VL53L4CX Latency

[Lutfullah1]

Hi,

I am conducting a rapid object detection test using the VL53L4CX sensor. I am moving an object towards the sensor at approximately 5 km/h and aiming to light an LED if the object is detected at around 60 cm to measure the sensor's response time. However, when checking the LED with a camera, it lights up at approximately 50 cm. I am trying to determine the cause of this delay. The MeasurementTimingBudget is set to 20 ms. When the object is stationary at 60 cm, the LED lights up as expected. Could the sensor be providing outdated data, or do you have any theories on the potential cause of this delay?

[John E KVAM · Accepted solution]

in that 20ms of integration time, there is a lot going on. About 5ms is just getting a measure of ambient light and adjusting for any temperature differences. Then there are actually two measurements. Call them A and B. The A measure has one pulse repetition interval, and the B has a different interval. This is to deal with something called radar aliasing. 

But aside from that, you are getting the AVERAGE distance of your return times. So at the start of your 20ms you have one distance and at the end you have a different distance. Setting aside the 5ms of setup, you also have to consider that closer objects return more photons. So in your experiment, if the object were at 50cm at the start of the 20ms interval and 40cm at the end, you might get something like 44cm.

I suggest that you switch to the VL53L4CX. It's a bit more complex, but it's faster. What happens is that we still do the A,B trick, but we report back to you after every step. So you get the result of A and then the result of the B range. Lower latency, same accuracy.

Good news is you can try it with your existing setup. Just change the software to the VL53L4CX software. It will work in your experiment. BUT PLEASE DONT go into production like this. The L4CX sensors are guaranteed to run the more demanding software. The L4CD chips are not. But for a test. I think you will be happier.