TSX Accurate Polar Alignment... What am I doing wrong?

[Atmos]

Taken me a few attempts to get used to new software, mount, telescope and camera. Taken on everything all at once haha

Was doing my first T-Point iteration last night (~60 stars) and did the Supermodel. It said that I had to move ~105 arc minutes East and ~30 arc minutes up. Started the accurate polar alignment routine, slew to Acrux, centred it and then pressed "Commit". It made this massive slew (relative) wanting me to lower the mount ~10 degrees and move ~15 degrees in azimuth. I started on it thinking it knew what it was doing, went this is stupid, pressed Done, parked the telescope and then slew back to Acrux l, re-centred Acrux.

Did another 60ish routine and got down to ~3 arc min below and 12 arc min East. Did the Accurate polar alignment routine again, over corrected the elevation (5 arc min overshoot) and had me move ~12 arc min West instead of East.
I found the above out after doing another 60ish model, had to move 5.5 arc min down and 24 arc min East. Just ended up guessing it and left it there.

What am I doing wrong?

[gregbradley]

Hi Colin,

Firstly congrats on setting everything up! And getting TPoint up and running and plate solving. That took me a while. The key being having the accurate image scale.

I take it you did the rough polar alignment first as in the manual? If not do that first to get you close.

So you do your 60 point tpoint run. You click on super model. It comes up with a better RMS (average of the errors). You click apply to apply the supermodel and watch the new RMS figure go lower.

The Polar Alignment report gives you 2 options. How much you need to adjust your RA and Dec or use accurate polar alignment.

So you clicked on accurate polar alignment. It verifies you have done the super model. Then it asks you to slew to a star in the north or the south and gives a little percentage of the existing location. As you move you will see that % appropriate change. I go for about 95%. I usually point to a low star in the north as its easier for my observatory.

Once you have a bright star centred at this location you click commit. The software now moves the mount so that same bright star needs to be adjusted manually to be brought back to the centre cross hairs of your image (you need to activate centre cross hairs so you can see that is exactly centre).

Now adjust the RA and Dec adjustment knobs taking new short exposure to see where it is until you have manually adjusted the mount so that same bright star above is now centred again in the cross hairs. Now click done and the Tpoint model takes into account the physical change you just did.

You should now be very well polar aligned. It should be good enough to image but if you want to take it further perhaps do a 200 or 300 point model and repeat and will get it that little bit closer. 60 point may not be enough to get a really accurate accurate polar alignment but if you watch the polar alignment report which updates as it adds more points it probably won't change much with more points but it does a bit.

I also use 2x2 binning when centring the bright star but I have used 1x1 for that little bit extra accuracy. I would not use 3x3 as it would be too sloppy.


I would do that on another night so you can get some imaging done.

The accurate polar alignment feature is a terrific improvement to the TPoint program.

Greg.

[Atmos]

Thanks for that comprehensive explanation Greg. It sounds like I've done it all correctly, just been following the steps that the Accurate Polar Alignment routine gives. The problem that I ran into is that on the first iteration (slew away from the star) it was HUGE. It should have been out of my cameras FOV but not by too much, instead, T-Point moved approx. two Crux's away. Then the second iteration moved in the wrong direction along the azimuth axis (went from ~12 to 24 arc min away).

I think next clear night what I'll do is push Commit so that it slews away, press Done and THEN go and centre the star. It worked on the first iteration, started moving towards its suggested location and then went "screw that", pushed Done, parked the telescope (to contemplate my next move) and then slewed back to Acrux and centred from there which got me pretty close (~3-10 arc min on both axis, that was on 3x3 and not 100% accurate mind you, wasn't sure what was going to happen so I wasn't too stringent).

Thanks for the reply Greg, just wanted to double check that I hadn't gone an forgotten some silly little parameter that I NEEDED to fill out for it to work. I'll centre it after it thinks it has been done :P

[gregbradley]

Did you do the rough polar alignment first? It sounds like your PA is way off.

I would do a rough polar alignment first.

