Peak detection- problems finding x values of location

[kikemart]

You know the previous program does have a faster write to the power supply. We're using an HP 8341b synthesized sweeper. I though using the driver would be too slow so instead I used a GPIB CW (Constant Wave) write, but the guy before me writes and "FP" string instead of the "CW". I tried using his subVI for this but whenever I do the sweeper's power output jumps all over the place and gives me an "Overload" warning, so I simply stayed away from that. 

In that case i guess my job is to figure out why the subVI doesn't work in my program but will work in his, and then implement it into mine. Thanks! I hadn't though that would be such a big deal.

One question, however. How does the data aquicsition rate impact the peak detection? Won't the same data points be recorded no matter how quickly? Thanks

[johnsold · Accepted solution]

I cannot help with the commands to the sweep generator. I have seen instruments which behave somewhat counter-intuitively so you may want to re-read the manual carefully to make sure the intstrument is doing what you think it is.

The data acquisition rate does not affect the peak detection but if the frequency sweep and the analog acquisition are not synchronized properly you may be getting measurements which actually reprot the resutls of a frequency different from where you think it is.

Lynn

[kikemart]

It seems you were right, the Fast Phaselock command made all the difference, my peak detector is every bit as accurate as the last programmer's. Thank you very much! 

All I had to do to get the FP GPIB command working was work with the numbers the exact same way the last guys did, and although it was the least logical and most redundant math black magic I have seen in my life. But hey, it works! +Kudos for all your help.

[johnsold]

When working with complicated or sophisticated intstruments (or when you cannot tell the difference), you sometimes need to be very carfeul to determine whether the instrument is doing what you think it is.  Anyway, glad you got it working.

After you take a deep breath, it might be a good time to "reverse engineer" the code which works so that you can really understand what it is doing. Then document the algorithm or equations so that when new questions arise, you can start from a basis of understanding.

Lynn