PXI Line Trigger Question

[mcduff]

I have a PXIe-1092 Chassis, a NI-6683H GPS Card, and a PXIe-5172 Oscilloscope. I am trying to synchronize the absolute time to the GPS time. I also have an independent Trimble GPS.

For the plots used below, I used NI Instrument Studio and NI-MAX. I'll assume NI's programming is up to par.

Experiment: Compare two PPS signals from two GPS modules and see if they are synchronized within a reasonable amount.

Experiment 1: Use NI-Max to output a PPS on the PFI port of the NI-6683H GPS Card and input that into the PXIe-5172 Oscilloscope. On another channel of the Scope input a PPS from the Trimble GPS. Trigger the scope on the edge of one of the PPS signals. Plot shown below. There is roughly a 10ns difference between the pulses. This seems completely reasonable to me. Watching the traces continuously that jitter jumps around a little but again seems well within specs.

Experiment 2: Use NI-Max to output a PPS on a PXI trigger line from the NI-6683H GPS Card and use that PXI Trigger as a trigger signal for the PXIe-5172 Oscilloscope. On another channel of the Scope input a PPS from the Trimble GPS. Since the PPS from the NI-6683H is an internal trigger it cannot be seen on the trace. Time 0 is the mid-point of the plot. Plot shown below. Now with a different trigger there is roughly a 500ns difference between the pulses. This seems strange to me.

Why is the timing so different? It appears that the PPS on the PXI line is outputted before a whole second interval, why is it doing that? Can anyone explain what is going on and if so, is there anyway to calibrate this difference?

Thanks

Two independent PPS's from different GPS systems recorded by Instrument Studio. Trigger is edge of the blue pulse. A small and reasonable difference between the pulse, about 10ns.Use the NI-6683H to output a trigger pulse on a PXI line trigger. Now the PPS from the Trimble System is delayed about 500ns from the trigger.