Then a 30 point t-point model and do the adjustments per the polar alignment report.

Then do a 100 point t-point model and do the accurate polar alignment. Then you will find the bright star only moves away from the cross hairs a small amount and you bring it back, centre it manually adjusting the mount and you should be very close now.

You could finish off with a 200 or 300 point model and then another accurate polar alignment and it should be another smallish adjustment and you will be really accurate.

If you activate tracking corrections in Protrack you may find it will improve star roundness in your autoguided images.

I think you are jumping too quickly ahead on the polar alignment process. Its best done with several iterations.Not one.

Greg.

[Atmos]

Both the Rough Polar Alignment and ProTrack are for SB mounts only I believe. The Rough Polar Alignment uses its homing sensors, something that the EQ6 doesn't have :P

What really threw me was that on my second iteration it didn't move far (was only ~3 arc min down and ~12 arc min East). When I went to move my mount East I found that TSX had moved in the opposite direction that I thought it should have. At first I thought I must have just read it wrong but when I did my third run I found that I was in fact ~24 arc min too far too the West (needed to go East), so it moved the correct distance but in the wrong direction from what I can tell!

Autoguiding is my next step to figure out, never done it before. Tried doing it in MaxIm DL a couple of times when connected to TSX but it never seems to be sending the "move" signal for calibration. I THINK I need to have it as "Telescope" and not connected directly to the ASCOM TheSkyX thing. More testing :)
Tried using TheSkyX to auto guide, just kept spitting out errors during calibration. I probably haven't filled out all of the information for it though, I don't think I have ever put my guider scope information and stuff in so that could be why.

[peter_4059]

or you could drift align :screwy:

[PRejto]

I don't know the ins and outs of TSX when not using a Bisque mount. But I think the accurate PA is going to work quite well when you are already pretty close to good PA. If the star moves too far it will be out of the camera FOV and I'm unsure how you could go about re-centering it. When I've used APA the star move was well within the FOV.

The other issue is how accurate your mount adjusters are? Even on the Bisque mounts they are not perfect, especially in MA. ME is pretty good.

You might want to try using the jog control in TSX and see if you can measure how accurately your mount responds to arc-sec moves. You might be dealing with some backlash which would make the jog control difficult to use for PA. But, you might try something I used to do to get extremely accurate adjustment in ME. You do the supermodel and see what it says for ME. Perhaps it say lower your mount by 1 arc-min. What I've done is slew to any star on the meridian and center it. Then use the jog control to slew 1 arc-min higher. Then lower the mount until the original star is re-centered. You have lowered the mount exactly 1 arc-min! The same technique can be used for MA but is much more tricky because the move is by the cosine of the altitude. The best position is probably pointing east or west.

Personally I think you will find T-Point is going to give you very accurate pointing. But, it may prove less satisfying for PA given your non-Paramount. If I were you I would use PEMpro to drift align. Once you are satisfied with PA then run your model and ignore the polar alignment report. You will get a very good result. (Extremely accurate PA is a compromise at best anyway so there is nothing wrong with a drift alignment.) Also, realize that even if you are not permanently setup you can reuse the model by just recalibrating into the large model (maybe only 8-10 points are necessary). You just need to mount your equipment as close as you can from night to night.

Peter

[allan gould]

Colin
You really appear to be dressing up an "Intermediate" mount with a lot of software/features that is/are not necessary.
EQMOD, CDC and PhD guiding along with SGPro is all you need to be fully functional for imaging and finding your way around the sky.
Some of the programs you are using are crippled by their origins ie why do you need a Tpoint model on temporary setup? Tried it with The Sky and it was a waste of time if you are polar aligned and have plate solving. All the programs I mentioned are well supported and free (except for SGP).

Anyway that's just my 2 cents but enjoy the ins and outs of the system you are trying to setup with the EQ6.
Allan

[gregbradley]

I use Sky X and Tpoint with an AP mount and it works just fine no different really to a Paramount mount.

