Dave, Kai, Mike, Kevin, et al,
While writing the white paper on "how to guide unguidable mounts (:-)", something else just hit me like a ton of bricks.
Remember that sawtooth that describes the dynamics of guiding?
The amplitude of the sawtooth determines the limit of how well you can guide. If the sawtooth is 0.3 arcsec peak to peak, for example, you are not going to be able to guide better than that.
The leading edge of the sawtooth has a slope equal to the slope of the periodic error minus the guide rate.
Now, if the slope of your periodic error is less than say, 1.9 arcsec/second, you can use a slower (0.25x sidereal) guide rate and reduce the amplitude of that sawtooth! It almost certainly is; I have yet to come across a strain wave gear mount that is that poor. But you folks know your numbers, so you know for sure.
Recall that 0.5x sidereal rate is equal to 3.75 arcsec/sec. Your mounts' slopes probably come in at much smaller number, which means that you probably can choose 0.25x sidereal rate instead of 0.5x sidereal rate to produce a shallower sawtooth, which I superimposed as blue dash lines below to the original figure:
Notice that the dashed blue sawtooth is now flattened (and thus make the guiding even more gentle).
The Max Pulse duration is now double of what you had earlier tailored for 0.5x sidereal rate to match your mount's periodic error slope. But that's OK, since the mount is moving slower when you apply the pulse. The dither recovery is also going to be similar even though the guide rate is lower, because the max pulse duration is doubled.
I should have thought of this earlier, since this approaches the penultimate solution of my "guide by guide-rate" paradigm.
I won't have a chance to try it in the Pacific Northwest Rainforest for another couple of months (with my RST-135) when the rainy season comes to an end. Heck, it just dumped 2" of snow just yesterday morning (although not sticking).
Good luck, and smooth guiding.
Chen
