VL53L5CX API Size "vl53l5cx_buffers.h" seems to be too large for the MCU to handle. Is splitting the buffers ( "VL53L5CX_FIRMWARE[]" )into smaller 'chunks' to send to the SPAD a reasonable solution?

[DAzu.1]

I am trying to use the VL53L5CX with an AVR128DB48. 

The .h file "vl53l5cx_buffers.h" seems to be too large for the MCU to handle. 

I think, as it stands, it breaks the 32727 byte limit that AVR is capable of. 

Is splitting "VL53L5CX_FIRMWARE[]" into smaller 'chunks' to send to the SPAD a reasonable solution? 

Has anyone else encountered this problem?  

Is there an alternative that I have missed that is more compatible with an AVR MCU?

Is there any reason that this will not work with the VL53L5CX? 

[John E KVAM]

Absolutely acceptable. Lots of MCUs need to do this with various max lengths.

You need to modify it in the platform.c file.

So just go ahead and break up the transfer into whatever size you need.

It will work fine.

• john

[DAzu.1]

Great, thank you.

Do you mean vl53l5cx_api.c?

I can't see anything in platform.c that relates directly to writing the firmware, apart from the I2C read/write functions.

In vl53l5cx_api.c the firmware is written using the following:

/* Download FW into VL53L5 */
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x09);
	status |= WrMulti(&(p_dev->platform),0,
		(uint8_t*)&VL53L5CX_FIRMWARE[0],0x8000);
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x0a);
	status |= WrMulti(&(p_dev->platform),0,
		(uint8_t*)&VL53L5CX_FIRMWARE[0x8000],0x8000);
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x0b);
	status |= WrMulti(&(p_dev->platform),0,
		(uint8_t*)&VL53L5CX_FIRMWARE[0x10000],0x5000);
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x01);

I have modified this to to account for the limited array size in my MCU. Splitting 'VL53L5CX_FIRMWARE[]' into 5 sections. VL53L5CX_FIRMWARE[]1 ...VL53L5CX_FIRMWARE[]4.

status |= WrByte(&(p_dev->platform), 0x7fff, 0x09);
    
	status |= WrMulti(&(p_dev->platform),0,
		(uint8_t*)&VL53L5CX_FIRMWARE[0],0x4000);
    
    status |= WrMulti(&(p_dev->platform),0x4000,
		(uint8_t*)&VL53L5CX_FIRMWARE1[0],0x4000);
    
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x0a);
    
	status |= WrMulti(&(p_dev->platform),0,
		(uint8_t*)&VL53L5CX_FIRMWARE2[0],0x4000);
    
    status |= WrMulti(&(p_dev->platform),0x4000,
		(uint8_t*)&VL53L5CX_FIRMWARE3[0],0x4000);
    
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x0b);
    
	status |= WrMulti(&(p_dev->platform),0,
		(uint8_t*)&VL53L5CX_FIRMWARE4[0],0x5000);
    
	status |= WrByte(&(p_dev->platform), 0x7fff, 0x01);

I have already bricked one module testing my code. Because there is no documentation on the register structure in the SPAD, I would like to double check with you that this would work.

Is this correct?

Kind regards

[DAzu.1]

@John E KVAM​ 

[John E KVAM]

I think that approach is going to get in into trouble. A much more general approach would be to put the change in the platform.c file.

The WrMulti() and RdMulti() functions were intended to be implemented in the platform.c file which we conveniently left empty for this purpose.

Rather than change a lot of places, just make these functions more useful.

if the size is greater than MAXSIZE,

While sent_size >0 {Send MAXSIZE bytes; sent_size -=MAXSIZE}

And don't forget to do the same thing for the RdMulti.

I don't understand how you could 'brick' the sensor by issuing poorly constructed I2C commands. I mess them up all the time and have never ruined a chip that way.

(Although I must admit to ruining a lot of sensors in other creative ways.)

• john