If I were doing a temporary setup I might consider using PemPro polar alignment wizard as it works in real time and is also very accurate and fairly fast. If you make an error you can see it in real time worsening results and you go the other way. Kind of simple. Its also very bug free mature software. I think it even tests for backlash.

Perhaps your mount as has some backlash as Peter pointed out and that is throwing you a problem (check these other items below first though).

The advantage though of the Sky X and Tpoint is if you get past the learning curve I think you could be polar aligned very accurately quite fast.

If you use a digital inclinometer (cheap from ebay) and use that to set your angle of the mount (you will now be very close) and then set the mount to approximately due south you again will be quite close.

Now do a 15 point model and do the adjustments then another and then maybe a 60 point model and the accurate polar aligment routine and you are done. Perhaps only 40 minutes.

Pempro polar alignment wizard may only take 30 minutes. There is a Polar Align Max but I haven't used it.

Autoguiding in Sky X. What guide camera are you using? Did you connect to it using the autoguider tab?

The guide camera ideally is square to the mount so make sure its not at a weird angle. Sky X will compensate if it is but its better for it to be square, its easy enough.

Autoguider settings: This will depend on your mount and seeing to some degree but lets start with 3 second guide exposures, 2x2 binning, set it to relays (you need an autoguiding cable to go from the camera to the mount)
I use pulse but I am using an AP mount. I am not sure if that works on other mounts.It may and you don't need an autoguiding cable so its convenient. Min move .01 maximum move .5.

Backlash settings. Your mount may have backlash, mine don't so you may be best to ask others what they set backlash to for EQ6.

Guide stars. Make sure your guide camera is well focused. The idea it does not matter if the star is a bit out of focus is false.

Pick a medium bright star that looks nice and sharp and has no double star and no brighter star nearby that the software could confuse as the guide star. You can also use auto select.

Callibration. Take a 5 second exposure at 2x2 binning. Pick a medium bright star that is not a double star. Click on the framing icon above the photo window. Drag it across the star to form a reasonable sized box (not too small and not so large it include other bright stars which can throw off the calibration as it simply picks the brightest star so if another bright star moves into the frame it will erroneously pick that instead of your guide star).

Set time for calibration. About 5 seconds is about right perhaps a tad more. Too much and it will move the star out of the frame. Too little and it may not move enough for a good measurement.

Exposure time for the guide camera: This varies with mounts. A poor mount may need shorter exposure times to keep up with the errors. A better one takes longer exposures. I have run different exposure lengths with different mounts from 1 second all the way t0 10 seconds.

Also autoguiding will not work well if polar alignment is off by too much. So the first thing if autoguiding is not giving good results is to try another guide star (it can make a massive difference) if still no go then tweak polar alignment.

There is a fantastic autoguiding interview from one of the software guys who works on PHD2. Watch that and you'll get a fabulous education on autoguiding. Its the best resource I have seen on autoguiding. PHD2 gives superior autoguiding to Sky X and its free.

Autoguiding is a bit of a black art so feel free to ask questions.

It sounds like you don' have your guide camera attached to your mount via an autoguiding cable. If so make sure its connected and you can take images. If you are then make sure the autoguiding method (autoguider tab/ setup/there is menu with the different methods of autoguiding direct guide (Paramount mounts only) pulse (the one I use but depends on your mount?) relays (this requires an autoguiding cable and is the common one used).

Also I use auto darks on my autoguider. Hot pixels on noisy guide cameras can be confused for the guide star and will make callibration fail and guiding fail at times.

Feel free to ask more questions.

Greg.

[peter_4059]

I've never tried this approach but it seems very complicated for what you are trying to achieve - polar alignment and accurate pointing with a temporary setup of an EQ6.

PHD2 has an excellent drift alignment wizard that achieves very accurate alignment quite quickly and it is free. It will probably take you 15 minutes to read the instructions and 30 minutes to achieve alignment the first time out. The most complicated bit is adjusting the EQ6 altitude as the mechanism doesn't offer very fine control and tends to overshoot. I purchased Pempro for drift alignment but actually prefer PHD2 as it is far easier to use.

