OPT8241: OPT8241-CDK-EVM: phase shifting of emitted light

Hi Tobias,

There are a few challenges to be considered here. The product isn't really meant to do this. So we have not ensured that the modulation output of 1 camera would be coherent with modulation output of another camera. They will have some non-zero correlation. But it will be very noisy as the PLL jitter of each camera would have to be factored into the amount of noise picked up by this cross-correlation. Moreover, there is no feature present inside the chip to dynamically assign quads as the camera is meant to work on a fixed pattern of quads. You have a limited choice of these "patterns" like 4 quad has 0, pi/2, pi, 3pi/2. 6 quad would have pi/3 increments.
So, although it is possible in principle, we don't think it is practical to use these cameras to communicate using PSK. Even if you use the same crystal/clock source for both cameras, the lack of coherence is a fundamental limitation.

If you want synchronised emission of light and captures between 2 cameras, you could set up the cameras in slave mode by setting register slave_mode in Table 2 of OPT9221 datasheet. Then, you would see that both cameras emimt quad 0, quad 1 etc at the same time. If you adjust the register value of frame_sync_delay = pix_cnt_max, you can see that quad 0 of a camera would interact with quad 1 of the other camera and so forth.
Another useful feature present in the product is quad_hop_en and quad_hop_offset_f1 described in Table 11 of OPT9221 datasheet. You could have some kind of a makeshift system which can work using these features. However, we may not be able to support an entirely custom set of phase sequences.

Also, this does not solve the problem of lack of coherence of the 2 modulated sources. (I don't refer to the coherence of the electromagnetic light frequency. I am talking about the coherence of the modulation signal itself.)

Please let us know if this helps.

Regards,
Subhash

Thank you for your detailed response, that answered my question!