I want to track the amplitude at 828 cpm throughout a database using the parameter profile graph so I can show up to 144 points at a time.

I set my AP set to track only 1 parameter described sorter speed, units type is veloc, type of parameter is cpm int, lower frequency 820.0 and upper frequency is 830.0. (I tried a narrower band of 828 to 830-no change).

In plotdata/parameter profile I select the measurement points on a machine and it displays two graphs. One of the overall values and one of the sorter speed paramter for each measurement point. Everything seems fine, but the values displayed in the parameter profile graph are not the same as the values shown in the referenced spectrum.

Any ideas what's going on?

Danny,
Is your LOR fine enough to fit in that tight of a band? Don't forget that your Hanning window multiplies LOR by 1.5 or did I totally misunderstand what you are doing?

David Eason

David,

Maybe this will clarify it for you

parameter_profile.doc (43 Kb, 18 downloads)

The answer is:

Once the data is collected it is stored as a number based upon the parameter as it was collected. I changed the parameter after collecting and dumping the data, so the parameter is a number stored in the database rather than an actual band in my spectrum.

Lesson learned: Changes to the parameters effect only data collected in the future. Data already collected and stored is just a number in the database and not a reference to a band in the spectrum.

Sorry Danny, it's been this way with CSI forever! Based on your posting, I thought CSI had actually added a new feature; something I wish was there!

Walt

quote:

Lesson learned: Changes to the parameters effect only data collected in the future. Data already collected and stored is just a number in the database and not a reference to a band in the spectrum.

Not exactly true Danny....if you have Autostat, then you can recalculate these parameters using the old data (I recommend copying the database with Dbutly and then playing with the copy in case something goes wrong)....you just have to make sure that the units for the AL set you plan to use match the updated AP set...if not, then just create a new AL set....make sense??....or am I not understanding the question correctly??

This message has been edited. Last edited by: Billy,

Billy,

Yes, that sounds right to me, but I don't have autostat.

I wasn't any big deal anyway. I just used multiple point plots and looked at 18 at a time.

Thanks to all.

Danny, I think it's exactly as you described. The parameter is calculated at the time the spectrum/waveform is collected and is then just a "static" number in a database.

Also, if you change a parameter band, the new data goes in the same place in the database (i.e., parameter 6 is always stored in the same location no matter how you define it. For this reason, you DO NOT WANT TO CHANGE A DEFINED PARAMETER BAND once it is set up. If I am not sure what parameters I am going to want to define for a new machine, I will start with a basic parameter set and then add parameters to the set as I discover what they need to be.

Hate to be a stickler for detail, but just want to make sure that everyone knows that parameter bands in AP sets can be changed if you have AUTOSTAT. If you don't have this feature then you are correct...you're just out of luck...Again, you have to be careful that the units from your changed AP set match the units for your AL set....if not, then you'll get some unusual results...

This message has been edited. Last edited by: Billy,