MLX90393 DOA?

Qwiic System
1

Hi,

I’ve tried (2) MLX90393 sensors with a variety of boards including Arduino Uno and the ReadBoard Qwiic. Neither sensor works with the library mentioned in the hookup guide. I get spurious readings. I also checked to see if the sensor address was correct - it shows as 0x0C.

When I run the example debug helper sketch, I get some disturbing results as shown below. I get the same for both sensors.

(C) Udo Klein 2018

License: GPL v3

burst mode: 1

woc mode: 1

sm mode: 1

error: 1

single bit error detected: 1

rs: 1

d1: 1

d0: 1

0x00 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x01 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x02 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x03 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x04 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x05 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x06 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x07 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x08 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

0x09 9415 10010100 00010101 read failure 111 1 1111 (ERROR) burst mode, woc mode, sm mode, error, single bit error, rs, d1, d0

I’ve also tried removing the pull-up resistors and adding a MicroOLED in the daisy chain. No luck.

Do I have dead sensors? Should I request a return or is there some other arcane knowledge I do not yet possess to help get these working?

Thanks

2

Give the example sketches found in the “Software” folder [in this repo a try and see if those work for you.](https://github.com/sparkfun/Qwiic_Magnetometer_MLX90393/archive/master.zip)

3

Nope. Values read are all over the place.

4

OK, that sounds like a bad part.

If you purchased the board directly off our website, just [fill in the form on this page with your order number and the URL to this forum post and we can replace it for you.

If you purchased from a distributor or Amazon you will need to let them know you have a defective product and arrange with them for an exchange.](Return Policy - SparkFun Electronics)

5

I solved the problems. There were several.

First, the debug sketch included with the library is missing wire.begin(). Add that and whammo! The modules appear like magic. :stuck_out_tongue:

As for the wild readings? One of the modules worked only after I turned up the gain. The other when I lowered the gain. Weird. But they work now. Calibration was interesting as each had a different range of compensation for all 3 axis. That part might be normal.

So, I’m good now. I might file a ticket on github for that missing statement.

6

One more observation. I was using the MLX90393 in a building with a tin roof. Once taken out of that environment, the gain settings were not as severe. I should’ve thought of all that metal above me. :roll: Lesson learned: go outside and play!