses the address your are configuring.

what "I2C Addresses" says?

[ErikLundberg]

This does not look like a bug, it looks like your sensors are damaged or misconfigured.
Why are you getting 2 errors? are both sensors connected and configured? try just one.
Make sure the sensor uses the address your are configuring.
what "I2C Addresses" says?

Here is what I2C Addresses shows:

    0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:                        -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- 68 -- -- -- -- -- -- --
70: -- -- -- -- -- -- 76 --                       

I started with a single BMP280, which has address 0x76.  It worked fine for a few weeks, then simply stopped sending data, with no obvious trigger.  Yesterday, I removed the BMP280 and connected a new BME280 (just received it), which shows up with address 0x68.   So to answer your question, just one sensor is attached, and I've never had two attached at the same time.  

In the I2C app, all three sensors on the BME280 (temp, pressure, humidity) showed up in the Sensors tab, but not in Signal K Data Browser.  In fact, the Sensors display still shows the BMP280, even though it is not connected.  I've attached a screenshot, taken with only the BME280 connected.

Coincidentally, I have experienced something similar with 1-Wire sensors.  I had some DS18B20 sensors connected that worked fine, then simply stopped, again for no obvious reason.

I understand these issues may be due to bad sensors, but I bought them through Amazon from sellers with good ratings.  Or maybe I've damaged the I2C sensors when I solder on the headers.  (That doesn't explain the behavior of the 1-Wire sensors, though.)  I have a few more that I can experiment with.  But I'm wondering if there may be a hardware fault on the MacArthur HAT.   (I have checked my wiring pretty carefully, and it seems good.)


00:                         -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- 68 -- -- -- -- -- -- --
70: -- -- -- -- -- -- 76 --                         
00:                         -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- 68 -- -- -- -- -- -- --
70: -- -- -- -- -- -- 76 --                         

---

I agree that it may not be a software bug, but still seems a bit suspicious. Please let me know if it would be more appropriate to post this issue in one of the MacArthur HAT forums. -Erik