VL6180X unstable range results

[John E KVAM]

Hi,

Readings from my VL6180X sensor are very unstable. They vary about +-8 mm from measure to measure.

I tried with Nucleo board and official api, and also with Arduino board and Sparkfun code. 2,8V power supply is from linear regulator, measured object is solid white at distance 50 mm.

I tried to supply it from battery instead of usb, nothing helps.

[John E KVAM]

There might be a couple of things going on here.

The accuracy spec for the VL6180X is given by a graph, and it looks like in might vary by 3 or 4mm, but +/-8 seems too high.

1) Ambient light can be an issue. Did you range with any sunlight in the room? (Ambient can be returned by the sensor. It will affect the measurements.)

2) Surface texture. Is your measured object a mirror-like finish? One expects the best measurements from a Lambertian (non-mirror like or matte) surface texture.

3) Did you get a RangeStatus of 0? if there is someting odd going on, the RangeStatus will give you an idea of what it is.

Can you include a picture of you setup?

[John E KVAM]

From Jan:

Thank you for answer, i'm sorry for choosing wrong forum category (i didn't see tof category at all).

1. I tried sensor with daily ambient light and also i closed sensor and object in black box. There was no differences.
2. White phone charger, shiny plastic, 50 mm distance.
3. No, closing object to sensor at < 10 mm readings was always 10-20 mm.

Of course, my 8 mm error is not offset error. Rather a noise. With stable sensor and object (no movement) differences in continuous measure between min and max are about 16-20 mm.

Finally i will use cover glass and object will be matte white. I will check sensor with target setup. Do you think, that i can reach 2, 3 mm of error with it? Or maybe have to leave this project?

[John E KVAM · Accepted solution]

In the datasheet there is a picture - look for the word 'accuraccy'.

It shows a small hook at the very close distances. I explain this error as an echo. We want the light to echo once, but at the very near distances it echo's several times and basically increases the reported distance.

I suggest to everyone that they move the sensor back inside some sort of housing so the target can never get closer than 1cm.

That geneally solves the problem.