For pointing accuracy there are various free platesolving solutions that achieve excellent results with almost no setup once the image scale is entered. I no longer bother creating a 2/3 star pointing model for my EQ6 - just slew to approximate location, plate solve and one more slew and the object is centred on the CCD.

Creating a Tpoint model with 10's/100's of stars for a temporary setup with a mass produced mount seems like overkill. In my trade we'd call that "polishing a turd".

Just my thoughts.

[PRejto]

The only thing I don't entirely agree with the other Peter (hope you don't mind!) is that if you make a quite large model with T-Point it can easily be used from night to night if the equipment setup remains the same. And, T-Point will work very well after a drift align with just a small number of recalibration points. I think using T-Point for polar alignment will be very difficult on a mount that doesn't have accurate adjusters.

Peter

[Atmos]

Thank you everyone for your input. Going to have another read through it all and properly process it when I am not dog tired!

In the interim however, I did start off with using EQAlign as a way of doing the drift alignments. Free software and it does work really well, doesn't have the "real time" feature that PemPro does but it works well all the same. I have seriously considered purchasing PemPro and may still do it in the future, even if just as a way of getting some damn good PEC curves :P

I do agree with the sentiment that I am probably going a little over the top for a lowly EQ6 Pro :P It doesn't have the most accurate encoders, many seem to suffer with backlash (although I haven't really noticed any yet with my limited usage). Not say it's not there, just haven't pushed it far yet.

There are a couple of reason I did choose to go down the path of TSX over PemPro for alignments though. Using TPoint takes an all sky model, takes refraction out of the equation. For a temporary setup it is debatable as to whether the difference is worth the effort, I do agree with that. For what I am planning on doing, I do need pointing accuracy though. When I get everything up and running I plan to start doing astrometric studies on a few dozen globular clusters over several months. I have an interest in RR Lyrae stars and want to do some research into the variability of variable stars (they have a ~40 day cycle). This involves slewing from one cluster to another, taking one image and then moving on. Accurate pointing makes my life much easier to say the least.

I am also planning on having a semi-permanent setup both at my house in Melbourne and my dark site in Heathcote. When they get established, having strong TPoint models already done means I'll only have to do a resync to them, 15 minutes and I am up and running :)

Right now my mission is to work on one aspect at a time. First it is getting to know TSX, then auto guiding, then sequencing. Sequencing will probably be done through SGP running with TSX for mount control. As much as I love MaxIm, I haven't sat down to learn Visual Basic to write sequencing scripts for it yet. Maybe in the future!

I know that within about 40 minutes from power up I should be able to be pretty accurately polar aligned (enough for me to be happy with a temp setup), have a 60-100 TPoint model and be ready to go. I have a digital level which gets me quite easily to within half a degree and I am looking at making a mount attachment for my compass to get me within a degree from cold setup.

[PRejto]

Colin,

I'm sure you know this but as long as T-Point knows where it is at you can have pretty sucky polar alignment together with extremely accurate pointing. If your exposures are going to be short pointing can take priority.

Peter

[troypiggo]

I've tried and used EQAlign a while ago. OpenPHD/PHD2 is also free, and it's drift align tool is realtime. It's also much easier and quicker than EQAlign in my opinion.



Accuracy of encoders and backlash has nothing really to do with it. As per my comment to Peter below, if you have accurate polar alignment, everything else falls into place and is easier - pointing, guiding etc. You can't have pinpoint stars without accurate polar alignment unless your exposure times are unrealistically short for the sort of gear I understand you're using.

PHD2 (free) has a very simple and quick drift align tool for accurate polar alignment. You'll be polar aligned in minutes.

Astrotortilla (free), Elbrus (free), PlateSolve2 (free), Astronomy.net local server (free) are all plate solving tools that will solve an image taken in any part of the sky and sync your mount.



If you're talking about using SGP, all of my comments above are even more relevant, as they all work seamlessly with SGP. You tell it you want to go to a target, it will slew there, automatically take an image, plate solve it, if you're not bang on it will adjust and autoslew to target, solve again just to check, and off you go. Seamless, simple, quick, accurate.



