How can I measure an exact frame duration for my CMOS/Camera Link/1428 system?

[Fantazzer]

Hello,

I'm building a system that combines a mashine vision component (CMOS/Camera Link camera, 1428 board) and a galvo scanner driven by PCI-6731 analog board. I have LabView 7.1 and Vision module. I need to precisely sync the galvo with the camera in a following matter: the galvo must scan start motion at the beginning of a frame acquisition and stop at the end of frame acquisition in a single frame acquisition mode.

Before I was using a Epix Camera Link interface where I had a physical access to FVAL signal from the board. I was using this TTL signal to (i) trigger my galvo and (ii) measure the frame duration to adjust my galvo driving signal. For my new 1428/6731 combo I'm planning to use RTSI to trigger my galvo, but it is still unclear how I can reliable measure the frame duration since 1428 does not provide FVAL output. I need to know the FVAL duration with ~ 0.1% accuracy, i.e. ~ 0.1 ms.

Questions:

1. Is there any way to get an access to analog FVAL signal from 1428?
2. Is there any way to measure the FVAL duration with required accuracy using LabView?

I would appreciate any suggestions. Thanks.

Pavel

[Remzi]

Hi Pavel,

You might use IMAQ Trigger Drive2 with Vsync as the signal. For the 1428 this should give you FVAL, but you can test it to make sure. Then you can route this signal to an external line or RTSI line and measure the duration on a scope. Make sure it agrees with the timing you expect for the camera you're testing. Also try adjusting the height of the acquisition window and make sure the pulse width does not vary. If you want to see the duration of just the lines you're acquiring, you can use Frame Start as the signal. Keep in mind there's no way to measure the duration of this pulse with the 1428. You will need some other device. I would recommend using RTSI and a 660x board. I hope this helps.

Best Regards,
Remzi A.
National Instruments

[BruceAmmons]

The 6731 has two counter timers on it, which should be able to do what you need. You should be able to measure pulse widths within microseconds easily. You won't need to buy a separate card just for the timing.

Bruce