next up previous contents
Next: CONCLUSIONS Up: The Trimble AcutimeTMaccurate event Previous: The Trimble AcutimeTMaccurate event   Contents

Tests and Results

This section presents here the results obtained while testing the event timestamping with the AcutimeTM
interface. We connected the event input of the AcutimeTM Synchronization Kit to a Quartz-Controlled Time Base Generator, set to 59.2 seconds. The time differences between each timestamp were then plotted on a Time Series plot, as shown in Fig.19. At a first glance we are able to see that the maximum error is about 250 microsec, to be taken into account for seismic purposes. The signal rise side started at 2004-02-11 09:00, there was lack of data 16:30 to 17:30 and a steep descending step began at 18:00. This behaviour should be investigated further, either on the GPS constellation management or on the TB side. In addition, there were some spikes, both for positive or negative time differences. In spite of the above problems, we can see that the data seem to be quite stable, and well within the AcutimeTMspecification.

Figure 19: EVENT/TIMESTAMP. The time differences between events are shown.
\begin{figure}\centering
\epsfig{ file=IMG/ACU2K.eps,width=12.5cm}
\end{figure}


next up previous contents
Next: CONCLUSIONS Up: The Trimble AcutimeTMaccurate event Previous: The Trimble AcutimeTMaccurate event   Contents
2004-03-24