D405 triggering
Is it possible to externally trigger capture for the new D405 camera? Looking at the specs for the 400 series, I see:
"For D415 and D455, both depth and RGB sensors can be synced by external trigger. For D435 and D435i, only depth sensor can be synced by external trigger. For D401, external sync mechanism is not exposed."
But nothing about the D405.
-
Hi Jane Haslam D401 is the camera circuit board inside the D405's casing. The D405 therefore cannot be used with external triggering.
In the RealSense Viewer, the D405 does though support the ability to generate a trigger from the camera to an external device once per frame when the 'Stereo Module' option Output Trigger Enabled is enabled. The equivalent SDK script instruction for controlling its status would be RS2_OPTION_OUTPUT_TRIGGER_ENABLED
-
Hi Timothy P Baxter Social Vision Processor D4 board versions typically work with certain RealSense models. For example, the D450 depth camera board (the circuit board inside the D455) requires Vision Processor D4 V3 boards, whilst the V2 board is used with the D430 camera board to provide IMU capability like the cased D435i model has. The V4 is designed for pairing only with the D401 camera board inside the D405.
Furthermore, the D401 module's design is not suited for mounting by means other than in the D405 casing.
The bottom of page 62 of the current March 2022 edition of the data sheet document for the 400 Series cameras states though that "The V1 ASIC (without IMU) is validated and supports all D400 depth modules".
https://dev.intelrealsense.com/docs/intel-realsense-d400-series-product-family-datasheet
Whilst you are free to experiment with non-recommended combinations of camera components, any faults or breakages that occur from doing so will be at your own risk, and opening the D405's casing would also invalidate its sales warranty.
-
I had further discussions with Intel about the D401 module. I am permitted to say that the module will be released for sale separately, consistent with the listing on the Digi-Key website. There is not a confirmed release date that I can provide at the time of writing this though.
Regarding the possibility of switching the Vision Processor D4 V4 board for another D4 version such as V1, I am informed that this will not be possible due to the hardware design differences of the D4 V4 / D401 combination.
The D405 information page at the link below provides a helpful disassembled view of the D405's internal structure.
-
The Digi-Key website is now listing the D401 depth module as In Stock.
https://www.digikey.com/en/products/detail/intel-realsense/82635DSD401/16265563
-
Hi Tim Rensen The D4 V4 board used with the D401 depth module does not have a set of hardware sync pins and so inter_cam_sync_mode cannot be used with it. Without a pin set, trigger signal output cannot be tested by monitoring the sync pin (pin 5).
Another model of D4 board with a sync pin set cannot be used with the D401 module. It will only work with the pin-less D4 V4 model.
Below is an image of what the pin set looks like on another RealSense model that is equipped with it.
-
Hi Marty, thanks for your swift reply. I deleted my comment because I have found the answer with https://support.intelrealsense.com/hc/en-us/community/posts/9425447647635--D405-Function-of-FSTROBE-signal-. Pin-25 of the 50 pin connector between D4 processor and camera module is FSTOBE or exposure signal and is exposed as TP2 just behind the 50-pin connector on the camera module. We are very space constrained so the D435 is too large. Probing TP2 we see that the signal is a 1.8V peak square wave with the duration of the exposure setting and frequency matching the camera fps. Will verify if the signal is in phase with the exposure tomorrow.
Tim
-
After some testing we have deduced that the FSTROBE signal overlaps the actual sensor exposure as expected - we used a digital isolator driving an LED and looked for the LED in the frames. The LED is visible and changes brightness with exposure setting, but disappears suddenly at 71us exposure or less suggesting there is likely a 70us delay between sensor exposure start and FSTROBE rising. Could be the other way round also, but for our purposes of triggering a strobe light this is fine.
Please sign in to leave a comment.
Comments
14 comments