Forget the TPoint model and you can be ready to go in less than half that time, without losing any accuracy as per comment above ;)



Surely for imaging an accurate polar alignment should be priority number one? If that's bang on, everything else falls into place - your pointing is more accurate, your guiding is easier, you don't get field rotation etc.

[rustigsmed]

yep for a temporary setup that's probably not worth it.

I can do 30 min guided subs with just the handset synscan polar alignment routine and PHD2 FL at 1200mm.

It probably takes 4-5 PA routine iterations to achieve this... depending how far off PA is to begin with, so not too long at all. :thumbsup:

[Octane]

+1 for PEMPro. It makes it so easy.

Will see if PHD2 supports an STL-11000M and give that a go, too.

H

[allan gould]

Ive used pempro, EQalign, Tpoint, WCS by Wolfgang Ruthner and helped Andreas set up Alignmaster for the Southern Hemisphere.
PhD2 knocks all of these into a cocked hat. I used it for the first time at Astrofest on the recommendation of a few people whose opinions I respect.

With no instructions, I was polar aligned in 10 min top and it stayed that way for the 10 days with perfect guiding. I wont be using anything else.
Allan

[gregbradley]

You're getting some good advice here. I haven't used the PHD2 drift align tool but the guiding graph does show a trendline just like the Pempro one does.

Just as a point though, I have used Pempro polar alignment wizard a few times (especially when TPoint was having trouble with plate solving etc, it doesn't like some cameras like the SX Trius694 for example).

Tpoint polar alignments using the accurate polar alignment button has always been more accurate and gave better guiding.

Additionally, not really relevant for a temp setup, but if you have a decent Tpoint model you can activate tracking corrections using Protrack and you can find sometimes it makes the difference between slightly eggy stars and round ones. It corrects for the slow flexes and autoguiding corrects for the PE errors.

Greg

[Atmos]

There is a LOT of information for me contemplate here, thanks everyone for your input. Just finished reading the PHD2 manual, the Drift Align tool looks pretty good, will have to give that a go on the next clear night.

Should probably go and reread this thread again :)

[bugeater]

I use the PHD2 drift align tool exclusively as well and have found it pretty easy to use. I do take a bit longer than others here to get aligned though. I find the charts will follow a trend for a while and then change direction, so you aren't entirely sure whether you've got it spot on or not. Not sure why this happens. Wind perhaps?

[Atmos]

As someone very new to this game feel free to ignore me ;) but it sounds like potentially something on your worm gear? If you're getting a sudden change then it has to be something sudden and pretty nasty causing it I would imagine.

[troypiggo]

Depending on how much backlash etc in your system, early on the trend line can change direction a bit. The more time you give it, the more data it has to settle down and get more accurate. 30secs might be enough. If doing longer narrowband exposures you might leave it 2mins.

[gregbradley]

Its the same with a high end mount perhaps not as much shifting. Any mount has PE. It tends to look like a sine wave when smoothed right down with wiggles within the sine wave.

So a system that works on a trend line could at first show part of the curve up or down of the PE and it would be a false reading.

Only after its been averaged for a while would that become more accurate.

Pempro works the same with a trend line being formed on the basis of the plotted guide errors. The instructions for that clearly state it may need to run for a little while to determine an accurate trend line. However its pretty clear if you are a long way off. To get it so the trendline is level (the goal) it has to be very close to correct PA.

I haven't used PHD2 PA but I see the trendlines available in the guide error graphs. It may even be faster than Pempro if it shows both RA and Dec trendlines at the same time if that is possible without slewing to different parts of the sky.

Greg.

[bugeater]

That was my initial theory, but it's the drift in dec you are watching, not the RA. So PE shouldn't matter. At least that's what I was told and it makes sense.



I have done some work on reducing the backlash, but haven't seen much change really. But yes, if you leave it longer, you should average out any non-polar alignment errors. I often leave it for many minutes before tweaking it (I'll go inside and watch TV or something).

