Ad Astra Observatory

A,

an excerpt from
http://www.russellstreetobservatory.co.uk/equipment/G11pt4.html, a guy
whose demand is 5 minutes unguided at 1000mm. I don’t think he’s going

to get it, but he took great notes about his learning curve with the
G-11. Makes me want to start, too, except I’m pretty well over the hump
except for polar alignment.

His “part 3” page had his initial reactions to polar axis correction
(which is about where I’m at now), but by the time he wrote this, he was
getting pretty good at it, and was just needing to get the alignment

perfect for his observatory… (red emphasis mine)

14/05/2007 – the TODO list

It’s time to get the mount properly setup in the observatory. First up
is really good polar alignment, then getting the pointing model as good
as I can. Until now i’ve done a quick alignment with the polar ’scope

and maybe an attempt at Gemini’s Polar Align Correct routine, but it’s
time to get it right. Likewise, the pointing model has just been a few
‘additional aligns’. Then it’s finally time to tackle that 76-second
error…

22/5/2007 – Polar alignment and pointing model.

Five iterations through the Polar Axis Correction (PAC) routine reduced

my alignment error from almost 90 minutes off in both axes to an
impressive A:0 E:-1, which means that i’m very close to perfectly polar
aligned (the ‘worst case’ would be about 1.5 minutes of error, the ‘best
case’ would be a third of that). For each iteration I sync’d on Bootes
and then went back and forward across the sky for eight additional

aligns, taking in Vega, Unukalhai, Kornephoros, Mizar, Dubhe, Seginus,
Alphecca and Alderamin (not in that order!). By the end of the alignment
model Gemini’s estimate of polar misalignment had stabilized, and I
returned to Bootes for the PAC adjustment. After two iterations I was
within 10 arcminutes, which then dropped to five, three and finally one.
After the final iteration I ran roughly fifteen ‘additional aligns’ to
build up the pointing model.

The result of this is outstanding GOTO performance, better than I’ve
ever seen from the G11 before. That’s not too surprising; when I was
setting up each night I took some care over the polar alignment (within
10′, and usually within 5′) but I never bothered much about building a
pointing model because two or three ‘additional aligns’ was enough to

put the target somewhere on the CCD and I could centre it manually. Now
it puts a target star right the middle of a 5mm eyepiece wherever I go
on the sky. Impressive stuff.

Doing all that work took about an hour, and was a bit much for anything
other than a permanent setup (for field use without a polar ’scope I
think three iterations would have been fine, while with a polar ’scope I
used to get very good alignment with only a single PAC iteration). But,

for an observatory-mounted G11 it’s excellent.

29/5/2007 – Polar alignment and pointing model, part two.

I rebalanced the mount and switched from 21lb and 7lb counterweights to
a pair of 21lb weights, so had to redo the pointing model. This proved
to be a bit harder than last time, with Gemini producing some slightly
odd results. Polar alignment was still very close, but Polar Axis
Correct appeared to overshoot on the second iteration. More surprising

was that two runs through the list of eight additional aligns produced
different estimates of polar misalignment when a different order was
used (I did one set, got something like A:+5 E:-3, then reset the
alignment and did it again, which gave me A:+2 E:-1). Not sure why that
would happen, but it’s possible I’m not doing a sufficient number of
‘additional aligns’ for the model to converge?

After a slightly frustrating hour of work Gemini was once again

reporting A:0 E:-1, although I’m not sure how much I believe it. Polar
alignment is certainly very close, within five arc-minutes or so, but
I’m not sure exactly how close. Goto is brilliant again.

One thing I’m weirded out about is that he re-did his polar alignment
after changing counterweights. I would think that maybe you need to
re-do the pointing model, but no way you should have to redo the
alignment. The alignment is literally set in stone.

shrug. need to set up a blazing good pointing model and then start
running PAC until eyes bleed.

J