No Fred not loose and not too tight, just as the instruction manual states. Now you see what I am going through. Wysi just confirmed his mount has zero play in it too. So now to fix this problem.
OK, well Brad Moore is paid to know and your last chance, he knows PME better than SB. Other than that just kill yourself, itll be less painfull, cheaper and faster.
Or.........strip it down and learn yourself (that slop must be the problem).
I just started following this thread. I went out to my observatory just then to test my counterweight shaft to see how much slop it has. It has none. It flexes slightly under a push but it is firm right from the start with no noticeable slop. So if yours has 10mm slop then something is wrong and your T-point model is detecting that.
As far as the mapping process itself. I used an ML8300 camera with 1 second download times at 2x2 binning, no subframe, and 10 second exposures. I got about 99% success rate and one map took 20 seconds to do from slew to finish. 200 points took about 45 minutes.
I could not get it to work until I added the USNO A2.0 database of stars that Marcus gave me a copy of. Once I added that to the database and let the Sky 6 know the location of the file it worked like a charm.
Use new astronomy press CCD calculator (its free) to calculate your
arcsec/pixel. As an example my ML8300 on the CDK17 at 2959mm focal length gives .38 arcsecs/pixel. It has 5.4um sized pixels. As you realised flatteners aren't all neutral to focal length. You'd have to get a factor for your flattener or as you did not use it. Also north angle has to be very accurate for it to work. Mine was around 2.33.
If the point of the exercise is accurate polar alignment then Pempro does have a wonderful polar aligment wizard. I used it once (I missed the last step) and later did the 200 point model and I was only a little bit off.
I check the autoguiding errors to make sure the theoretical produces practical observable gains otherwise what is the point. In my case the very adjustments did make a gain. Although I was surprised to see go-tos were not spot on. Anyway, accurate polar alignment and adequate go-tos was what I was after not necessarily object dead centre every slew. I frame anyway so I just don't want to spend time searching for the object.
So my conclusion is Pempro polar alignment wizard will get you to within a hair of precise PA and a proper T-point model will get you that last tiny bit.
By the way with a 200 point model started from scratch and synched to the first star manually got me 28.5RMS on the CDK17. Most of the points are grouped within a 50 arc sec circle. I deleted perhaps 4 or 5 outliers out of 205 points that left me with 200 accurate points.
As I say the USNO database made the difference for me. Thanks Marcus!
I just did the calculations:
TSA102 816mm focal length 102mm aperture F8.01 KAF8300 chip 5.4um pixels 3326 x 2504 1.36 arcseconds/pixel
with flattener focal length is 800mm F7.84 = 1.39 arcseconds/pixel.
Thanks Gary, I will try to get those requests sorted.
As to random looking data. I now think you are right.
On my mount if I stand with the RA pointed towards me and I gently grab the counter weight bar there is noticable slop in the RA axis (I am talking 10mm here; up and down perpendicular to the axis); so if you are holding the bar and the scope is say laying over in RA I can move the bar up and down by 10 mm. My camera is set up so that long axis of the sensor is perpendicular to the Dec axis. That should make the north at almost 90 to the RA. Would this slop account for the north south separation?
Hi Paul,
Good find and it may turn out to be the culprit.
Quote:
Originally Posted by Paul
When I took the face plate off some time ago I noted that the whole worm and gear move as one and that the assembly does not appear to be bolted to the body fo the mount, it sort of floats. Is this normal? I would have thought that this sort of slop would prevent good pointing. I think this is most likely the cause of the problem.
It certainly would be a good question to put forward to Bisque.
However I see in the thread another owner is kindly coming to the rescue with a PDF of the mechanicals.
Quote:
Gary, the odd thing is that many images are ok on the astrometric solution in CCDsoft but they get rejected in the sky or Tpoint (not sure which is rejecting, but I get the message in automapper that it has been rejected.. I get two main errors; insert WSC failure and cannot map. Last night I got 140 errors and 93 positives on one run. That means there is something really wrong physcially. Often the circled star in CCDsoft (the star being pointed toward) is all over the place. Sometimes near the cross hairs, but most often at random distances and places from the cross hairs. I think it might well be the mount work that was done before I bought it that is causing the problem.
Sounds as if the pointing is sufficiently out that the automapper is failing.
It would be instructive to even manually slew to 6 bright stars you know on
either side of the meridian and sample them and see what the RMS is for just
ID, IH, ME & MA. With the amount of slop you mention in RA, I think we
can anticipate that the RMS will be high and the value/sigma ratios still
very low.
So the next port of call will be to confirm with Bisque or another ME owner (Brad?)
as to how it really should be inside the mount physically with regards the RA
axis. Or as Fred mentioned, you might be able to figure it out for yourself with the aid of the drawings.
Once you nail that, I'd say you will be a long way along the path to dramatically improving your pointing performance.
Certainly TPOINT is all you will ever need to clean-up anything systematic remaining, including the polar alignment.
The plate solve will then find a lots of its work done for it already.
Certainly 10mm of slop in RA would explain the poor model!!! And it's a very poor model for a PME! Hopefully this is just a spring plunger adjustment on RA. BTW, sigma values in a good model should all be under 10. Sigma for the MA and ME terms should be less than 5 or so to use them reliably to set polar alignment. Usually the only way of doing this is to add harmonic terms until your RMS and PSD stops improving (or improves less than 10%).
Greg, I got the same figure for calculation. Although I have read in one of the manuals that anywhere around 0.25 of the correct scale will be find for astrometic calculations.
I also got the USNO A2.0 database from Marcus too . I did read on another forum though that greater success rates for plate solving are gained when only one data base is selected. Incidently did you select it by Data/file locations/data applications/core data bases and then in stellar options? Then saving the settings?
I honestly cannot see how I could get my PA any better now though after doing it myself with drift alignment. 20 minutes per star and nill movement is pretty good but I stand to be corrected if you think I can get better than that.
Gary, I will probably do the six star model once I get this adjustment sorted out. Thanks for the suggestions.
Marcus, I think it might be either the spring plunger or the locking bolt located up in the top of the mount. I need to remove the side plates (fortunately I removed the 6 little allen screws on those plates for future access so my PA will not be affected) and take a good look around. Brad is going to organise what adjustments will be need via PDF once I figure out where the problem really lies. There is a PDF on my way for checking of the RA slop to ensure the bearings are in correct tension. I have access via a mate here to any type of tension wrenches or vernier gauges that I should need. Spent plenty of time rebuilding motors in my petrol head days so I know all about tight tolerances. Having already taken a good look around inside one of these mounts they look pretty simplistic and yet have such good performance.
One this is for sure I will report back what is up with the mount and how it got fixed. It might be instructive to others to know how to fix similar problems.
I have definitely found the source of the problem and I hope I have fixed it too. On the advice of Brad I went and investigated the mount today. I have opened this mount before and am very familiar with most of it. I began my making a video of the slap (see below) It shows the slap in the mount and RA. This does not happen on other mounts.
I then removed the side panels to see if I could find the source. I had thought I could avoid disturbing the PA I got the other night but this was not to be. The side covers I had previously removed the bottom screws on the panels which means you don't normally disturb your polar alignment. Unfortunately to get the grub screws which hold the RA worm assembly in place you need to remove the very bottom narrow plates. This means I had to jack up the altitude all the way. I got the grub screws to discover that each one on each side was quite loose with the locking screws being loose too. I used an allen key to tighten up the grub screws on each side at an equal depth on each side until it was just snug or just finger tight (I figure that finger tight is the correct tension; too tight and the thrust bearings inside the assembly could be damaged). Making it dead center too prevents the worm from being one end or the other when it meshes with the drive gear. Contrary to some opinion the entire worm assembly was shipped to Brad and so any angle adjustment of the worm to the gear is not necessary as the unit bolts staight on and swivels up and down. It cannot be placed at an angle at all. So here is the video of the grub screws in question.
