Sequence Generator Pro & Plate Solving... Help please!

[PhotoGav]

I'm hoping that there is a bright spark out there who can get me on the right track with this...

I am trying to get Sequence Generator Pro working and have installed Elbrus as the plate solver. I tried it all out on the night sky last night and SGP worked perfectly for all functions except for plate solving. Elbrus wouldn't solve anything... It took the picture fine, but then after about a minute threw this error:

"Failed to plate solve: Elbrus failed to plate solve: The server threw an exception. (Exception from HRESULT: 0x80010105 (RPC_E_SERVERFAULT))"

I must have the installation of Elbrus wrong, so will fiddle with that and try and sort it.

Anyway, realising that Elbrus wasn't going to get me on target, I tried astrometry.net and it started solving plates. However, that's when my confusion of the way it all should be used kicked in...

I am used to using Cartes du Ciel, APT and Astrotortilla. I slew to a target in CdC, plate solve and recentre object in Astrotortilla and within a couple of iterations I am on target, ready to image. It just doesn't seem to work that way in SGP. So, please can somebody explain to me the workflow for getting on target with SGP!

What I have managed to glean is that you can solve an image file and then center on that result. Great, but what if you have no image! Is it not possible to do a similar thing to CdC & Astrotortilla?

Hope you can help.

Thanks.

[Barry-Wilson]

Hi Gav

About to go out with family and will give longer reply later today if I can.

Have you re-installed Elbrus using the new installer from the SGPro website?  May help.  The SGPro forum is very helpful http://forum.mainsequencesoftware.com for matters plate solving.  Ken and Jared are very quick with replies, especially if you are able to post up your SGPro log file.

The 'centre here' function is brilliant, whether on its own or as part of a sequence; persevere it's worth it.

Barry

[ChrisLX200]

Thing about Elbrus is that it is _very_ fussy about image scale, so if you have that set wrong it will fail to solve and throw that error. You must get an accurate estimation of the image scale and there are various ways you can do that. You get a rough estimate by calculation from your scope's focal length and pixel size, and that may work - but if there is a flattener or reducer in there then likely it won't be accurate enough (2 decimal places). You can take an image and submit it to Astonometry.net (online plate solver) and that will blind-solve and provide you with the image statistics, or in my case I use AstroArt 5's built-in plate solver (which is far more forgiving for image scale) and that will solve the plate, then by looking at Image Statistics you can work out the image scale in arcsec/pixel. Transfer that figure to Elbrus (well, SGPro under the Camera tab - SGPro transfers it automatically to Elbrus) and you'll find it works OK.  It is a pain to have to go through this process, and I had to do it for about 6 different configurations, but once working it is very reliable and works a treat for centering objects and re-centering the same object on another night.

ChrisH

PS - if you get stuck then post an image here, give me a clue where in the sky it is (faster that way) and I'll try to provide you with the image scale.

[Barry-Wilson]

Gav

Chris is right about Elbrus needing an accurate image scale.  I set mine to 3 decimal places and then it worked very well.  Set the blind fail over option too if you have internet access.  I never had to use the Elbrus interface as SGPro handles that.

When setting up a sequence I more often than not look up the target's plate solved co-ordinates in Astrobin (from an image, right clicking to copy the link in the address bar) and the paste these into target options, solve this, and then use this as the reference for centring.  In the mosaic wizard you can just type in the common name (e.g. American Nebula) and it will pull the co-ordinates from the web.  You can also paste the link to your own image, solve, and then use this as the reference.  Clicking 'centre now' in the target options will carry out a centring operation.  This is crucial if you have a number of targets or have automated your meridian flip.  Watching the sequence unfold with an auto-meridian flip (i.e. stopping the sequence, stopping PHD2, plate solving, flipping, re-centring, re-starting PHD2, re-autofocusing, and then finally re-commencing the sequence) is awesome.

My workflow: once I've identified my target, I run an AF, then click on the option on the target, centre now (centring takes place), then once centred I calibrate PHD2, I re-autofocus once guiding and then commence the sequence.

Good luck and I hope you get the plate solving working - happy to help if you have any more questions.

Clear skies.

Barry

I too have CdC, but now only really use it for planning rather than slewing and synching.  SGPro's plate solving synchs with EQMOD and so a pointing model is built over time (if you've a permanent set up).

I have to say that I switched to PinPoint a couple of months ago - it is faster.  I use the GSC catalogue downloaded from the dc3 website.

[ChrisLX200]

I select targets via Starry Night Pro, saved as a target list SGPro then has a built-in function to import them together with respective co-ordinate data. Starry Night also allows you to overlay images from Deep Sky Survey and other image catalogues so you can really see the oject you're about to image (rather than an outline on a map etc.,) which helps with framing. If you have a camera rotator (and I don't) SGPro will also rotate the camera to correct angle.. It's a powerful combination forautomated imaging.

ChrisH

[PhotoGav]

Barry and Chris, thank you both so much for your help.

That gives me plenty to work on and it certainly sounds as though it will be worth persevering with the clamber up the learning curve! As for target acquisition, I think that I just need to tweak my work-flow and be a little more organised and informed about what I'm going to image. That should be nothing but a good thing! Sounds like the process to re-acquire a previous target is very efficient and as for automating everything in my permanent set-up... I dream of such things! At the moment it's a scramble out into the garden when the forecast is clear to set-up my fairly limited kit, followed by an early hours break-down of everything. Especially when the forecast is for rain the next day.... Very tedious! Occasionally I do get to leave the tripod and mount in the garden until the next night and it is so easy to get up and running for that second session. Top of my wish-list really is an obsy of some sort... Just got to convince the boss, the children and the dog!

No doubt there will be further questions....

Thanks.

[PhotoGav]

Well, I have completely failed to get Elbrus working... However, Astrometry.net works perfectly every time! So, next time it's dark and clear I will give it all a go using that.

I've had communication with Jared from SGP about trying to get Elbrus up and running, but no joy so far. If anyone has any brainwaves as to how to make it work that would be interesting to hear.

Thanks,

[ChrisLX200]

Check that the parameter for Scale provided by Astrometry.net is the same as that passed to Elbrus when you right-click an image and choose Plate Solve.

ChrisH

[StarryEyed]

I will just add that AstroArt is very forgiving and U can do an allsky solution for a dslr frame in less than 45 seconds. You could just download the demo and solve some plates to get the fov etc.

I had a little trouble like this at first and just went with the all sky until I could tune the parameters.

Regards Kevin

[PhotoGav]

I've gone for a slightly left field solution now...

I'm using APT in CCD mode with Astrotortilla running alongside. There isn't quite the same functionality to automate great long sequences, but I am very pleasantly surprised with what is possible in APT. The target saving and slewing is perfect. And for the money, I'd say it's unbeatable!

Sent from my iPhone using Tapatalk

[StarryEyed]

Nice you worked out a solution.

[Astroscot2]

Sorry to bring up an old topic and go off subject slightly.

Chris / Kevin - can you tell me your process for plate solving using AA5? ive never managed to get it to work,  The AA forum has not been much help with regards to the "process"

Appreciate your help

Mark