Polar Alignment - How close is good enough for guiding to compensate

[chris_g]

I was working with N.I.N.A.'s TPPA last night and wanted to know, how good is good enough to let guiding take over any deviations? I do know that I would probably never get quite as close as I did with TPPA last night with just my Polar scope.

[Juno16]

Hi Chris,

It will be interesting to see what other folks say, but for me I just try to get it close. By close I mean under five arc minutes.
Of course, I’ll shoot for close to zero, but I always run TPPA or Sharpcap from the east side of the mount.
When I review the PHD2 guide logs the next day, my PA is always higher on the west side of the mount. Sometimes as high as 8 or 10 arcmin. Doesn’t seem to affect the guiding (or star quality) at all. My guiding total rms error is usually similar before and after the flip.
No software is perfect. It’s just designed to get you close, which is all you need to be.

[chris_g]

Hi Chris,

It will be interesting to see what other folks say, but for me I just try to get it close. By close I mean under five arc minutes.
Of course, I’ll shoot for close to zero, but I always run TPPA or Sharpcap from the east side of the mount.
When I review the PHD2 guide logs the next day, my PA is always higher on the west side of the mount. Sometimes as high as 8 or 10 arcmin. Doesn’t seem to affect the guiding (or star quality) at all. My guiding total rms error is usually similar before and after the flip.
No software is perfect. It’s just designed to get you close, which is all you need to be.

That also answers a question I had but raises another. The screen shot was from when I did the TPPA routine starting from Jupiter, on the south side. When I reran the process after parking the scope at home, it was way off. So I should start from the north east closer to the NCP instead of the south east?

[Juno16]

Really not sure Chris.

You could always look up the N.I.N.A.documentation or better you get on Discord and ask. It would be a good idea to get familiar with Discord in case you have any issues with N.I. N.A.

By the way, how much is way off?

[chris_g]

Really not sure Chris.

You could always look up the N.I.N.A.documentation or better you get on Discord and ask. It would be a good idea to get familiar with Discord in case you have any issues with N.I. N.A.

By the way, how much is way off?

Double digit arc minutes in the teens in the RA axis and it was off in the Dec axis as well though not as much, I should have screen capped it for a comparison. It was also after I stopped an inadvertent meridian flip. Of course I don't really know what way off really is yet

I'm on Discord for a game I ply, I'll join the group for NINA

[Juno16]

Double digit arc minutes in the teens in the RA axis and it was off in the Dec axis as well though not as much, I should have screen capped it for a comparison. It was also after I stopped an inadvertent meridian flip. Of course I don't really know what way off really is yet

I'm on Disqus for a game I ply, I'll join the group for NINA

That is quite a bit. In my limited experience with TPPA (and also with Sharpcap PA), if I repeat the process, I might be off up to 1 arcmin, but usually, it is very close to the first TPPA run.

Since it only takes a minute or so to run the routine, if I see a discrepancy (>1 arcmin) between runs, I will repeat until they agree very closely (<1 arcmin).

I think that the biggest discrepancy that I have seen between TPPA runs has been 1 or 2 orcmin, but as I mentioned above, successive runs are usually much closer than that.

Please share what you learn on Discord.

[chris_g]

Double digit arc minutes in the teens in the RA axis and it was off in the Dec axis as well though not as much, I should have screen capped it for a comparison. It was also after I stopped an inadvertent meridian flip. Of course I don't really know what way off really is yet

I'm on Disqus for a game I ply, I'll join the group for NINA

That is quite a bit. In my limited experience with TPPA (and also with Sharpcap PA), if I repeat the process, I might be off up to 1 arcmin, but usually, it is very close to the first TPPA run.

Since it only takes a minute or so to run the routine, if I see a discrepancy (>1 arcmin) between runs, I will repeat until they agree very closely (<1 arcmin).

I think that the biggest discrepancy that I have seen between TPPA runs has been 1 or 2 orcmin, but as I mentioned above, successive runs are usually much closer than that.

Please share what you learn on Discord.

It was probably the flip, I did stop it as it started to push my guide scope against the tripod. I stopped it, parked the scope and reinitialized everything. I'm going back out tonight and will run it again and see what I come up with. If anything is really off, I'll post it there...

[Juno16]

It’s kind of good that you guide scope crashed into the tripod! Just kidding, but it might explain why PA was off the second time.
Good luck tonight. If you don’t mind me asking why are you using your guide scope and camera to polar align instead of your main imaging scope?

[OzEclipse]

I just use the optical polar finder and usually get it better than 2 minutes. But I live in the southern hemisphere so using the optical finder is much easier. There are two stars a few minutes from the pole that make a nice little equilateral triangle with the SCP.

