We would like to use the VL6180 sensor in an underwater robot and we would like to know some details about the sensor in order to find the most suitable configuration for our needs.

[jmont.2b]

Is it possible to adjust the LED current ? (increase the current a little bit to compensate for water absorption).

Could you give us more details about the WAF? because we've observed a strange behavior during underwater measurements (for targets with a low reflectance).

Could you give us some recommendations about the cross talk calibration procedure because the one indicated in the documentation is not applicable in water (400mm with a black target does not work; the absorption at 850 nm is high in water).

[John E KVAM]

Water is interesting. I absorbs the light from out 940nm VL53 sensors. It doesn't absorb a ton of it, but it's measureable. The 850nm VL6180 however is unaffected by this absorbtion loss.

What does happen in the light is slowed down by a factor of 1.3, but it's just straight math.

We build the chip to absolutely insure that it will never output a non-eyesafe amount of energy. So there is no way to increase the VCSEL power.

The VL6180 continues to range until it gets enough photons to return a distance, or it runs out of time. (You supply the time.)

We spec the max ranging distance, but are careful to state what the reflectivity of the target is. (Black is the worst case, as few photons reflect off it.)

If you are having trouble seeing a black target at 400mm, I'd switch to the VL53L4CD. It does use a 940nm light (with absorbtion loss) but it has a range of 1.3meters in open air and you should be able to detect the black target.

If you invest $56 dollars for the P-NUCLEO-53L4A1 and it doesn't work for you, try the VL53L4CX. This chip can see 6M in the best conditions, and so in water, against black it should still do quite a ways. It's eval kit is the P-Nucleo-53L4A2. That last digit is important.

But to save you some $$ buy the P-Nucleo-53L4A2. It can run the L4CX and the L4CD software. Then evaluate both. If the L4CD works for you do that. The sensor is less expensive and the code is much simpler. (The VL53L4CX is a more capable VL53L4CD. it can be used to evaluate both.)

[jmont.2b]

Hi John,

thank you for your reply.

You say that 850nm is unaffected by water absorption but we did some research and we found these following absorption coefficient (cm-1) for water:

• more or less 0.02 for 850nm
• more or less 0,06 for 940nm

Using Beer-Lambert-Bouguer equaiton (exponential) the difference of absorption between the two wavelength is significant. For this reason we've decided to use 850nm but we will do some tests with VL53L4CX.

Could you give us more details about receiver sensitivity and emitter power? (for VL6180 and VL53L4CX)

If we have minimum sensitivity and emitter power (for each sensor), maybe we could do a bit more theoretical study.

In our case the robot work in the dark (closed environment), so maybe we can adjust the SNR value to detect weaker signals.

What do you think about the WAF ? Is it better to disable it in our case?