Click here to view the video of the mount stripped.
Once I got it all back together I connected the mount to the computer and just tested to see if the mount was sound and all the motors and gears ran correctly. The video speaks for itself.
I will be talking directly to Daniel from Software Bisque and seeking advice about tension required. I suspect my finger tight will be just fine and I will not be required to do further adjustments. Hopefully the pointing will be out of sight now. It is going to be cloudy and raining now for a week so I will post my next mapping run when I can see the sky again. Failing any need to do further adjustment to the grub screws.
I am a firm believer in sorting things out yourself. I now have a greater understanding of the mount and it reminds me of my first RA drive in the way the worm was engaged.
That's great news Paul - you must be relieved! That's one of the selling points of the PME isn't it - nothing is hidden from view. You have the peace of mind knowing you can can get inside to see what's going on and, more often than not, fix stuff yourself (with guidance from SB of course)!
I am very happy and releaved that you have done such good work finding out the problem and the solution. I am so sorry you went through so much on this problem and I want to publicly apologize to you as it was my mount that was sold to you. I believed it was in like-new condition when you recieved it and paid good money so it would be that way. I am saddened to find out that this happened and it is one of the real issues of being thousands of miles away from your eqipment.
The good news is once the problem was found you were able to fix it and you know more about your mount than most people do. I believe you will be able to get a fine T-point model and great pointing and guiding.
On Another note - I spent allot of time cleaning my equipment I just recieved back from Moorook. There is lots of brown fine dust in everything including my rack-pinion gearing on my AP130. So, I would yearly re-grease the worms just to make sure they are free of the fine grit. This stuff was everywhere . . .
Thanks Ken, yes I did a big clean when I got the mount. Don't get me wrong it had already had a good clean but I am a little more keen to have a really clean mount. Inside the mount was fine, no bull dust there, just only on the surface. The dust was the result of a couple of large dust storms a couple of years ago, one of which was recorded here is you are interested in doing a search. It covered a lot of NSW and Eastern South Australia.
Anyway, I have to wait to find out the results now. Cloud and rain predicted for next week.
Just thought I was give an update to the issues with the mount. Using automapper II and using the map current function I got my six stars for PA to a pointing average of 5.9 arc seconds and PA close enough to what I wanted.
Next I ran let automapper II make a list of stars of the entire sky and then let it go. I found that my error rate had gone down from 150% rate to 50%. I was still getting quite a few outliners (nearly 40 of them) but finally got my pointing down to with 150 odd points out of 195. See the scatter diagram below.
I think the software has some problems or I am still doing something wrong to get such high error rates, but for now I just want to get on with imaging again. If the sky ever clears.
I do think though that the dec worm block needs some adjustment too. There is a little slop in that too, and that was not changed but probably considering it had such a large scope sitting on it, maybe it needs some service work. I will be sorting a few things out to get this mount working really well but it seems for now I have sorted the main problems.
If you think there is something I am missing or have any final tips please let me know. I am all ears when it comes to this. I have read every manual inside and out and think that I have got it sorted.
One thing that bothers me is that when I do research in CCD and it does not insert the camera angle I cannot get the image to link and that makes for problems trying to sink that first star. Is it easier to just find a bright star and take that image at the center and sinc that rather than just pointing it at a constellation?
I have been running Tpoint and Automapper II the last 2 nights on the CDK.
Your final result there is similar to what I got tonight. Mine went as low as RMS 17 with 102 points mapped. I did several runs each time I adjusted the Polar Alignment after fitting the data and getting rid of outliers. It took 9 terms to get there. I simply used the suggest terms button and accepted it and watched the RMS number. I ran about 50 to 100 points in a run and then adjusted the PA then slewed to a bright known star, centred it, deleted the T-point model, then started a new one, then synched the star then started the new mapping run after mappung the current position (I don't think that is required for a run though).
