File format used by the dmfit program

Reading

The "dmfit" & "Edit NMR" programs can read in different file formats that they read in using automatic detection routines :

+ Native Bruker XWinNMR format from the spectrometer (Unix, WindowsNT and Linux ) (read only)
+ WinNMR (Windows PC) file format (read and save)
+ Ascii files (list of X and Y coordinate with even spacing in X - Hz) (read and savedownload an example
typical file :
ti: string title
##freq <frequency in MHz>        // optional
x1 y1
x2 y2
...
xn yn
 
Important remark :
+ the dX increment is computed from xn and x0 dX = (xn-x0)/(n-1)
+ the sweep width Sw is taken as n*dX to be compatible with Fourier transfromed data
+ Simpson 1D & 2D ascii files ((read and save download an example

Warning : the coding of the SIMPSON Header previously used combinations of X0, Sr and dX values resulting in redundant coding of the frequency and inconsistent results [marginally in most cases]. From dmfit #20170801 the header has been reduced to X0 only as described below. The compatibility is maintained as much as possible.
Many thanks to Simone Weigelt for mentioning and identifiying this issue.


for a spectrum : TYPE=SPE - taken to be described in decreasing frequency
+ the X0 parameter is the Hz position of the left most point
+ the X increment is taken as dX = -SW/NP
+ the SF value is the reference frequency for computing ppm scale and NMR interactions

example of SIMPSON header :
SIMP
Comment= Saved by dmfit #20170801
NP=2048
X0=12237.3
SW=20000
Sf=50.3292
TYPE=SPE
DATA
 -0.0783812 0 -0.117981 0
...

for a 2D spectrum X0_F1 and SF1 are added to the SW1 parameter in the same wayy
+ Tecmag, ".tnt" files  ((((read only, save in Simpson format)

FID can be loaded and basically processed
There are some remaining problems for referencing the spectra, put me a mail with an attached file if you have documented exemples

+ Varian/Chemagnetics ... in progress (read only, save in Simpson format)

FID can be loaded and basically processed
There are some remaining problems for referencing the spectra, put me a mail with an attached file if you have documented exemples

+ Grandinetti's RMN, in some cases (read only) Ohio State University - P.J.Grandinetti

Saving
The "dmfit" & "Edit NMR" programs only save using the  following formats :
+ WinNMR (Windows PC) file format
+ Ascii files (list of X and Y coordinate with even spacing in X - Hz)

Native Bruker XWinNMR format 

guest is the username
exam1d is the experiment name
1 is the experiment nb, it contains fid, aqus, acqu2s, ser, pulprogram  files
pdata contains the processed data
1 is the process nb, it contains the 1r, 1i, procs, 2rr, 2ii, 2ir, 2ri, proc2s files and many others

WinNMR PC format 

All files reside in a single directory named with the experiment name.
eeeppp.xxx correspond to experiment e and process p files (007001.xxx is exp7 proc 1)
dmfit and EditNMR programs accept names as well as numbers
The correspondance between XWinNMR is the following :

1D dataset    
XWinNMR WinNMR PC Content
acqus 007001.aqs acquisition paramters
procs 007001.fqs process parameters
fid 007001.fid raw fid data points
1r 007001.1r spectrum real part data points
1i 007001.1i spectrum imaginary part data points

 

2D dataset    
XWinNMR WinNMR PC Content
acqus 007001.fa2 acquisition paramters F2
procs 007001.fp2 process parameters F2
acqu2s 007001.fp1 acquisition paramters F1
proc2s 007001.fp2 process parameters F1
ser 007001.ser raw ser data points
2rr 007001.rr 2D spectrum real part data points
2ii 007001.ii 2D spectrum imaginary part data points

The acqu & proc files are now in ascii and were previously binary coded, dmfit and EditNMR can read and save both.
fid, ser, 1r, 1i, 2rr, 2ii, 2ri and 2ir contain integers with encoding depending upon the OS used.
WinNMR data files (ser, fid, 1r, 1i, rr, ii...) are usually stored in float format.