stm32duino / vl53l1 Goto Github PK

I am using this library to control 4 VL53L1 devices. For this I defined 4 objects of the class VL53L1. Unfortunately each object uses 9kB dynamic memory. This meets my expectations from the information given in the user manual UM2133 (8kB).

I am compiling with xtensa-lx106-elf-g++ (GCC) 4.8.2 and -Os.
I defined VL53L1_MAX_RANGE_RESULTS to 1.

These 4 objects use nearly half of the RAM available on the ESP8266 I use.

Do you have any recommendations on how to safe some memory? I only need simple range measurement data from each device.
Is it a good idea to use only one object and switch between the devices? What are the drawbacks?

Using the current version (bdd6387) of the VL53L1_Sat_HelloWorld.ino example I do not get any data from the sensor unless I add an additional clear interrupt.

The program execution is stuck in this loop:

do
{
  status = sensor_vl53l1_sat.VL53L1_GetMeasurementDataReady(&NewDataReady);
} while (!NewDataReady);

status is always 0 and !NewDataReady == true.

But if instead I modify the end of the setup procedure as follows:

-   sensor_vl53l1_sat.VL53L1_StartMeasurement();
+   sensor_vl53l1_sat.VL53L1_ClearInterruptAndStartMeasurement();

Then everything works as expected: I get correct data.

But according to the user manual UM2133 rev 6 fig. 6 the original procedure in the code is correct:

• StartMeasurement in the setup phase
• ClearInterruptAndStartMeasurement at the end of the loop

So I wonder, where is the error?

I have this issue only with an ESP8266 (NodeMCU board). I also tested it with a Nucleo Board:

In order to use it with the NodeMCU I obviously had to use a different pin for XSDN_I.

I know that the VL53L1_Sat_HelloWorld.ino is designed for a Nucleo Board. But due to the abstraction by the Arduino Library I would have expected to work for both boards.