As far as 6 points for PA I would think that is highly inadequate. Even just fitting the data and getting rid of outliers you see the suggestions for the PA adjustment change.
I have used 2 cameras now. The FLI ML8300 which is similar to your camera and the PL16803. The ML8300 I used 10 second exposures at 2x2. I added the USNO catalague from Marcus to my database in the Sky.
With the PL16803 I use 3x3 and 10 second exposures and it works well and usually gets more stars. Both are fast as the FLI's have super fast downloads of 1 sec or less. So total time is about 19 seconds per mapping including slewing time. So a 100 point model only take 40 minutes or less.
I found I had to take an image in CCDsoft then insert WCS research and it plate solves and gives the north angle. I then added that into the automapper default settings. Then it plate solved. It would fail without that north angle. Its done that twice now where it would not get the north angle when you press the get button next to north angle. So if you are not doing that try getting it through CCDSoft. I take it you have CCDsoft server settings turned on per the manual.
The only failures I got were because of cloud. Its probably a bit harder to do with a full moon. Perhaps that was all it was.
I have done 3 iterations in 2 nights and tweaked the polar alignment each time. Make sure you are turning the dec knob the right way.
I also check how its going by doing a bit of autoguiding and watching the errors and seeing if they are in fact lower than before. They have been so far.
As far as 6 points for PA I would think that is highly inadequate. Even just fitting the data and getting rid of outliers you see the suggestions for the PA adjustment change.
I found I had to take an image in CCDsoft then insert WCS research and it plate solves and gives the north angle. I then added that into the automapper default settings. Then it plate solved. It would fail without that north angle. Its done that twice now where it would not get the north angle when you press the get button next to north angle. So if you are not doing that try getting it through CCDSoft. I take it you have CCDsoft server settings turned on per the manual.
Greg.
The manual seems to indicate that 6 points is enough for getting PA. Several other people have indicated that too. With just the basic terms I have got it quite close. And; the final PA results are very similar to the 6 points results. Still 89 seconds away from the poll though in the alt and 7 seconds in azimuth.
I hardly ever get a plate solve in CCDsoft under insert WSC. Sometimes I might jag it and it gets the north angle, but most times it just will not play. I suspect this is part of the problem really. I set all the parameters correct, although on my version of the sky and CCDsoft some of the things that need checking are not present. It all seems to talk to each other though, which is weird. Everything is running in administrator mode. I will try this for now and see what happens. Maybe it is worth getting skyX and the new Tpoint version. I have heard this plate solves better anyway.
I checked the manuals and the 6 point mentioned in the manual is just to get the scope pointing closer to the object so you can find things.
If your PA is accurate with 6 then that is good. But per the manual and the T-Point workflow pdf it recommends a few hundred points for permanent installations.
Only once you have added terms and gotten the sigmas and the RMS down is the PA recommendation considered valid. This is all from the T-point workflow PDF.
That has been my experience with it. Recommendations on few points were inconsistent and did not always improve the PA I already had from drift alignment.
I have just done several iterations of models and after model fitting I got 0.0 recommendation on the elevation and .7 tick adjustment on the dec.
But thats 3 models, 1st one about 100 points, second one about 38 third one 102 points.
None of this of course is the problem here.
When you try to plate solve what error message is showing? When mine failed it was initially because of incorrect north angle.
When you zoom in on a starfield in Sky 6 do you see all those USNO stars showing up? Do you have the check box ticked in the display explorer for the USNO database? I take it your CCDsoft has the server settings set properly and you have both a CCDsoft box saying listening to port as well as one in the Sky 6? You need both going. Did you set the file location of this database in the Sky so the Sky can find the USNO database?
What operating system are you using and do you have enough RAM?
I doubt getting the later software will change anything as it should be working already. Something is off. Odd it works half the time. What exposure setting are you using? Are you binning? I take it you entered the correct arc/second/pixel for your camera/scope unbinned?
I used 10 seconds and 2x2 binning for my FLI ML8300. It worked well.