[alistairsam]

Have you installed the Daily build update for TSX?
Software bugs are probably why tsx slewed so much.
I had a night where I have a myt, freshly installed tsx, had all sorts of issues. would not even home properly, something that cannot go wrong.
after 2 hrs, installed the daily update, "everything" worked from then on. couldn't believe it.
unless you've already done this, then ignore.

I've been using tsx with a paramount myt and find the PA routine extremely accurate. more than sufficient for 40min subs with 2 secs guiding on a 1m scope.

I make sure the first slew to a star has the star in the fov.
then a 24 point run, difference here is that it tells me how many ticks to turn each knob. after 3 x 24 point runs I don't need to adjust it any more according to it. and I clear the model every time I do a run.
This has worked for me in a field setup every time.

good luck.

Cheers
Alistair

[LewisM]

I use a "lowly" GPD2 mount, fitted with a Synscan GOTO system. If I do multiple iterative 2 star alignments, manually moving alt and az in between with the knobs using the reported alignment errors as a guide (NOT using the useless Synscan Polar Realignment feature), I can get the mount down to 30 sec or less out in both axes - depends how lazy I feel on the night. I usually aim for sub-10sec out maximum.

I then autoguide, and find, if I did a good iterative alignment, that I will get a guide correction every 7 to 8 minutes, the remainder of the time it just tracks. 30 minute subs no issue.

That is using either PHD2 or MaxIM (preferring PHD2 at the moment) for guiding, MaxIM for image acquisition and just the bog standard Synscan handset as "the brain".

Never once tried controlling via ASCOM, though it is all there if I want to.

[Peter.M]

The gpd2 has a worm period of 10 minutes, having the guider issue one correction every 7 minutes would be unwise unless the periodic error of the mount was considerably better than the local seeing. Even if the mount was performing at 5 arc seconds peak to peak which would be outstanding it should be issuing more corrections than that.

[LewisM]

I dunno Peter - that's all MaxIM seems to do - it continually checks, but the movements readouts are saying 0.00 in both axes, except once every 7 to 8 minutes when I see it make a small correction.

PHD2 keeps the centroid continually within 0.5, usually MUCH less (0.25 or less). RMS is VERY small with this mount.

I dunno the settings - it just works! I let PHD2 decide what, and I use default in MaxIm.

I have had this mount completely apart - polished worms and spurs, regreased, fiddled with meshing for hours. it just plain works right, with stuff all effort. I had to tear it down again maybe 3 months ago after NIL use for 4 months, as the first time out I found slackness in Dec and RA - hence the remesh etc.

I will try to make a graph tonight IF the weather holds - the dec and RA traces are essentially flat, on the smallest measurement scale.

[peter_4059]

Might be guiding on a hot pixel.

[Octane]

Lewis,

What are your min/max move settings in MaxIm DL? That will explain why you are only guiding once every few minutes.

H

[troypiggo]

Could it be that your guide camera is set to 7min subs? :poke:

[LewisM]

Probably a hot pixel, and the tracking is so faultless it don't matter none :)

H, I will look it up later when on the astro comp, but I have been using PHD2 instead lately - bloody good stuff, SO much better than PHD1.x!

There is something about this GPD2 - it's been sold twice, and I bought it back each time (just ask JRC :) He's since bought the LAST one out of Germany ) when I found other more expensive mounts didn't come CLOSE to this one - I originally bought it from Kunama, who bought it new, and he can attest too to how darned good this one is. Just incredible really, and light enough to be portable (seeing I have about a 2 hour window on most targets, transportability is paramount to me)

Anyway, the GPD2 is good only up to an FSQ-106 maximum - I doubt Colin's 130 would be any good on it, let alone any RC over say 8". It's pretty much a refractor mount, or light Newt/Cat (R200SS / VC200L)

[gregbradley]

Sounds pretty amazing. Your mods must have enhanced it. I have heard GPD2 is an accurate mount though. Although mostly zero errors I find hard to believe(do you use auto darks?). No errors does sound like your settings are unusual or a hot pixel which of course does not move is being used. But if you get round stars 30 minutes who can argue? That is a stunning result no matter how high end your gear is.

