About x,y axis movement programs in LabVIEW

[ikeda_kaoru]

I am planning to acquire a 256×256 pixel image by scanning the probe and acquiring data for each pixel.

Currently, I have programmed the probe to move 1 pixel along the y-axis when it moves 256pixel along the x-axis.

After the y-axis reaches 256pixel, it goes back to the 0pixel position and then moves again incrementally by 1pixel. I would like to improve the program so that after reaching 256pixel, it decreases by 1pixel, i.e., the x-axis direction is now triangulated.

In other words, the current behavior is that of a triangular wave in the x-axis direction and a sawtooth wave in the y-axis direction, but I would like to change it to a triangular wave in both the x and y axes.

Current:：x-axis⇒0,1,2...256,255,254...2,1,0,1,2,3,...256...

　　　  ：y-axis⇒0,1,2...256,0,1,2...256,0,1,2...256...

After improvement：x-axis⇒0,1,2...256,255,254...2,1,0,1,2,3,...256...

　　                   　：y-axis⇒0,1,2...256,255,254...2,1,0,1,2,3,...256...

I am attaching the program and would appreciate your review and advice.

[altenbach]

Instead of many words, a picture explanation would be clearer, but here's one possible  interpretation..

(note that if the number of columns is odd, slight changes are required)

[Bob_Schor]

Kon-nichi wa, Kaoru-san.

I'm not sure that you described "how to scan a 256 x 256 area".  The first few sentences seem to say that the "frequent movement" is along the X axis, from 0 to 255 (which is 256 different position).  Once this movement is complete, the next movement is to increase Y by one, then make X move from 255 down to 0, increase Y, move X from 0 to 255, etc. until you end up back at (0, 255). 

Consider only the X movement.  Let X start at 0.  What kind of LabVIEW Structure (on the Structure palette) will let you do something exactly 256 times?  What do you want to do in those 256 times to X?  [Hint -- there are two functions on the Numeric palette that describe how to make X "grow" or "shrink"].

If you can figure out how to do this, you'll have a method to move X from 0 to 256 (why 256, and not 255?  What are you doing to X, and how many times are you doing it?

So you've managed to get X from the left pixel position to one past the right pixel position (if you look at the "wire" carrying the value of X that goes across the "unnamed Structure" I described that runs 256 times, it should start at 0 on the left side of the Structure, you do something to it, and it increases by 1 when the wire reaches the right side of the structure).  Does the term "Shift Register" mean something to you?

OK, you know how to get X to shuttle back and forth.  What about Y?  Well, Y also needs to move 256 times, each time letting X shuttle right or shuttle left.

I wrote (as an exercise for myself) a simple VI that has three Structures (two to "count" X and Y through 256 positions, and one to decide if X was increasing or decreasing) and two outputs -- one I called the "Scan Array", as it started [[0,0], [0,1], [0,2]...[0,255],[1,255],[1,254]... and ended [2,255],[1,255],[0,255] and the other I called "Array of Scans", which was a 3-D array with the first element being the scan "number" (0 being the first left-to-right scan of 256 X-Y "pairs" where Y is 0, and 255  being the last Y scan where Y is 255 and X is decreasing down to 0.

I'm not going to post this now so that you can try to do what I tried to describe in words (pictures, i.e. LabVIEW code, is so much easier to understand once you've gotten the hang of it), but after you have a chance to respond to the comments on your Post, I'll post my "Scan VI".

Bob Schor

[ikeda_kaoru]

Thank you for your response.

I would like to add the program shown by @altenbach to the program I am currently using in my research (attached file: testq.vi).
Also, please tell me how to set the time to move 1 pixel to 300μs.
Thank you in advance.

[ikeda_kaoru]

Kon-nichi wa, Bob_Schor-san.

Thanks for answering my question.

The program currently in use scans 1 pixel at 300μs, scanning 1 pixel in the x-axis direction and moving 1 pixel in the y-axis direction after scanning to 255 pixels. It then scans in a decreasing direction of 1 pixel in the x-axis direction and scans to the 0 pixel point.

The problem is that after scanning 255pixel in the y-axis direction, it returns to the 0pixel point. I am trying to use the decrement function, the selection function, and the shift register, but I can't seem to get it to work, so please let me know.

Thank you in advance for your help.

[altenbach]

Again, it would be so much clearer to show a sketch how you want to scan. Your description is ambiguous.

[BigBri]

Having done similar scans of a measurement plane using probes in wind tunnels, I think I understand what he's after. What he has right now is what we'd call a 'Z' scan arrangement where the 'X' coordinate resets to 0 whenever 'Y' is changed. Think like a typewriter doing a carriage return after every line. What he wants is what we'd call an 'S' scan where 'X' does not return to 0 for every 'Y' but starts at the last 'X' of the previous line and travels in the opposite direction of the previous line. Image the same typewriter but, after finishing the first line typing left-to-right, drops down and continues the typing right-to-left back towards the starting 'X' coordinate. This motion profile minimizes total time spent moving by the traverse.

[altenbach]

I guess that's what my example already does, right?

[ikeda_kaoru]

Sorry. Attached is a file (scan.pdf) with a sketch of the scan, the first page shows the current scan and the second page shows the program after improvements.
As shown in the figure, the current scan returns to the 0-pixel point as soon as a 255 x 255 pixel scan is made, and the scan is started again. In other words, the behavior of the scan in the x-axis direction is a triangular wave and in the y-axis direction is a sawtooth wave.
In the improved scan, we would like the scanner not to return to the 0-pixel point immediately after a 255 x 255 pixel scan, but to return one pixel at a time from the (255,255) point. This means that both x- and y-axis scans will have a triangular wave behavior.
Sorry for the confusing explanation.

[ikeda_kaoru]

Your program was very helpful.
I am attaching the original program(fpga_test3.vi), which I am currently using.
I would like to incorporate your program into this program so that I can scan.

I would very much appreciate your help.

The area circled in red in the file(original_scan.pdf) is the scan program described here.

Thank you for your cooperation.