As for as i understood your question! you want to apply qfil. its very sample just measure the region you want to clean from unwanted signals, it would be in ppm. try to make it as lean as possible region to avoid from base line distortion problems. Put this value there and again FT the experiment. You got it what you want ? or if you still need any help can contact me.

When I enter the search term "xwinnmr qfil" into Google, I find the XWINNMR and TOPSPIN manuals posted on some Bruker user's websites. From that, I get the following:

qfil: This mode suppresses signals (e.g. water) in the center of the spectrum [...] The algorithm is derived from Marion, Ikura, Bax, J. Magn. Res. 84, 425-430 (1989).

Further, when I Google "J. Magn. Res. 84, 425-430 (1989)" I find the full article at Ad Bax's website: http://spin.niddk.nih.gov/bax/lit/508/115b.pdf

It appears to be a Gaussian high-pass filter applied to the FIDs prior to multi-dimensional Fourier transformation.

P.S. The Bax article describes the application of the filter in the t2 dimension, which makes sense if we are talking about solvent suppression. The quotation in your posting states that qfil was applied in the t1 dimension. As such, it cannot be for solvent suppression, but must be intended to suppress axial peaks which were not suppressed by other means (phase cycling or gradients). This sounds like a cosmetic application, i.e., beautifying the spectrum but not actually enhancing the spectrum for purposes of analysis. Axial peaks, if present, may simply be ignored in the analysis. Therefore I would not advocate the use of this, or any other high-pass filter processing, to any indirect dimension in multidimensional NMR. Does anyone know of a case where it is actually necessary or helpful?