5MS/s acquisition w/ the 5105

[hankypanky]

Hello, I've been trying to get continous capture working on my 5105 at 5MS/s for days now.  While I can sample at lower rates with minimum "error" (I sample a sin wave and compare it to a perfect wave generated in software), sampling a 1Mhz sin at 5MS/s doesn't work well at all.  In particular, sampling the sin should yield the same values across sin wave periods, since 5M is a multiple of 1M, but I'm essentially missing by a small "phase" each time, so that comparing the 1Mhz sin to one generated in software (that is, just taking the difference of the two) yields, over long time periods, a triangle-wave looking.  Perhaps the function generator generating the 1Mhz sin I'm capturing is off, but I doubt it, since it claims to support up to 80Mhz.  My VI is based on the EX Fetch Forever example.  One thing I don't understand is the difference between Min Record Length and Max Points Per Fetch.  How should I be choosing these parameters if I want to fetch optimally at 5MS/s and higher? 

[hankypanky]

So I think I've gotten to the bottom of this.  The Fetch Forever example seems to be "broken" (at least under my settings and on my machine).  In short, it seems that each fetch overlaps with the fetch immediately before it.  I.e. when I try to capture a simple sin wave, look at what I get (I changed the scroller so I could see data across multiple fetches instead of just the data from one fetch) -- see attached image.  This isn't what's supposed to happen, right?  How do I get rid of this overlap???

[hankypanky]

Ok, I don't see where it attached the image (if it did at all), so for those of you like me, here's a link to the image:
http://people.virginia.edu/~hwa5c/sshot.JPG

[Sean_C.]

Hello,

I am not able to reproduce what you are seeing.  Will you please attach your code to this post, so I can take a look?

Regards,
Sean Close
Product Support Engineer: High Speed Digitizers

[hankypanky]

sure

[hankypanky]

(warning: press stop quickly, b/c i simply save everything in that buffer that gets plotted!)

[hankypanky]

Ok, I just didn't understand the whole Fetch Relative To attributes.  I had it set to Read Pointer instead of Start.  Silly me -- maybe I should read the instructions next time.  (I think this is the problem -- if it isn't please let me know, heh).

[Sean_C.]

ok...  You were concatenating your fetched data to the beginneing of your buffer, not the end.  Reverse those 2 wires and it works perfectly.

[hankypanky]

Darn it, this doesn't seem to be what I want either.  Now the graph appears smooth, but I'm thinking that's because it's always fetching the same points (the ones at the start; hence the word Start).  I want to fetch points like Read Pointer does, but I want the graph to be smooth...

[hankypanky]

Aha!  Whoops, heh!  Thanks!