AP1600 has the best tracking of PMX, PME, Tak NJP, MI250, Vixen Sphinx and I get minor corrections every 10 seconds with perhaps no correction on half the errors.

I often get about .9 to 1 arc second PE which is the best I have seen. PMX was about 1.5-1.8, not sure what NJP was as no graph but probably 3 to 6, MI250 was probably about 6, PME is about 1.5. Sphinx was about 30!

PHD2 guide assistant will make recommendations for min/max move and length of guide exposures plus give a rough idea of how accurate your polar alignment is.

I am imaging at 1159mm and 40 to 43kgs of weight. The PME is carrying around 80kgs.

Usual seeing is around 2.5 to 3.5 arc secs judging by the guide camera guide star which again is rough, as focus is not as perfect on the guide camera (done by hand) compared to the scope. I have seen a few guide stars at 1.6 arc secs but not for long. A bad night must be 5 arc secs seeing. Not too common.

Greg.

[LewisM]

Greg,

Yes, I do use autodarks. I was tongue in cheek re hot pixel - I always make sure it is not.

I just don't have many issues at all any more - it just works, and I am happy with that. I had even been leaving it out covered (sprayed down with Ballistol once every 2 weeks to be safe).

I was even discussing with H how long this one will go - watching the PHD2 status bar, it is "silent" most of the time. The last image I made (http://www.pbase.com/lewism/image/160890603) I watched for the entire 2 hours (OK, I let the focus shift with temp change - my fault - and yes, there is some bottom right tilt in the system) - the guiding was flawless and it issued a minor correction once every 4.5mins (seeing was unusual and fluctuating - thanks for living near a river and ocean!). The centroid graph was scaled to 0.5, and it stayed below 0.25 the entire 2 hrs. The auto settings just seem to work!

I am a complete gumby with the software - I don't even pretend to know what everything does.

[gregbradley]

Well you are lucky and I would not take that for granted as normal or usual. Its not.

By the way I can see you need to adjust your camera for tilt. The top right is nice and sharp but the top left the stars are quite a bit larger.

The bottom 2 corners are showing coma so there is significant tilt going on.

First step is to work out where those corners of the image are on your camera.

I used a dim torch and did a focus exposure. I lit up the top half and watched where that appeared on the computer image. In my case the top and bottom of the image was inverted. Then I did the same left and right. Again for my scope left and right were inverted compared to the computer image.

So now you know what corner of the camera corresponds to which corner of the image. Now you can start packing it out.

Then you focus say the bottom right of that image until its sharper. Note does the focus have to go in or out. That tells you how to pack it out. If it has to go in then you need to pack out the other side.. If it has to go out you pack out that side. Keep taking test focus images until you get all 4 corners the same and top and bottom stars, left and right sides show sharp tight star images.

I have found sometimes misshapen stars are easy to mistake as guiding errors when its non squareness of the camera. A lot of systems have some tilt and lesser model CCD cameras may not have well levelled sensors.

Greg.

[cdavmd]

Hi, just wanted to pass on some info. What you describe with the APA routine with T-point and TSX is exactly what I encountered when using a Mach -1. When I checked polar alignment with PHD after performing an APA, I found that the alignment was way way off. I investigated further comparing the recommendations in the T-point Polar alignment report with the actual fiducial star movement and noticed that the star was displaced in the wrong direction. I have posted to the Bisque site and worked with them over the last few days. This is a bug in the software. It will be corrected in the next daily build. I tested a preliminary new build tonight for Bisque and I can confirm that the bug is fixed.
Cheers
Carlos

[alistairsam]

Thanks Carlos

I presume you mean the Altitude recommendation
it tells you that the polar axis needs to be lowered and the next line says raise it.

good to know it's fixed.
I'm very happy with my 3 x 24 point T point routine. gets me close to 10arc sec and good for 60 min subs at 1m

cheers
Alistair

[Atmos]

That's good to know, I'll have to actually go and get the daily build stuff setup.

