Improving PA and GOTO-The easy way?

[Merlin66]

While being eaten alive by the mozzies last night.....I found the alignment of my NEQ6 was a bit out and the GOTO accuracy marginal...
Then it came to me -
Assumptions:
1. Your time and location have been correctly and acurately entered into the Syscan (I'm still using V3.27)
2. You're happy with the SW capabilities - ie GOTO accuracy about one arc min.
3. Any "cone" error has been corrected. See SW Manual.
4. The mounting is "roughly" aligned to the pole.

a. OK. Connect the handcontroller to your PC (or use EQmod) and open your planetarium package (I use CdC).
b. Connect to your mount and make sure all the normal movement functions are operating.
c. Set the scope up to get a known bright star in the middle of the FOV (you can start by using the finder cross-wires).
d. Find the same star on CdC and sync to this star.
e. Then, using the CdC screen, locate another bright star - start with whatever is available....
f. Use the CdC "Slew" command to move the telescope to the second star.
Check the finder/ FOV to locate the target star -
g. Now, USING ONLY THE ALT/AZ adjusting bolts (Don't touch the RA/Dec) move the mount to bring the target star central.
h. Check by slewing back to the first star - repeat e/f/g as necessary (usually not!)

That's it - I feel this method is a bit like extreme drift alignment.
The first slew uses the location/time and sky cordinates to move the scope - if we were in perfect registration with the sky it would land on the target..by moving the mount in Alt/Az we are re-aligning the mount's axis to the sky...

Let me know what you think.

[Merlin66]

Based on feedback so far it looks very promising!!
Think of it as registering a piece of graph paper with a similar transparent graph paper overlay ( both in RA/ Dec)
One is the "actual" sky and the other the "mis-aligned" mounting...
We have one fixed point (the original "sync" star) and by using another distant star can determine how much the "transparent" overlay has to be rotated (in Alt/Az) to bring them into align.
Much easier than three star alignment etc etc - which can still show errors....

[Steffen]

I've used this method to get the mount into reasonable polar alignment in the field, too. However, it is not a replacement for 2- or 3-star alignment, the purpose of which is to inform the hand controller of the mount's home position and polar alignment error.

Cheers
Steffen.

[Merlin66]

Steffen,
What I like about this proposed method it doesn't rely on anything else..no problems with park position ( you can use that later) or any other "corrections"

[Steffen]

Yes, I find it useful since I don't have a fixed setup, and need to align every time before observing.

Cheers
Steffen.

[lazjen]

Another process you might like to try is using Astrotortilla + Planetarium package (I use Stellarium). Similar type assumptions apply: rough pole alignment, level mount helps, time, location set correctly.

AT gives you plate solving capability so you when you ask your planetarium package to point the scope somewhere, it adjusts and makes sure it really does go there. Plus AT has a nifty PA routine. For this you point East or West, and it measures your Alt error (then adjust appropriately) and repeat until happy; then you point to the Meridian and it measures your Az error (adjust again, etc).

AT can re-sync your scope, so for me at least, if I then use the hand controller to slew to somewhere else, it'll be reasonably accurate - but I can then refine by doing another AT plate solve anyway.

I also don't have a fixed setup and I found drift alignment to be tedious, so I'm very happy with this process (kudos to Rally for the idea, although he uses different software, the principle is basically the same).

[naskies]

Yep, I used to do a similar thing by first slewing to a star and loosening the clutches, centering the mount on the star, and then tightening the clutches. It's now redundant because I do an initial slew, plate solve the initial unknown position, and then add a single sync point to EQMOD.

The part about adjusting polar alignment after slewing between star pairs is what some of the software-assisted approaches such as AlignMaster and the SynScan handset do. However, they also separate out the alt vs az components of the error so you can adjust the mount more precisely (arc secs at a time).

I'm not sure how generally true this is, but I've personally found that if I start with a horizontally level mount/tripod, and slew between star pairs so that they are as far apart as possible (i.e. one star lowish on the horizon, the other high near the zenith) but on the same pier side, my PA reliably converges below 30'' in both alt/az within three iterations.

More advanced methods such as those used by PoleAlignMax and MaxPoint use plate solving of multiple points in the sky to calculate an average PA error, i.e. it takes into account flex in the mount. I've found AlignMaster more reliable for me than MaxPoint on my EQ6.

[Merlin66]

Dave et al,
I'm not sure I fully understand - loosening the clutches and re-centring doesn't change the alignment of the mount axis....
I assume somewhere along the line you apply Alt/Az corrections to the mount....

[naskies]

I only loosen the clutches only to perform a very rudimentary "one star alignment" to help with initial slewing to star pairs... but I don't need to do this anymore with plate solving.

Afterwards, its obviously adjustment using the PA bolts only. The software helps with separating out az from alt PA error because depending upon where your stars are in the sky, the error may be a contribution of either or both axes.

[rustigsmed]

hi dave does your method end up with you being able to take photos? sounds really simple!

Rusty

[Merlin66]

I think we may have to differentiate between EQ5/EQ6 running under Syscan V3.xx and the AZ/EQ6 and EQ8 running V3.35 (this ver does provide an Alt/Az error reading....)

[naskies]

Sorry, by "software" I should clarify that I mean AlignMaster controlled from my computer.

Yes, if I choose the star pair well (maximise the distance in both altitude and azimuth from each other) then I can start with my polar alignment severely out (e.g. +/- 10 degrees in azimuth) and within three iterations - slewing from one star to another, followed by corrections - I'll generally be within 30 arc sec of the SCP. This is sufficient for me to take 60 min subs using a KAF-8300M chip at 1625 mm focal length, and have no field rotation.

What I was trying to point out is that Ken's approach gets you closer to the SCP, but it can't separate small azimuth and altitude errors in polar alignment. This is because the conversion from celestial coordinates (RA/DEC) to terrestrial coordinates (alt/az) is not linear, i.e. moving in RA will usually result in both alt and az changes. Software-assisted approaches help by crunching the maths to determine the alt and az errors separately, so you know exactly how much your mount bolts need to be adjusted by.

[Merlin66]

Dave,
Understood....
In my method you don't need to know the alignment error....just be able, using the Alt/Az adjustments to re-centre a star in the FOV.....

[Merlin66]

To bring the required adjustments into perspective....
If the radius of action of the Alt/Az bolts (the distance from the mount axis to the centreline of the bolt) is 50mm, then one min arc would represent a movement at the bolt of only 0.015mm.
Let's assume the pitch of the bolt thread is 1.5mm, then a rotation of the bolt of 0.015/1.5 * 360 = 3.6 degrees would make the correction of one arc min......:eyepop: