Precise timing of the SMB trigger on the PCIe-1429 board?

[pwebster]

Greetings,

I am using the SMB input on my PCIe-1429 board to trigger a line camera. I am used to having a "sync out " connection on my camera to tell me precisely (better than 1 us) when it is integrating, but my new camera does not have one. Therefore, unless there is some pin inside the frame grabber I can measure, I must calculate precisely when it will integrate using only what I can measure from the signal going into the PCIe-1429. The camera specifications show that it begins integrating 1210 ± 20 ns after it recieves a trigger event from the frame grabber. The PCIe-1429e manual does not specify any related timing (which is strange for laboratory-grade equipment). So my questions are:

1) Is the SMB trigger connection wired directly through the CL cables to interface directly with the camera, or does the frame grabber do something to the signal first?

2) If the latter, how long does that operation take?

3) How precisely then can I know the time between when I send a trigger pulse to the frame grabber and when the camera will integrate?

[Andy_C.]

Hello pwebster,

SMB trigger connection is wired directly through the camera link cables to interface with the camera. What is the timing information that you wish to know about PCI2-1429? If you want to measure the time it takes for the camera to acquire the image and send it back to the computer, you will have to rely on the manual of the camera.

[pwebster]

Hi Andy,

Thanks for your reply. My goal is to get a precise measurement of when the camera integration is occurring vs. when I send the trigger to the framegrabber. Time for readout and transmission to the framegrabber/system memory is not as important to me. I made an estimate by correllating the camera signal with a laser that I could pulse and found that the total delay was about 1700 ns. If the camera specifications are correct, then there is approximately a 500 ns delay from the time I send the trigger to the time the camera gets it. That would suggest that there is more than just a direct connection.

Also, what is the input impedance of that trigger in?

Paul Webster

[BlueCheese]

Hi pwebster,

I'm not sure Andy is correct on this one. All the signals to/from the camera and framegrabber go through a routing matrix that allows the board to route and drive triggers. Because there is some digital logic and clock boundary crossing there is going to be some delay and jitter associated with the line from the camera routed out to the SMB. I don't know these numbers offhand although they might have been calculated previously (which I'll check on). However, they are *likely* under a couple hundred nanoseconds based off a quick guess as to the implementation. I could ask some of the hardware engineers to look into this and describe/calculate the exact timings, but it might take some time for that request. Quite possibly you could measure this with a scope as well and characterize it.

Eric