Good point Bill, thanks
Alignment stars should be separate in both RA and Dec, and should?also avoid near the poles (i.e., don't align on polaris)
One thing worth mentioning regarding model building (especially for the first two alignments) is to avoid using stars that are close in hour angle.? Here¡¯s excerpt from the Gemini 1 Level 4 User Manual.? I think the same thing applies to the Gemini 2. ? ¡°The first 3 alignment stars should be selected from Gemini's "Bright Stars" catalog
(database) and must differ in hour angle (distance in RA from the meridian) by at least
one or two hours. This means you either need to choose stars that differ in RA by at least
a couple hours, or wait a couple hours between alignments so that your next alignment
star will differ in hour angle from where your first alignment star was. For example,
doing an alignment on Capella and then on Rigel or Bellatrix is not a good idea; the same
is true for alignments on Procyon, followed by Castor or Pollux ¨C there are only tiny
differences in RA between them.¡± ? I typically try find two stars for the first two alignments that are at least 2 hours apart in hour angle.? My understanding is that when the two stars are close in hour angle, any errors in the model fit due to things like centering the object, etc. will get multiplied for later GoTos. ? Bill ? ? >>>I am very confident that the polar alignment is spot on.? Not only do I do a very careful PoleMaster alignment, but I never see any drift when viewing an object. ? That's great Mark. It helps when viewing, but as you pointed out, it doesn't really impact pointing accuracy, especially if you build enough?alignment points to compensate for polar misalignment. But if you are using polemaster, i probably wouldn't even bother with that many points ? On Wed, Aug 17, 2022 at 1:05 PM Mark de Regt <deregt@...> wrote: Brian, ? Thanks for this very thorough reply. ? I am very confident that the polar alignment is spot on.? Not only do I do a very careful PoleMaster alignment, but I never see any drift when viewing an object. ? I¡¯ll try your suggestion for how to create the model, and see how that goes. ? Mark ? ? Hi Mark I see a few problems with your approach, specifically the use of Sync instead of Align Here's my take on what should happen using your procedure, along with additional comments.? - Cold start at CWD position erases the model - polar align is important for good tracking (as always) - CWD position accuracy is really only for the first goto, as you pointed out. i wouldn't sweat it too hard - slew speed setting should not impact accuracy for gotos FIRST GOTO, Altair in eastern sky - no model means the first slew will be the least accurate, that's expected - Doesn't matter how you center the target: use slew/move/center, Although typically if the goto is farther out, you start the centering routine with slew, then move, then center, and perhaps guiding if you are fine tuning the position. There's no hard and fast rule here - Then Menu-> Align ->Align.? ?You don't need to add Sync at this point, you aren't syncing a model yet, you just started your first model point - "It worked very well!? Everything I told it to go to was reasonably well centered." WIth an Altair Align, you have one point model, so that's surprising it was that good but hey, who's arguing if it's accurate. If you needed to refine the goto accuracy for any of the subsequent goto targets in the eastern sky, you would goto, center (with hand controller, etc.) and then Menu->Align->Align. Typically a 3 point model is what I would consider a baseline, but i've often done 1 point as well. You can see this if you go to Menu->Align-> Model Parameters. It should show model terms for the first two (HA and DEC) only.? - this is where i think part of your issue is happening. The Sky Model is actually two separate models, one for the eastern sky, one for the western sky. If you go back to the Model Parameters and press Switch, you will see it Building a model on one side does not do anything for the opposite side. So we are back to the "first goto will be the least accurate" - Goto Mizar as the first western goto in an ampty?western sky model. This goto will therefore be as inaccurate as Altair, your first goto in the eastern sky.?? - At this point, you would repeat the same procedure for Altair, goto, center and ALIGN. Sync won't do anything at this point because there's no western sky model you are syncing to.? - I would also add two more Goto/center/aligns on this side, but at a minimum one Align.? NEXT - Gotos returning to eastern sky - at this point things should be fairly accurate on both sides - if you goto back to the eastern sky and it's slightly off, you would center and then SYNC to get your existing eastern sky model back in line with where the mount is pointing - why would eastern sky be slightly off when it was spot on earlier? equipment shift, polar misalignment (which is accounted for in the model only when you have 3+1 alignment points), mirror flop, ground shifting, etc. many things could impact the move from one side of the sky to the other. If you are using an SCT mirror flop can and will play a major role here - this will be the case anytime you switch sides of the sky I appreciate how important understanding this is, because I am doing planetary imaging with a miniscule fov (0.06 x 0.03 degrees) and a modest finder scope with a cmos camera attached (2.18"/pixel). I am not using any eyepieces so I can't just switch magnifications, it has to be really accurate to get it into the fov. With my Edge HD 11" my rough calculation of the difference in position when switching between sides of the sky is 2-4 arcmin? I'm estimating based on my memory of things but I will look closer next time. I attribute this primarily to equipment shifting and possibly mirror movement, even though I have mirror locks on the 11" So i would say if you give the above a go, i hope and expect things will improve for you ? On Wed, Aug 17, 2022 at 8:58 AM Mark de Regt <deregt@...> wrote: I went through all of the suggestions made, and collected those that seemed might possibly be connected to my problem.? Then I implemented them. ? Last night was a test. ? I carefully determined the CWD position.? Although, to be honest, there is absolutely no way I know of to know when the counterweight bar really is pointing as down as it can, since a bubble level depends on catching exactly the right part of the curve, and not moving it at all.? But I also understand that that matters only for the first slew. ? I slowed the speed at which the mount slews, to lessen the probability that anything skips with the sudden torque. ? I did a careful polar alignment with PoleMaster. ? I made sure that time, date, and place were all very accurate. ? I deleted any existing model. ? I sent the mount to Altair.? It missed by a fair amount.? I decided that maybe I should only ¡°center¡± using ¡°center¡± or ¡°guide,¡± not ¡°move¡± or ¡°slew,¡± so it took a while to center Altair.? After carefully centering Altair on my illuminated-crosshair eyepiece, I selected ¡°Menu,¡± then ¡°Align,¡± then ¡°synchronize.¡± ? I told it to go to Vega.? It was almost perfectly centered.? I centered, synced, then told it to go to Deneb. ? It was perfectly centered. ? The west side of my sky is pretty much destroyed with Seattle sky glow and trees.? So I decided to play on the east side, to see how it worked,?
It worked very well!? Everything I told it to go to was reasonably well centered. ? Then I told it to go to Mizar, very much on the west side, but easy to see. ? It missed by a lot.? I centered it, and synced. ? I told it to go back to Deneb.? It was on the eyepiece (17mm Nagler Type 4, on a Celestron 11¡± EdgeHD), but quite far off center.? I centered and synced.? Told it to go to M27.? It was centered pretty well,? I told it to go back to Mizar, and it barely was on the eyepiece. ? And so the night went.? I found no way to get it to converge on a model that got everything reasonably close to the center of the eyepiece (I wouldn¡¯t care so much, except that it¡¯s a royal PITA to center anything, to me, on a GEM, since I have no idea which of the four buttons to hit at any time to get it to move in any particular direction). ? It was somewhat promising, in that it always got the target on the (very, very wide) eyepiece, but it is far, far from perfect. ? Mark
--
--
-- Brian?
Brian Valente astro portfolio? portfolio astrobin?
|
