Plate solved, slew but no object

[Palantir]

Hello

Just started and first attempt at astrophotography failed spectacularly (obviously).  I did some research and bought as good a mount as I could.

Kit:

Avalon Fast Linear Reverse

Sony A7rii with 70-200mm f2.8

MacBook

Software:

AstroDSLR and AstroTelescope form CloudMakers

Process:

Polar aligned

Mounted and balanced camera (to light for any counterweight so just used bar)

Plate solved using AstroDSLR and http://nova.astrometry.net, twice and synced to mount

Slew to Pinwheel galaxy and took a test frame, couldn't see the galaxy put pressed on with sequence of 30 second exposures at iso 100 at f2.8.

After downloading light frames and stretching definitely no  success.  I can see stars and they are in focus, at the edge the stars

My question is what is the most likely step that failed?  Is plate solving 100% reliable?  The mount is second hand but appears to be working and is in very good order.  Even though the DSLR was only clamped on using a Vixen to DSLR 'L' plate I assumed it doesn't matter if it is exactly lined up with the mount as I aligned using an image form the camera (if that makes sense).  Anyway if someone spots something I would much appreciate it.

[Erling G-P]

Did you stack your light frames ?

If not, it may not show up in a single image at ISO 100 and 30 secs.  Also, what focal length did you use ? - even at 200mm, it's bound to be pretty tiny. Here is a single 30-sec frame at ISO 1600 through my PDS200 (focal length 1000mm):

[vlaiv]

Plate solving is not 100% reliable, there are number of things that can go "wrong". However, I would say that astrometry.net is 99% reliable, and you would certainly get the message if plate solving failed.

M101 is very low surface galaxy, and it might not show in single sub. Have you tried stacking?

Probably best course of action (if you've finished your session) would be to stack frames and see what you get out of them. You can upload finished stack to astrometry.net and have it plate solved again - this will tell you if you've got the target right, or which part of the sky lens was pointed at.

If you are still "in session" - maybe try to see where lens is pointed at to get rough idea if it is correct part of the sky - it should be above big dipper, somewhere like this:

[david_taurus83]

M101 is very faint. You'll need longer exposures and higher ISO to reveal it. It's not an easy object to begin with.

You probably did everything right in regards to platesolving and aligning. Can you post an image?

[wimvb]

31 minutes ago, david_taurus83 said:

You'll need longer exposures and higher ISO to reveal it. It's not an easy object to begin with.

+1 for this.

[Palantir]

Thanks everyone, let me have another look at the lights, I’ll stack them and post results.

[kirkster501]

With a DSLR you should using ISO 800 or 1600 on deep sky astronomy subjects.  Also, 30s is not long enough, even at ISO 800.  One to two minutes is where you need to be. 

[ollypenrice]

Just drop a reasonably large JPEG version of one of your subs onto here and I'll use Registar to drop my M101 onto it to try and see where you are. A high quality image isn't needed because Registar will just read the star pattern.

People seem to feel that plate solving is compulsory. I don't use it, though. Anything software/PC/USB will be less reliable than the rest of your kit. Once you've synched onto a star (or two or three) you can just drive manually to M101's RA and Dec co-ordinates.

Olly

[Palantir]

Hi Olly, I did run the stacking anyway as I set it off before I left for work.  Attached is the stacked image.

Hi Kirkster501, I did have the ISO at 800 to begin with but it was over exposing quite quickly.

Cheers

Tariq

[Palantir]

Actually that didn't work, here is a better jpeg:

 

[Palantir]

Hi all, mystery solved, I was pointing at the galaxy and the plate solve did work just too faint to see as many suggested.

Apparently the galaxy is there (see pic).  I should say, for anyone using a Mac, the support was excellent from cloudmakers (http://www.cloudmakers.eu)  who developed the AstroDSLR and AstroTelescope software (I was emailing with Peter the developer).  I can also recommend their software as it just works without much fiddling.  I set up a windows 10 laptop with EQMod and my Mac with the Astro suite from and can say that the astro suite of software was very clean and did what it said, certainly from download to set up and running was a lot quicker on the Mac.  I think it is helped by the use of the INDIGO and INDI platform which (putting my day job hat on) is a well architected software platform IMHO.

[david_taurus83]

Hi. Your pointing at Mizar in that image with Alkaid to the left of frame. M101 is just out of frame a little more to the left.

[david_taurus83]

Oops. I've just checked again. It's possible you may have caught it north of Mizar, not to the left like I suggested.

