Showing results for 
Search instead for 
Did you mean: 

Defining an array of clusters

Go to solution

I am using DFD toolkit to build a least pth norm filter with arbitrary coefficients. The example has the band specifications in a 1D array of a cluster with 4 elements (frequency, magnitude, weight and phase) that allows the user to type in the values. I want to read my values from a file and  have a spreadsheet with 4 columns of frequency, magnitude, weight and phase. I have read the first four rows converted them to clusters and then built a cluster array. However if I try to connect this to my filter I get an error, that says I have connected two arrays of different type, but the explanation defines them as the same type. So could someone explain what I am doing wrong? I have attached the VI and an example spreadsheet. I ultimately want to read all the rows in the file, could you also help me to construct an iterative loop to build the cluster array?

Download All
0 Kudos
Message 1 of 3
Accepted by topic author kathyleak

You have an excel file (proprietary Microsoft format), not a delimited spreadsheet file, so you cannot read it this way. Have you ever placed an indicator at the 2D array to see how the raw data looks like? Easiest would be to export the excel file as tab delimited text and read it the way you currently do.


And yes, your current manipulation is pure Rube Goldberg, because your cluster elements will not have any names. Just autoindex on the correctly read 2D array and build the cluster using a template of the array element. (Sorry, I don't have your toolkit, so I am not sure how the input should look like. Can you disconnect the array from the function and right-click that terminal and "create constant" so we have something to work with?)


@kathyleak wrote:

... that says I have connected two arrays of different type, but the explanation defines them as the same type.

What is the exact error message?

0 Kudos
Message 2 of 3

Thank you for your help. I have used a purely tab delimited file and auto indexed. Much easier than my overly complicated and superfluous design! I am now building a correctly formed cluster array to feed into the filter design.

0 Kudos
Message 3 of 3