Clock delays due to filters for MKIV and VLBA stations

R01C1 was an experiment carried out in June 2001 to determine the clock delays introduced by different filters for MKIV and VLBA stations. The delays of the VLBA filters are different to those of the MKIV filters resulting in an apparent clock jump on baselines composed of VLBA and MKIV stations whenever the filters are changed.

An ASCII table of the results can be found here.

Experiment details

For those interested in the details of the experiment, the (absolute) delays between each set of filters for every baseline in the experiment are given below. There is a separate plot for the delays between each pair of filters. The different baselines are colour coded according to whether they contain two stations of the same type (indicated by "identical" in the following plots), one MKIV and one VLBA station, one MKIV and one EVN_VLBA station, or one VLBA and one EVN_VLBA station. For the purposes of this experiment an EVN_VLBA station is defined as an EVN station with a VLBA rack (irrespective of the formatter). The MKIV stations were Jb, Mc, On. The VLBA stations were Nl, Hn. The EVN_VLBA stations were Eb (Efflesberg VLBA terminal), Nt, Tr. Note that the delays for the EVN_VLBA stations (including Eb) are slightly different to those of the VLBA stations.

There are two delays given for each baseline - one for the LL and one for the RR correlations.

The solid lines show the mean value of the data points and are colour coded in the same way as the data points to which they correspond. The standard error on the mean is approximately 2 ns in each case.

• 2 and 4 MHz
• 2 and 8 MHz
• 2 and 16 MHz*
• 4 and 8 MHz
• 4 and 16 MHz*
• 8 and 16 MHz*

*Please note that the 16 MHz results are preliminary and require confirmation.