Notes on Voyager Quick Look Data:  

Since Oct 2, 2006, "quick-look" magnetic field data for Voyagers 1 & 2 
have been posted on the Internet within a day or two of acquisition by 
the spacecraft. The rapid availabililty is achieved at the expense of 
the quality of the data. The quick-look data are processed with the last 
available roll and zero level data. The quick-look data are very 
preliminary and must be used with the greatest caution.  At the time of 
experiment proposal, the mission (then called "Mariner-Jupiter/Saturn") 
was designed to investigate Jupiter, Saturn and the interplanetary 
medium out to Saturn at 10 AU, where the interplanetary magnetic field 
strength is 0.6 nT. These objectives determined the required accuracy 
of the measurements, which together with the nature of the spacecraft 
determined the design of the instrument. The spacecraft magnetic field 
at the outboard magnetic field sensor, referred to as the primary unit, 
was expected to be 0.2 nT and highly variable, consistent with current 
estimates. Hence, the dual magnetometer design (Ness et al., 1971; 
Behannon et al. 1977). At distances > 80 AU, the magnetic field strength 
is less than ~0.05 nT in the solar wind, an order of magnitude smaller 
than the instrument was designed to measure and a fraction of the 
spacecraft magnetic field strength at the outboard sensor. Thus physical 
signal measured by the Voyagers is a fraction of the highly variable 
non-Gaussian noise that produced primarily by the spacecraft.  

The use of roll calibrations lasting ~6 hours permits determination of 
the effective zero levels for the two independent magnetic axes that are 
perpendicular to the roll axis (which is nearly parallel to the radius 
vector to the Sun) at intervals of  ~ 3 - 4 months. Two successive rolls 
are required to extract the signal from the noise. There is no roll 
calibration for the third magnetic axis; data for a month after an 
additional month after a roll are required to calibrate the third axis. 
Data between two successive rolls + 1 month are required in order to 
extract the signal from the noise.  Given this information, a comparison 
of the two derived magnetic vectors from the two magnetometers permits 
validation of the primary magnetometer data with an accuracy of 
0.015 nT - 0.05 nT. A discussion of the uncertainties that must be 
considered when using the fully processed data is given in the Appendix 
of Burlaga et al. [1994] and in Appendix A of Burlaga et al. [2002]. 
     
Since the quick look data are necessarily processed with information from 
only one (the last) roll, the information required to extract the signal 
from the noise is not available after that roll, and the uncertainties 
in the quick-look data can be very large. Indeed, in general, the 
quick-look data are dominated by the variable spacecraft magnetic field.