[SkyHiker]

Let's compare the effects of periodic error and PA error because both errors must be guided out.

Intermediate and beginner telescope mounts usually have a periodic error of 10" amplitude over a period of 2 to 4 minutes, say, 5" to 2.5" per minute. If the PA is off by 10' or 600" this leads to a worst case of 600" per 6 hours or 100" per 60 minutes or 1.6" per minute. This is less than the periodic error. So if a guide scope can guide out periodic error, it can easily deal with 10' PA error.

If TPPA gives you trouble then just do a one time drift alignment with PHD2 and mark the location of the feet. In my experience it is easy to reposition the mount with 10' accuracy. So if you have a guide scope, this is what I would do.

BTW You can use adaptive PEC to compensate for periodic error as well. I tried this in Ekos and I believe it helps while it does not make a major difference. But I achieved 0.7" total RMS several times just plunking the mount back in place where I marked the feet without redoing the PA so it can work quite well.

[SkyHiker]

Let's compare the effects of periodic error and PA error because both errors must be guided out.

Intermediate and beginner telescope mounts usually have a periodic error of 10" amplitude over a period of 2 to 4 minutes, say, 5" to 2.5" per minute. If the PA is off by 10' or 600" this leads to a worst case of 600" per 6 hours or 100" per 60 minutes or 1.6" per minute. This is less than the periodic error. So if a guide scope can guide out periodic error, it can easily deal with 10' PA error.

If TPPA gives you trouble then just do a one time drift alignment with PHD2 and mark the location of the feet. In my experience it is easy to reposition the mount with 10' accuracy. So if you have a guide scope, this is what I would do.

BTW You can use adaptive PEC to compensate for periodic error as well. I tried this in Ekos and I believe it helps while it does not make a major difference. But I achieved 0.7" total RMS several times just plunking the mount back in place where I marked the feet without redoing the PA so it can work quite well.

The PE is actually worse because the 10" in the example has 4 phases within the 2 to 4 minute timeframe so if it were a triangular wave it would be 20" to 10" per minute. In that case the PA can still be a whole lot more off before it becomes a problem relative ro PE. It looks like 30' should still easily be OK maybe even a degree. Adaptive PEC should also work for PA errors because it is a near constant error, which is periodic.

Aside from the guiding RMS there's also field rotation, which is not included in the above. If the PA were off by 1 degree the FR would be 1 degree per 24 hours. For an image with a 2 degree FOV the worst case would be in the corners at 1 degree per 6 hours, which is 1" per 6 seconds. Medium / small bright stars on an AP image are about 10" to 20" in diameter (for a color camera) so you want to limit your exposure to 60 seconds at the most before it starts ruining the image. A mono camera has much smaller stars so in that case the 1 degree PA ruins the image much quicker. Typically the exposures are much longer too in that case so for mono, PA should not be more than 10' off or so, guessing without redoing the math but these calculations are pretty simple.

[OzEclipse]

Chris,

You might also find this calculator useful in giving you an idea of the magnitude of drift caused by varying amounts of misalignment.

http://celestialwonders.com/tools/polarErrorCalc.html

Joe

[chris_g]

That's awesome, I love the math! While I am not formally trained as a mathematician, I crunch numbers, design databases and design formulas to develop custom reporting for the company I work for. Thanks to both you and @SkyHiker

Chris,

You might also find this calculator useful in giving you an idea of the magnitude of drift caused by varying amounts of misalignment.

http://celestialwonders.com/tools/polarErrorCalc.html

Joe

[chris_g]

I laid out my pathway stones and shimmed them to .1 of an inch for every foot with my phone's pitch and roll indicating 0, 0 with the tripod legs fully extended, it isn't going to get any leveler than that. I also positioned one of the northern facing stone corners as close to due north as I could get. Before I fired up N.I.NA. I slewed to Jupiter and it was within the FOV of my red dot finder instead of being way out of whack, I only had to make a few adjustments to get it within my imaging camera's FOV to begin the focusing so N.I.N.A. could plate solve. I fired up TPPA and got to the image below in about 5 minutes, most of that time was waiting for N.IN.A. to plate solve. It was so much easier to achieve by making adjustments in one axis and not have the other go off. I didn't have the permanent marker until close to the end of the evening, so the tripod positioning on the stones will be marked today when I set up tonight. Polar alignment was a major obstacle for me to overcome and with that done I am almost at the zenith of my learning curve. Thanks for the help and advice from everyone...

Update: I just noted the bottom numbers are where I started from, so in about 5 minutes I got it from over just over a 143 minute total error to 41 seconds. I updated to the latest nightly build yesterday, another nice feature of TPPA.