WAPP data with possible wrong frequency settings
================================================

Data affected: all previous pulsar WAPP-files up to 2006-Oct-11 which
have been using the WAPPs with 100 MHz bandwidth and (A) have been
setting up the IF/LO path using bad CIMA .gui- or .iflo-files (see
list below) or (B) have been setting up the IF/LO path AFTER loading
.wapp-files into CIMA.



AFFECTED PROJECTS

Projects that contain bad .gui-files:

P2017:  P2017set_1400.gui
P2019:  p2019setup.gui
P2099:  1906+07_Lwide_20050628.gui
P2179:  P2017set_1400.gui
P2203:  p2019setup.gui, p2203setup.gui
P2206:  old1913+16_Lwide.gui, old1913+16_LwideplusHI.gui,
                old1913+16_LwideplusHI_cal.gui, old1913+16_cal_Lwide.gui,
                old1913+16_cal_LwideplusHI.gui

Projects that contain bad .iflo-files:

P1810:  p1810_iflo_wapp_asp2.iflo, p1810_iflo_wapp_asp3.iflo
P1834:  p1834_iflo_wapp_asp.iflo, p1810_iflo_wapp_asp3.iflo
P1880:  ifsetup_lbw1398_1534.iflo
P2017:  ifsetup_lbw1398_1534.iflo
P2018:  ifsetup_lbw1398_1534.iflo
P2099:  ifsetup_lbw_1906.iflo, ifsetup_3w_asp_hi.iflo
P2179:  ifsetup_lbw1398_1534.iflo

Projects with problematic .wapp-files:

ALL pulsar projects which use .wapp-files and set the bandwidth to 100 MHz.

However, note that problematic .wapp-files are not a problem provided that the 
IF/LO set-up has been done BEFORE the .wapp-file is loaded into CIMA! Whether 
the bug affects a project using .wapp-files thus depends on the ORDER of 
things the observer did during the observation sessions!!!


POSSIBLE ERROR

WAPP data taken with 100 MHz bandwidth may have been taken with the 
destination frequency incorrectly set to 275 MHz instead of 250 MHz.

REASON

The bandwidth selection is stored in four variables called 'iflo(bwN)' where 
'N' is either 0, 1, 2 or 3. This variable is supposed to be a string 
containing a single numeric value. However, in the above-mentioned 
configuration files, these variables (or some of them) also contain the word 
'MHz' once or twice, for example:

        set iflo(bw2) "100 MHz"

instead of

        set iflo(bw2) "100"

This variable is used to determine if the destination frequency should be 275 
MHz (used for bandwidths of 50 MHz and smaller) or 250 MHz (used for 100 
MHz).

It turns out that if the variable does not contain a pure numeric value, CIMA 
will default to use 275 MHz for the destination frequency.

The reason that the word 'MHz' appears in some of these files comes from the 
fact that a similar variable in the .wapp-files ('wapp(bwN)' with 'N' being 
either 1, 2, 3 or 4) also contains the WAPP bandwidth. This variable, 
however, is supposed to be a string containing the word 'MHz'. The problem 
arises when a .wapp-file is loaded into CIMA since the data from 'wapp(bwN)' 
then is copied into 'iflo(bwN)' WITHOUT the word 'MHz' being removed. This 
turns into a problem if the IF/LO-setup is done AFTER the .wapp-file has been 
read into CIMA or if a .gui- or .iflo-file is created afterwards.

RESULT

Any project that have used any of the above-mentioned configuration files with 
bandwidths of 100 MHz have probably got data that is 25 MHz off.

Any project that uses a .wapp-file and does an IF/LO set-up AFTER the 
.wapp-file has been loaded into CIMA should also be affected in the same way.

BUG STATUS

CIMA version 2.2 (normal) was patched on 11 October 2006 so that the word 
'MHz' is stripped from 'iflo(bwN)' when a .gui-, .iflo-, or .wapp-file is 
read.



						Mikael Lerner

						Arecibo, 11 October 2006