Note: groups.io will be down for maintenance on Wednesday, October 5th, starting at 9AM Pacific Time (4PM Wednesday October 5, 2022 UTC), for approximately one hour.
I don’t think it matters which pier side you start from.
As it is written now, PemPro does a continuous run, with no way of splitting it into two parts, so that we could do a half on each side, and merge the two sample files together to make say, a total 6 cycle file of raw data.
Ideally, the span of cycles should be equal on both sides, since when you are finally CCD imaging, the BEST photos result from a “span” of exposures, taken either side of the PM. You would be taking your Blue exposures before the prime meridian, and the Red ones after, where the blue in the sky diffusion tails off in the denser air. Thus, the best imaging is achieved through “planning exposures”, that span the PM.
Would be nice if the PEC curve an average matching both sides, as well. Then, your curve accounts not only for the somewhat stochastic pattern from the small spur gears inside the gearbox, it also includes a span of the worm and worm wheel contact irregularities, in the collected curve. End result is that it would account for more than just the gearbox effect, in the generated PEC curve.
However, another problem would arise if I were to do a PemPro run, “right through” the meridian.
If the scope is “TOO perfectly” balanced, there may be some very minor worm tooth “gear play” during the transition ... causing a slight lag as the mount passes PM. Or, there could be a “loading change” on the worm/worm wheel, and the gears in the box, if the OTA is (purposely) slightly biased, off-balance. Either one could (perhaps) cause PEC curve inaccuracies depending on which side of the pier you are shooting from.
It would be ideal to actually perform a Meridian Flip after half the number of cycles – then have PemPro “Resume” the remaining raw data collection cycles. But, that is not possible with the present program method.
Either way – running right through PM, or “resuming” after a flip, would produce data that may be closer to a true “average cycle” for the ideally used CCD imaging span, across the PM.
Next time, I think I will go for the continuous PemPro right across the PM, and “trust” that this provides better results.