[Shiraz (Ray)]

mine measured at 7 arcsec p-p in PECprep as well, with two regions of fairly rapid change. These guide out OK at my typical seeing and at 0.91 arc sec sampling.

attempted PEC as follows:
1. autoPEC - why not give it a go I thought... It generated a reasonable curve, but assumed an EQ6 worm period. I redefined for an EQ8, but the new auto curve had a phase error????.
2. then tried using PECprep with data out of phd (not phd2) - it has to be done exactly by the book. I got a good curve that was very consistent across 5 worm periods and almost removed PE entirely in the simulator, but it had a phase shift as well and I could not get it synchronised, even though I had done time stamping and parking. The phase shift slider did not allow shifting across the whole of the worm cycle, so something? was clearly not quite right.
3. Then both phd and phd2 stopped guiding properly - presumably because EQMOD was applying crap PEC as a carryover from an earlier attempt. disabling PEC did not help and I could only get it all back working by reloading everything.

when I get another marginal night and bit more courage, will try again with PECprep, but not with autoPEC. I don't think it is actually too difficult, but everything has to be done exactly by the book. Rate based PEC is in principle a great idea and should clean up the PE very effectively. In the meantime, I get about 0.6 arcsec RMS guide error at 1 sec update on a good night, which is fairly good anyway.