Pixinsight users help needed

[barrie greenwood]

Hi guys .

I'm trying to plate solve through photometric colour correction in pixinsight and all I get back is failure when using my stacked m31 fitts file however if I solve 1 single neff file (raw) its has no issues solving and then colour correction .

I guess it doesn't like the fitts files but I use dss to stack and it willnonly give you fitts or tiffs to save the stacked image.

I've uploaded the stacked image to astro. Net and it solved it no issues 

Any help or direction more than appreciated 

TIA Baz

[scotty38]

I don’t think It’s fits in general as I just tried one and it calibrated ok for me

[barrie greenwood]

Not sure why it's failing to be honest .

Was it a single frame you solved 

[Anne S]

Had you cropped and resaved the stack? You may have lost the fits header info.

If so have you tried inputting the catalogue no and camera pixel size etc and looking the coordinates up first?

[barrie greenwood]

Yep put the m31 ra and Dec in but did crop it so I tried with an uncropped version of the same pic still not playing .

I've input my focal length and pixel size yeah

[Budgie1]

Which database server are you using? PI defaults to Strasbourg but try different servers and see if it will solve.

[barrie greenwood]

18 minutes ago, Budgie1 said:

Which database server are you using? PI defaults to Strasbourg but try different servers and see if it will solve.

Tried em all still no dice .

Its working in a single image raw file but not on the stacked images

[Laurin Dave]

I've had similar problems with PCC plate solving recently ...  when this happens I use the Image Solver in  Image Analysis scripts..   this usually works ... then pick up the co-ordinates for PCC from the image

[barrie greenwood]

I'll give it a try bear with me

[Aramcheck]

I used to have problems with PCC and a DSLR where the focal length in the FITS header was defaulting to 50mm. Checking "Ignore existing metadata" in the Plate Solving Parameters solved the problem.

Cheers
Ivor

[Scooot]

When you stacked it did you drizzle. I used to get a failure which was due to me not changing the arc/sec per pixel after I’d drizzled. I’ve since realised I don’t need to drizzle. 

Edited November 6, 2021 by Scooot

[barrie greenwood]

No but I restacked in pi and its done it now but strange really

[scotty38]

4 hours ago, barrie greenwood said:

Not sure why it's failing to be honest .

Was it a single frame you solved 

I realise it's a bit late now but it was a stacked image, colour of course.

[barrie greenwood]

16 hours ago, scotty38 said:

I realise it's a bit late now but it was a stacked image, colour of course.

Yeah it was a coloured image from my nikon dslr a D90 to be specific. 

It looks as it won't use the photometric process unless I run it through pixinsight to stack them.

Let me ask u guys what importance do you put on flats bias darks etc as some times I've found using them makes the image worse or us this a known issue and do you take out or use as required depending on results .

Regards Baz 

[scotty38]

I'm no expert but for my dedicated CMOS I don't bother with bias and use flat darks along with darks and flats. It may be different for a DSLR though. Have you seen Adam Block's WBPP series on youtube, worth a watch for any PI user

[barrie greenwood]

 Take a look 😊 

[Viktiste]

 

A little late too, but - under Image parameters you can press 'Acquire from Image' and select one of the raw files that does work. That will make sure your stacked image gets the correct date + coordinates. I have had better luck with the France server than the UK server.

Also, if you get an error similar to 'Error: Insufficient photometric data: Got 4 sample(s); at least 5 are required', then untick 'Automatic limit magnitude' and manually play with the 'Limit Magnitude'. As per this post on the Pixinsight forum: https://pixinsight.com/forum/index.php?threads/imagesolver-fails.16215/#post-105315

[kirkster501]

When this fails it is normally always because the image scale is set incorrectly.  Why not share your file here with your FL and pixel size to see if we can solve it?

[scotty38]

Lots of PCC failures are down to DB access so it's worth installing the Gaia and APASS DBs locally. This is another thing covered in one of Adam Block's videos that is worth watching. Again, I'm no expert, but I've not had an issue with PCC since I did this. You'll need to remove the "Automatic" selection in the tool and pick the DB needed too. He also explains why APASS9 or 10 will work/not work for some targets based on different coverage gaps in both.

Edited November 12, 2021 by scotty38

[daz]

On 07/11/2021 at 12:57, barrie greenwood said:

Let me ask u guys what importance do you put on flats bias darks etc as some times I've found using them makes the image worse or us this a known issue and do you take out or use as required depending on results .

Very important!

A lot of it comes down to consistency of your data and calibration files.

• Are the focus points the same in the flats as the lights?
• Have you moved the camera in relation to the telescope?
• Are your darks at the same temperature as the lights, and the same duration?
• Are you flat-fields bright enough or too bright?
• ..
• ..

For a DSLR, you won't need BIAS frames (DSLRs can't do zero-length exposure) so take flat-darks - darks that are the same length as the flat frames. They will still be very short, but will work perfectly.

Calibration frames need as much care to collect as the lights - possibly more because of the number of them!

Watch the Adam Block videos. Then watch them again!

 

And yes, post the troublesome frame, so we can have a play and see if we can advise further