How do the DALSA CL-P1 4096 attributes correspond to the i/o trigger signals?

[?????????????????????????????]

When attempting to use the PRIN Controlled Integration method with the DALSA CL-P1 4096 Piranha line scan camera with the NI-IMAQ PCI-1422 card to acquire wavefront data when a laser diode source is passed through turbulent flow and focused by a 64-lense array under regular lighting conditions, we have been having trouble acquiring acceptable data. We were hoping to have the ability to limit the exposure time of the camera using the PRIN signal, but the images acquired are almost black.

With the integration set to "Total" with no PRIN signal, the images are fine, although sometimes saturated. When changing the integration to the "Controlled" mode on the camera, however, and thereby enabling the PRIN signal, the image looks almost black. Taking an intensity distribution over the image, the pixel values range from only about 3 to 5 on average, and no matter what we set the line rate or the PRIN number of pixel clocks, we cannot seem to get away from these sorts of pixel values. We e-mailed DALSA and they said that the camera was working properly, but they did not know how the interfacing with the camera through the N.I. software (mostly the "Line Rate" and "PRIN" attributes) corresponded to the camera's input/output signals ("EXSYNC", "PRIN", "LVAL", and "STROBE"), and without knowing this, they could not really help us with our problem.

Attached, you will find two images acquired, one with the integration set to "Total" (PRIN disabled) and one with it set to "Controlled" (PRIN enabled). These are from the same lighting conditions and line rates, yet with very different intensity distributions. Also attached is the camera user manual for easy reference (see especially page 14 for timing). It would be very helpful to get any sort of information on how the i/o camera signals correspond to the camera attribute interface provided by National Instruments and possibly also any tips to program a new interface manually if the provided interface will not be sufficient to acquire acceptable data.

[JR_A.]

I took a look at what the camera file is doing and it should be correct based off of the timing diagram in the User Manual that you attached. The "EXSYNC" is controlled by Control Line 0 and the "PRIN" is Control Line 1. Since you are able to get good images with integration set to "Total" the EXSYNC should be working because otherwise you would get a timeout error. You can also check the EXSYNC by changing the Frame Rate in MAX and seeing if the frames acquired is correct since the line/frame rate is controlled by the EXSYNC signal. It then appears that the problem is the PRIN signal because in Controlled Mode without that signal the integration never starts. The PRIN signal could be seeing difficulties in the 1422, the cable, or the camera. Do you have the cable f
rom NI? If not it could be a cabling issue. Otherwise, we would need find out what is going on with the PRIN signal which can be difficult to probe since the camera must be attached to the cable in order for proper operation. If the cabling is correct and the frame rate changes according to the line rate that was set, it might be a good idea for you to contact NI Support directly by generation a web request for service (you can generate an email or phone request). If anyone else out there has experience with this please add to the thread. I hope this was helpful!

Cheers,

JR A.
Applications Engineer
National Instruments

[?????????????????????????????]

I believe we have made some progress in finding what the problem may be. I don't think it is the signals. Yesterday, when we started probing the signals, an odd thing started occuring when trying to take data. When a grab sequence was run, every once in a while a line would actually be taken from the camera (and the rest of the image was black). These lines seemed to be taken completely randomly. I thought that this was probably due to unsychronized signals. We went in and probed the LVAL signal---it was fine too. I started another grab sequence, and the data was in some ways okay, except it had black lines through it. The only strange thing was, the setup had not been changed at all.

I disconnected the probe setup (a m
ale and female connector wired together with extra leads to connect the oscilloscope to), and tried to take data, and the images were black again. I thought all our progress had been for nought, but I reconnected the probe setup and it started taking data like that which it was before. The connectors we use for probing seem to have a tighter connection and are gold plated, so I figured it must be a connection problem between the PRIN pins from the control bus. But one last development has completely baffled me: disconnecting the ground probe for the EXSYNC signal from the oscilloscope is what stops useful data from being acquired. All the leads that connect the camera and the control bus can be still intact, making whatever sort of connection it seems to need, but disconnecting the ground/reference probe from the oscilloscope is what stops the data. This completely baffles me, and any suggestions why this would be very helpful. I've attached one of these data frames with the pro
be and have attached control signal waveforms if they are of any use. Thanks again for all your time.

[JR_A.]

I'd say it is a grounding issue somewhere. The camera and the board shouldn't have to be on the same ground since the signal is differential. Take a look at page 23 in the user manual for the camera. In the troubleshooting steps it says to verify that all of the grounds are connected together on the connector so maybe this is a common issue.

Cheers,

JR A.
Application Engineering
National Instruments

[?????????????????????????????]

We have found out that the problem could not be a problem with the board since we have two NI-IMAQ PCI-1422 DAQ boards and both of them work in the same manner. The problem I believe must either be with (possibly) the program used to run the "Controlled" integration mode or possibly with the cable (we have the cable supplied by National Instruments). What DALSA thinks the problem is, is that the PRIN signal goes low before LVAL goes high. We believe that this might possibly be the problem. I have attached a detailed timing diagram from DALSA to show what the program should be doing. We will put together a full detailed diagram of what our signals look like next week. In the mean time, I was wondering if you coul
d give us some information about the IMAQ Set Camera Attribute.vi and how the control signals are sent from the board. DALSA said that if the PRIN signal is low when LVAL goes high, that would cause the black images we have been acquiring. Thanks for your help.

Jason Bourgeois.

[?????????????????????????????]

The PRIN signal does in fact go low before LVAL goes high. I have attached sketches of what DALSA says the waveforms should look like along with a sketch of oscilloscope data that we acquired. DALSA also drew on the sketch that EXSYNC goes high to when PRIN goes high should be a minimum of 2.4 µs, but our value was 200-240 ns. This means that the interface is not working properly. If possible, could you explain how the program can be modified so that these signals work together as they are supposed to. Thank you.

Jason Bourgeois.

[?????????????????????????????]

(slightly clearer)