[frolinmod]

A new daily build was released last night. If you haven't already fetched it, please do so now.

[gregbradley]

I thought that was fixed a few months ago. The polar alignment report was advising incorrectly the opposite corrections needed for altitude adjustment than it should, ie. raise when it should be lower.

That was fixed 2 months ago. So is this a specific bug for a Mach 1 mount or general? I have not noticed any bug using APA routine since that bug fix around 2 months or more ago.

Greg.

[PRejto]

Carlos, welcome to IIS! I hope when you get your mount MyT sorted you will post some images here!

Greg, the bug Carlos is reporting is different from the one you mentioned. I think it only is a problem with AP mounts. When you had the first problem you were using the MX right? Now, your new AP mount was behaving with the APA routine? Odd that it would only show up in the Mach 1.

Peter

[gregbradley]

Greg, the bug Carlos is reporting is different from the one you mentioned. I think it only is a problem with AP mounts. When you had the first problem you were using the MX right? Now, your new AP mount was behaving with the APA routine? Odd that it would only show up in the Mach 1.

Peter[/QUOTE]

Yes that's right Peter. I updated the new daily build last night.

They didn't include the FLI Filter wheel like they promised in the camera add on so lucky Hartmut's Ascom driver works. Now I am using full Sky X for camera and tracking but PHD2 for guiding.

My main issue with CCDSoft was always how it only allowed you to set 4 filters up for an imaging run. A lot of images are HaLRGB. Also for narrowband I can set LRGBHaO111 S11 for a run and get the LRGB stars I need for the narrowband.

Sky X is looking very good these days.

Greg.

[frolinmod]

So you should remind them. They were probably in a hurry to get folks some quick fixes, but still, if they're forgetting something, please be kind and remind them. (Alas, I suspect they don't use a real tracking system to track these things.) Thanks.

[gregbradley]

Good point. I'll make a post.

I must say though I am becoming increasingly impressed with Sky X. Its now becoming mature is a massive piece of software with tons of features.

I tried @focus2 the other night and it worked right away. I was amazed. I am used to being thwarted and having to find out why something did not work for a few days with most software in this game!!

Greg.

[frolinmod]

Greg, as a long time TheSky6 user I think it just took you a longer time to warm up to TheSkyX. New features such as Supermodel, all-sky image link and accurate polar alignment wizard (with the kinks now worked out of them) sure help make the transition worthwhile.

[gregbradley]

Yes that's true. Its a world within itself and a very complex and complete piece of software that is constantly updated and improved.

Plus SB write the best manuals. They leave others for dead in that area.

Greg.

[cdavmd]

Thank you Peter!

I had forgotten to subscribe to this thread so I never received anyones responses. Life has gotten in the way of astrophotography lately so I have been off the radar. I hope to be a more active participant on these forums. I really like the layout of this site!

So in regards to the bug, I am not sure if it was specific only to the Mach-1. From what I gleaned from Mathew it sounds like a bug affecting Astrophysics mounts in general. They never did answer me when I asked what exactly was the cause (just for my curiosity and education) but based on some discussions with Patrick....it appears there is some addition to the values of RA and DEC so as to keep track of which side of the Meridian the scope is on. The astrophysics routine when married to TSX requires it. I am guessing but I believe there may have been a "sign" error somewhere. When I was testing the movement of a laser pointer using the APA routine indoors it was essentially sending it in the wrong direction both in Azimuth and Altitude.

It appears to be fixed now.

In regards to the MYT....its on its way back and I am awaiting the new shipment. Hoping it gets here soon.....Freeze warning tonight....winter is not far away and with its arrival my scopes go into hibernation until late spring.

Thanks again for the welcome all.
Cheers
Carlos

[gregbradley]

It definitely is fixed in the last daily build as I have used it the last few nights and my guiding graph looks very good. I am using an AP1600GTO mount.

Greg.