- these glitches are clearly seen in the current monitor of NE while these are not seen in WE monitoring (see first plot).
- a zoom (2nd plot) on one of these glitches shows that the NE z correction is constant on 2 consecutive bins while the WE is not. Either Gc sends twice the same number either the DSP miss 1 sample: it is not pssible to know since the correction sent by Gc were not (yet!) in the raw data...
- 3rd plot shows the FFTof the signals during this period (purple) compared to normal condition (black).
It is probably not related but one can also see that there is a large 10kHz component in the correction signals (4th plot). This is probably the same problem as in the Gc alignment signals (big line at 250Hz) reported in 250Hz peak in Gc_Al_CommonEnd.
1/ the big line added by Gc is at 5kHz, not 10kHz
2/ the 2 problems are related: without the 5kHz oscillation, missing 1 sample would have a much smaller impact