I have measured fluorescence lifetimes that are near the response time of our
system. For this reason it is necessary to deconvolute the system's response
from the observed data.

Despite shortening my "resp" column to less than half the size of my observed
data I still get an error telling me to shorten my "resp" column.

Has anyone succesfully deconvoluted using Origin?
Do I have to FFT my sets of data first?

Any suggestions would be welcomed?

Thanks,

In general, Convolution and Deconvolution routines use datasets for Response curves that are significantly smaller than Signal curves. Origin's implementation of these algorithms additionally require an odd number of datapoints.

When you are making your Response dataset, there are circumstances in which it can appear that a dataset fulfills the requirements, but doesn't. For example, say your dataset should have 79 points (at most) and you generated a full column (160 points in this example) of data and then cleared (Edit:Clear) row 80 to 160. This would not work because the Clear command does not change the number of points, it only sets the selected range of points to a 'missing value'.

You can read (or write) what Origin thinks is your last row in the Script Window with:

get DatasetName -e last
set DatasetName -e 79

as in
get Data1_C -e last(press Enter key)
last=(press Enter key)
and
set Data1_C -e 79(press Enter key)

Given this worksheet (partially shown):

A(X) B(Y) C(Y)
1 4000 0.35889 23.93654
2 4020 1.29752 22.09621
3 4040 0.3931 17.38149
..
78 5540 10.35812 0.0
79 5560 22.61893 0.0
80 5580 23.39991
81 5600 11.30395
..
160 7180 0.86328

you can highlight columns B and C and select Analysis econvolute (or Convolute) from the Origin menu.