What sensor or combination of sensors would provide the best accuracy in a range of 1mm - 1100mm?

[GPoly.1]

I have read the data sheets for the VL53L0X, VL53L4CD and VL53L1CX ToF senors. I am unsure which one would be the most accurate as the datasheets present the accuracy data at different conditions.

Ideally the sensor would be accurate over 1mm - 1100mm, but for my purposes shorter range accuracy would be on a higher priority.

I have not used ToF sensors before, so I hope I am asking the right questions to better understand this technology.

[Anne BIGOT]

Thank you for your interest on our products.

Let me know if you can find the answer to your question in this article "how to choose a ToF" which is here.

[GPoly.1]

Thanks Anne,

I have read though the page that you linked.

Reading the data sheets for the sensors I am unsure which would be ideal for short range. I understand that the VL6180CX is meant for short distance, but I have also read that the VL53L4CD can be accurate in shorter ranges as well.

If you could provide some more information that would help me find a high accuracy sensor for a short range that would be greatly appreciated.

[John E KVAM]

If you need something beyond 50cm then the VL6180 is out.

The VL53L4CD is one of the newest Sensors, and it's also our least expensive.

(And it used invisible 940nm light.)

I'd buy the P-Nucleo-53L4A1 and give it a try. For $56 dollars you can prove it's accurate enough for your application. If it passes that test, you are all set.

The L4CD has a 'bigger brother' the VL53L4CX which is generally for longer distances, and it's more difficult to use. But if you buget allows for it, you might squeek out just a bit more acccuracy.

(Main difference is the L4CX uses Histograms, which is a definitely better at longer distances and maybe just a little more accurate at the short distances. But of course it comes with a driver that is more difficutlt to use, larger, and you lose the ability for the sensor to do all the work. (It's also more expensive.)

So try the L4CD. I'm betting you will be quite satified.

• john
•  

[GPoly.1]

Thank you John for the reply.

I will purchase the P-Nucleo-53L4A1 and run some tests to observe if the sensor will be suitable for my application.

I appreciate the help.

[John E KVAM]

I had a guy have trouble buying the P-Nucleo-53L4A1. If you can't find it buy the parts separately.

The X-Nucleo-53L4A1 is the sensor part and the Nucleo-F401RE is the MCU part.

Buying them separately is EXACTLY the same thing. It even saves you a few cents.

Good luck,

• john

[GPoly.1]

Thanks John for the information.

If I can not find anything I will buy the components separately.

I have a Nucleo-F411RE MCU, will this work as well?

[John E KVAM]

Unfortunately no.

In order to test the system, you need to download the Graphical User Interface (GUI) code from ST.com. And that GUI was written for the F401RE and, alas, the F411RE is a different MCU. It won't work.

For sure when you begin writing code the F411RE is a fine choice, but it won't run the GUI code.

But the F401RE is pretty cheap.

[GPoly.1]

Ok, good to know.

Thanks for the help John.

[GPoly.1]

In addition, I would like to conduct some tests with the ToF sensor (VL53L4CD) on reading the distance of objects as they pass by. For my application, it will be important to know how accurate the distance readings are when the ToF sensor is moving.

Do you know any limitations of the sensor in this type of application?

Is there a speed at which the ToF sensor does not provide an accurate reading?

[John E KVAM]

The accuracy of the sensor depends quite a bit on the integration time. We need time to gather enough photon strikes to get a statistically large sample. Acutally the photons are fast enough but we have to re-charge the SPAD element after each strike.

So a reaonably accurate reading takes 20ms. A better one would use 30ms. But if you are moving rapidly this would be a problem.

This is a sligthly different problem then if the object was moving an the sensor was stationary. In this case, the answer you get would have been exactly right at some point durning the collect. But not at the very begining or end of the integration time.

And of course lateral motion has it own set of issues.

All of this becomes clear when you get the evaluation kit and play with it.

• john