How to connect the external trigger input

[Rasika Dissanayake]

 I am using the same DAVIS 346 mono camera that Kai uses here. We are trying to trigger the Davis 346 camera using an external trigger input. 

This is the layout (Figure 1).

Figure 1: Camera triggering setup 

The trigger generated from the laser, which is 10 kHz, is converted to 1kHz/100 Hz and directed to the SRS DG 535 delay generator to create a pulse with a 15 µs pulse width after 100 µs delay from the original pulse (C=T0+100µs, D=D+15µs). CD pulse (load high Z, TTL) is directed to BNC 577, which generates the trigger for the camera (Pulse duration 20 µs, peak to peak voltage 4.5 V)

However, we realized that our trigger resets unexpectedly (Figure 2), which is not suitable for the work we intend to carry out

Figure 2.1: Trigger spacing for the rising edge detection

Figure 2.1.a: trigger spacing values (expected to see a constant value)

Figure 2.2: Trigger spacing for trigger type 3

Figure 2.2.a: Trigger spacing values (expected to have a constant value)

We tried different settings in DV software for the external trigger input, and I have attached the xml file for your reference. 

I have attached the output trigger files where we detected the rising edge and the pulse detection (Trigger type=1: and 3:)

We noticed that, trigger signal has an unstable peak-peak voltage (4.5-4.7 V) also when connected to the camera, and don’t know what causes it. 

Figure 3: Trigger signal from BNC577 to the camera when the camera is attached. 

But when the camera is not connected BNC 577 output pulse is stable at 4.9V

Figure 4: Tigger signals to the camera when the camera is not attached 

If anyone is familiar with this situation, looking forward to your help 

[Luca Longinotti]

Hello, this looks like a somewhat complex setup. Some jitter is expected in the trigger timestamps, but more in the order of tens of microseconds, not milliseconds.

I would suggset trying the simplest possible test:

- get an Arduino (or any signal generator), set one of its 3.3V pins to a 100HZ square wave, attach GND, 3.3V to both VDD and CLOCK_IN (CLOCK_IN should always be tied to VDD if unused to avoid spurious timestamp resets as per documentation), and the signal pin to SIGNAL_IN.

Then enable rising edge detection and see what you get. If that still looks weird, there might be a hardware issue (cable/connector most likely).

-- 

Luca Longinotti (llongi)

Senior Software Engineer
iniVation AG - https://inivation.com/
A SynSense Group company

[Rasika Dissanayake]

Dear Luca, 

Thank you for the suggestion. I simplified the setup as Laser--->Frequency Divider -->BNC577-->Camera

As you suggested, I connected GND, VDD, and Clock to the Clock out of the BNC577. I was able to get a constant trigger spacing with some deviations. 

Thank you for the help