VL53L1X offset when detaching cover glass

[mf101]

We are using the VL53L1X with the ULD API in our products. During the production process, we perform the offset calibration process as described in UM2510. Our cover glass (Luxycryl red 1600) is actually separated with a light blocker to prevent crosstalk and therefore we do not need to perform the crosstalk calibration.

After the offset calibration, the measured distance values are as expected. But for some strange reason, when detaching the cover glass the measurement values drop and we get an offset again. After restoring the original state by re-attaching the cover glass, the offset is gone.

Is there any potential explanation for this behaviour ?

[John E KVAM]

One thing to try - do the crosstalk calibration to see if your coverglass solution is as good as you think it is.

It could be that the plastic you use to make the light blocker is not as opaque as you think it might be.

(I've played with optics a lot, although I'm not an optical engineer. I've come to think of photons as really sneeky.)

Removing the coverglass might mean removing a lot of 0-distance photons. So your 'offest' would go up.

But you say your distances 'drop'.

Is there any part of your structure that the sensor is 'seeing' when you remove the coverglass? If something were in the FoV then you would get a lower number.

But that's all I can think of.

[mf101]

We are unable to perform the crosstalk calibration, because we cannot obtain the calibration distance. The measurement is too good, therefore we see no point of divergence. When detaching the cover glass the ranging results are very similar, but with the offset mentioned before. The sensor sees no structure, there is a large opening in the housing, into which the cover glass is fixed.

Like you say, an increased offset would be explainable by crosstalk. But in this case there must be another cause.

From my point of understanding the offset correction simply adds or subtracts the offset value from the ranging result. Is this true ?

[John E KVAM]

Yes there is a reason for it. I'm guessing crosstalk.

If there is a difference between the coverglass on and off, then you have crosstalk. It might not be much, but it's there.

the easy way to tell is to simply do the crosstalk - or the easy part anyway.

Find a dull gray target and start ranging. Is it good at 20cm? 40cm? and keep going.

(if you can't find anything else, normal cardboard is about 50% reflective. Use the unprinted side. But it would be best if you had a duller gray.)

Make sure the target is big enough to cover the entire field of view. ( diameter of the target >= 1/2 the distance to the target.)

If you find there is no degradation in distance accuracy, then your crosstalk is good.

But given that you see a difference with and without the glass, I'd expect some crosstalk.

You might see a difference in distaces if you change your target. There can be up to a 1cm delta between a very dull black and a sheet of aluminium foil.