What scope are you using? If a TSA102 you may need to do 30 second exposures and use subframes. I say that because it needs to see lots of stars. I am using the CDK17 and of course 17 inches gets a bright image in no time.
It sounds like your mount is now working nicely and it is a software problem or something not checked that needs it in the various settings.
When you try to plate solve what error message is showing?
Thanks Greg. Error messages in CCDsoft are north angle failed or image link failed check co-ordinates and image scale. I just checked on two images of stars one worked and the other did not.
Quote:
Originally Posted by gregbradley
When you zoom in on a starfield in Sky 6 do you see all those USNO stars showing up?
Zooming in I cannot determine which stars are in the USNO catelogue. How do you do that?
Quote:
Originally Posted by gregbradley
Do you have the check box ticked in the display explorer for the USNO database?
If you are referring to having it ticked in stellar options under astromety, then yes. In the data/file locations it is not in bold like the other catelogues but when I use verify it comes up with a tick (is it supposed to be bold like the other catelogues?).
Quote:
Originally Posted by gregbradley
I take it your CCDsoft has the server settings set properly and you have both a CCDsoft box saying listening to port as well as one in the Sky 6? You need both going.
Both CCDsoft and the sky have ports listening and I think I have all the settings correct. Both have checked allow remote connections, both have remote clients use orchestrate or RASCOM, then the other settings needed for each.
Quote:
Originally Posted by gregbradley
Did you set the file location of this database in the Sky so the Sky can find the USNO database?
Not sure, I verified under file locations but like I said it is not in bold now like the other catelogues but it does verify with a tick.
Quote:
Originally Posted by gregbradley
What operating system are you using and do you have enough RAM?
I am using Vista with 4 gig of RAM
Quote:
Originally Posted by gregbradley
Something is off. Odd it works half the time. What exposure setting are you using? Are you binning? I take it you entered the correct arc/second/pixel for your camera/scope unbinned?
I am exposing for 7 seconds at 3x3 bin. I used Rod Wodaskis CCD calculator and got 1.39" with the TSA/flattener and KAF8300. The TSA native has 816mm but with flattener it is 800mm that combined with KAF8300 should have 1.39" per pixel.
Quote:
Originally Posted by gregbradley
What scope are you using? If a TSA102 you may need to do 30 second exposures and use subframes.
Greg.
TSA with flattener. I seem to be getting plenty of star but I will try say 20 seconds to see if that works.
Just out of interest, once I do get a plate solve in CCDsoft, do I then go directly to the sky and sync and then map it. Even if it is not in the center of the field (the star I am aiming at)? This seems to be my reading of the instructions. Does it matter if the star is not dead center of the cross hairs in CCDsoft?
Just testing I found that if the image does not plate solve in CCDsoft. I have to then completely close down both the sky and CCDsoft to get it to plate solve on a known star. Odd really. Could be a Vista problem.
Just to let you know. I used one image several times in a row today and tried to get a plate solve. It would do it if:
The view in the sky was near to the stated image scale. It will not plate solve if the view is not near the employed image scale. That is just weird.
By way of example, if I used the USNO catalogue the calculated scale each time would reduce and plate solve more stars. See images. If I used the UCAC2 catalogue the reverse would happen. That is; the scale would continue to increase and the view would take in more skies and plate solve less stars.
I really don't know what to do now. I cannot get this software to work for me and I don't want to buy more software so I can run AAGautomapper. I would need to buy maxim and pinpoint. I have read and read the manuals and I cannot seem to get this right. This is no doubt why I get so many errors in the plate solves. I am tempted to ask Software Bisque for my money back and forget the whole thing. So fed up with this software.
What do I do now???
I know the USNO stars are there as I did not have the fainter star option bright enough. Got that sorted.
. I did read on another forum though that greater success rates for plate solving are gained when only one data base is selected. Incidently did you select it by Data/file locations/data applications/core data bases and then in stellar options? Then saving the settings?
