So apparently the 1550nm LiDAR system of the new Volvo can damage certain camera sensors when aimed at it (the viral video link)
The obvious question: what cameras are affected and are greater distances safe? 
Should I worry about my camera at all?
So apparently the 1550nm LiDAR system of the new Volvo can damage certain camera sensors when aimed at it (the viral video link)
1 Like
All Lidar systems can.
2 Likes
Yeah, but what about the distances, is there any reason to worry about it standing farther away? All videos showing damage were filmed while standing at point blank pretty much.
2 Likes
i would not point my camera at any laser at and distance tbh.
Funny. Usually self-driving cars use different types of sensors to see their environment - including cameras. Does that mean that this Volvo damages other cars?
3 Likes
It is said that cameras behind windshield are mostly protected from 1550nm light, other cameras, might not. We’ll see how it plays out, the issue is pretty fresh at roughly 2 months old.
The only camera that suffered damage in the Reddit video was the telephoto though, but I was not able to find a source with an explanation that made sense to me.
2 Likes
i guess we will find out when the density of those cars goes up.
1 Like
Given how many videos we have of waymos rolling down the streets with absolutely no damaged phone sensors, you have to wonder how common this damage will be.
3 Likes
That’s what doesn’t make sense to me. Why is no one talking about it? Like the risk of damaging a camera is real, but how bad is it…
Probably not too bad given the silence, but should I worry about my super expensive camera when on the street, or is it only when literally coming close to the car and filming the sensor…?
2 Likes
I live in a city where Waymo has been around for several years (they trained their cars here, including in my neighborhood) and I’ve never heard about this. Obviously this is just anecdotal but people love to complain about any minor thing that Waymos do wrong and if that had happened here it would have been widely reported on. I assume the risk for damage isn’t actually that great.
4 Likes
The kind of damage that that is caused on the sensor are analyzed in detail in the paper:
https://doi.org/10.1364/OE.515728
The authors actually tested until damage occurred, and then inspected what broke.
This doesn’t answer how likely any damage would be in practice.
Harri
4 Likes
So, who pays for the damage? I would assume the car’s owner or operator.
Exactly, I’d love to see someone test it over different distances and angles relative to sensor
I would say not because the main feature of a laser is that the beam doesn’t spread much with distance.
1 Like
I would agree it is not a great article. It should come as no surprise that a few 10’s of mJ/cm2 for very short pulses causes damage to Si devices, even at photon energies below the bandgap of Silicon. There are a variety of materials on or near the surface, and impurity doping of Silicon that increases absorption.
Something that’s missing in the article (and other fuss I’ve read about) is what is typical for lidar pulse energy density in the real world (mostly automotive Lidar)? I would guess it’s well below the levels used in this test.
1 Like
From reading multiple articles, it looks like volvo change the wavelength on the lidar to a “safer” for human eyes but it allows them a higher energy. The increase in energy is to attempt to “see” farther away. Interesting stuff.
3 Likes
Eye-safe is a term applied to the use of 1550 nm Lidar, but as with almost any technical topic there are some details that engineers need to consider.
Safety questions about 1550 nm Lidar
I think the assumption is that 1550 nm is safe because it can be absorbed in the vitreous humor of the eye, thus not reaching the retina. But if intensity is high enough there could be corneal damage.
By beyond that I have no references for typical intensities and how that compares to damage thresholds for humans (animals) and for devices.
2 Likes
I tend to agree. E.g., there are videos on youtube taken in concerts with a laser light show and you can see in the video how the camera is damaged more each time the laser beam crosses over the camera. And in the videos that I have seen the camera is not close to the stage.
Well, what about the fact that it’s a visible spectrum laser compared to 1550nm infrared… Does that change anything, I wonder?
isnt the point more how much energy is focused onto parts of the sensor to damage that pixel?
4 Likes