MX Pointing Ability

[CDKPhil]

The last few days have been fine and sunny here in the Canberra region.
I managed to get the the MX out for its first test run. I set it up on the pier, put the weights on then the OTA, balanced, then plugged it in to my Mac book Pro. Flicked the on switch, the red lights came on, the mount beeped twice and made a pleasant humming noise. I selected the connect telescope button in the Sky X, "Connected", "do you want to slew to the home position" yes, and away it went.

I followed the instructions to do a quick polar alignment. It was just that, very quick and easy. I then synchronised on a star and started to do a T point calibration run. After about twenty points I re adjusted the polar alignment. and started again. This is way to easy.
Last night I did a calibration run of 100 points. This was just using an illuminated reticle. I ended up with a pointing accuracy of 11 arc seconds RMS.
After the model was finished it was such a pleasure to use. Click on any thing that was in the chart and away the mount went, no fuss. Then looking into the eye piece and seeing the object dead centre or near enough was fabulous.
I might be able to refine this a little more by doing a larger model, and refining the polar alignment.

What a great piece of engineering and software. Money well spent.

Cheers
Phil

[cventer]

Agreed 100%

With good polar alignment and 150 point model I get 11 arc sec pointing accuracy.

That's within 5 pixels of dead center on my ccd chip.

Awesome bit of engineering.

[CDKPhil]

You got to love that.

[gregbradley]

Yes it is super accurate and gets accurate really fast.

Very impressive indeed.

Greg.

[pvelez]

You guys are making it hard - mine is due any week now. Has been since January

Sounds fantastic

Pete

[avandonk]

Stop it you lot. I am suffering from evaluation interruptus as I still need some pier bits.


I went out and repainted my pier matte black to match the new mount. A sure sign of madness for me.

So let me get this right my 300mm lens is about 5.8" per pixel with the Canon 5DH. So the pointing accuracy is about two pixels at 300mm!

With the new RH200 and camera it is about four pixels.

I have heard with more stars a better model will even do better than this.

It sounds like the PMX performs as good as it looks.

Bert

[frolinmod]

If you're doing manual rather than automated calibration runs, ouch, you're poking yourself with a sharp stick. Slap a CCD camera on that baby, bin 3x3 (or 4x4 with a QSI583) and run a 180+ point automated calibration. If you're permanently mounted in an observatory, a long winter night is perfect for doing a 700+ point calibration run that you won't repeat for a good long time.

[CDKPhil]

Doing it manually is not so bad. It was nice to be back out under the stars after a long period of cloud and rain.
I have no choice at the moment other than do it manually, I do not have a camera to put on it.:sadeyes: I have ordered an SBIG STI, that will not be here until the end of the month. And I still can't decide between the SBIG STX and the FLI Proline.

I do like the idea of doing an automated calibration run of 700 points. I would imagine the pointing error would end up very small. Mind you 11 arc sec is not bad after 100 points.

Cheers
Phil

[Jeffkop]

Awesome Phil ... nothing like the feeling of getting your monies worth eh !!

Hope my setup with the Titan ends up as successful

[CDKPhil]

Thats for sure Jeff.
The next test will be to see what the periodical error is.
I hope it will be within spec. I can't measure this until I get a camera, and get to a location with a lot better seeing.

I am sure you will do well with your Titan. It looks like a great mount.

Cheers

[gregbradley]

I have not been able to get automated callibration runs working yet with a Proline 16803 and Sky X. I am not sure what the issue is but it keep failing to plate solve.

I use 3x3 binning and 10 second exposures. I wonder if the chip is too large for it and I'd be better off using the ML8300.

Do you have to have a fairly exact north angle for it to work? That was the case with my Sky 6 and automapper 2 on the PME. Also the Sky 6 needed an extra library of stars. Does Sky X also need more libraries?


Greg.

[CDKPhil]

Hi Greg
Have you posted these questions on the Bisque forum?

I have yet to fully read and understand the instructions on automation but I wouldn't think CCD size should effect a plate solve.
Your binning sounds ok but maybe your exposure time is to long? The manual says for cooled CCD an exposure time of 1-2 sec should be sufficient. It also says "you only want a long enough exposure to acquire a minimum of six stars, but short enough to avoid blooming and saturation".
I would imagine your image scale and camera orientation would be correct.

I wonder if it is a compatibility problem with FLI cameras and Sky X. I am using a Mac so I only have a choice of SBIG cameras in the camera setup menu for now. What cameras can you chose on the Windows version?

It wouldn't hurt to try a different camera and see if it will work.

So have you been doing manual calibration runs? what has been your lowest pointing error?

Cheers
Phil

[cfranks]

I had a lot of plate solve failures in my early tests with TSX and the MX. I fixed them with the 'Image Scale' parameter. Other software that I am familiar with require the unbinned arcsec/pixel whereas TPoint requires the as/p of the actual exposure. eg my unbinned is 0.6 and I plate solve binned 2x2 @ 6 sec exposure with the as/p set at 1.2. My plate solve problems disappeared when I fixed that. In Telescope/TPoint's Automated Pointing Calibration Run tab as well as Source Extraction Setup Button. Also in Tools/Image Link Search Tab, Image Scale.

Charles

[Mighty_oz]

I've found 2 things mainly upset my automated t-point modelling,
inaccurate pointing ie too far away from where it thinks it is, usually fixed by a sync then plate-solve and
the image scale u use ,
u also have to remember that skyx uses diff terms for camera solves and automated plate solves.... doh.
Anyways once it works it's a dream, i use atm for my ST8300m 3secs with bin3 i was using 6 secs at 2bin but wanted to speed it up.
417 point automated model took just over 2 hrs i think.
Also if u try to plate solve too early in the night as i have done it doesn't work too well.

Just my 2c worth hope it helps

[cventer]

Greg

Check this setting

TheSkyX->Tools->Preferences->Advanced->Bad pointing sample criterion(degrees)

what is it ? Bump it up to 2 degrees for your camera. set your search area in the Setup tab of Automated Pointing Calibration Run window to 4 fields of view

see my settings in pic

No extra star libraries needed in The SkyX Greg.

I found the key to get automated t-point run working is that you sync before you start your run and then have your first automated grab somewhere close to our sync position.

If you home the sync and then do a meridian flip before starting automated run that's when it often fails.

[Jeffkop]

REALLY interesting thread this

[gregbradley]

Thanks very much for the tips. I am sure one of them or all of them will make it work. My FLI camera worked on that with the SB camera plug in.

I have my PME and automapper 11 and the Sky 6 worked on this point.

The only thing that makes me wonder about the Proline is the files are quite large. If I have trouble again after using your suggestions I'll switch the the ML8300 which has smaller files.

Cheers,

Greg.

[DavidNg]

Hi, I am new with astronomy and its complicated equipments. Would someone please advice: for the PMX home position the OTA is supposed to point the western side of the meridian as written in the manual? even if we are in southern hemisphere? (the instrument panel is at bottom rear) Thanks in advance.

[cventer]

In southern hemisphere, home postion is front of OTA pointing east.

[CDKPhil]

Hi David
Make sure you have set the MX for the southern hemisphere.
It is under the Bisque TCS window, Hemisphere setup.
Page 101 in the user guide.

The absolute home position will be hour angle -2 dec 0.
As Chris has said, the OTA will point to the east side of the meridian.

Cheers
Phil

[DavidNg]

Thanks for your help.