[vlaiv]

I'm sorry to say - M101 is outside of the frame, I've marked its approximate position in this screen capture of your posted frame:

Here is same region from Stellarium:

Notice bottom row of stars and almost perpendicular one going from second from the right upwards at an angle - M101 is somewhere between forth and fifth star a bit to the left ...

 

[AngryDonkey]

Was there a large slew to the target? The easiest thing would be to plate solve again once you are very close to the target (after the first slew) and then do another slew. That's how the centering process works in most software i.e. slew (and/or rotate), solve and repeat until you are close enough. SGP for example will do all this automatically until you are on target with the correct angle (manual rotation of scope/camera required if you don't have a rotator). Maybe your software has a similar 'center' function?

It's also very helpful to setup your field of view (camera/scope) and rotation in your planetarium software so that you already know what to expect. I use SkySafari on my mobile which works great (but I'm sure you can do this on most other programs). When you then take a longish test exposure you can easily see if you are on target.

[Palantir]

Hi David, Vlaiv and Mike, thanks, looks like a little more investigating on my side needed.  I will have another look.  Its a bad week this week in the centre of London so I will have to wait before I can try again.  There was a large slew (I guess its subjective) so will definitely try plate solving on the target object or nearby in future.  Do mounts normally loose their place between slews?  I thought once you were plate solved that was it for the session.  I'm also going o have a go at astrometry.net to have a go myself.  I did look at the website but couldn't see where you can do it.

[vlaiv]

Go to nova.astrometry.net and hit upload menu item:

That will bring you to the page where you can upload image and set basic parameters.

Mounts usually don't loose their place, but large slew can make it appear to loose their position for number of reasons:

1. Bad polar alignment - mount thinks it is always perfectly polar aligned, but if it is not it will not track perfectly. Slew is one big "tracking session" sped up. If there is DEC drift due to polar alignment error - large slew will make that error large (expected and final DEC)

2. Cone error - even if mount is perfectly polar aligned - scope need not be perfectly mounted. Cone error is when optical axis is not parallel to mount pointing axis - so mount is pointing to one place, but scope is slightly offset - so plate solving is telling mount that it is pointing slightly offset .... error compounds on a big slew.

3. Mount tracking accuracy. In order for mount to track perfectly it needs quite precise clock. More often than not in most budget mounts, actual sidereal rate will be very slightly different than true sidereal rate - it will be off maybe few seconds up to a minute in 24h. Slew only lasts for dozen or so seconds, but it covers what could be more than couple of hours of "tracking" time.

4. Atmospheric refraction will make apparent star position different than true star positions (bending of the light when it goes thru the atmosphere). This is very noticeable lower down - closer to horizon.

Then there could be other things that are "operator error" rather than system error - like selecting wrong tracking rate Solar/Lunar/Sidereal and such....

All of the above means that after large slew you can end up in vicinity of your target but not exactly on it - hence good recommendation - do first plate solve for mount to get idea where it is currently pointed, then slew to target (this can miss by some amount because of above reasons) - and then do another plate solve. This will tell mount where it is pointed and you can proceed with very small slew - any error from above reasons will be just couple of pixels in the image so second slew is almost guaranteed to put you on target if you ended up reasonably close after first slew.

[AngryDonkey]

17 minutes ago, Palantir said:

Do mounts normally loose their place between slews?

They don't loose their place but the model is not perfect because of imperfect polar alignment, mechanical issues etc. so there will always be some error. Your mount (which is excellent by the way) is capable of building up a pointing model i.e. the more points you plate solve and sync to the mount the better it will be (if the points are well spaced over the entire range). This model will then compensate for the above and make slewing more accurate. Avalon even have a tool for this (XSolver) where these models can easily be created and even saved (although this really only makes sense if you have a permanent setup).

For imaging I never build up a model though, I don't think it's required. I put the mount in the home position based on the markers on the mount. Then I use the 'Sync Home Position' function to carry out a rough sync. Then I slew the mount to a target, plate solve and sync and slew again, plate solve and sync etc. until I am on target.

[Palantir]

Thanks, Vlaiv, Mike.  So I think my polar alignment was good (I found what I believe was the pole star, downloaded the app and aligned it to the correct part of the circle according to the app) but I think that my DSLR was not aligned with the polar axis accurately as there was no was to line up except by eye.  Also because I used a 'L' plate it was perpendicular to the normal way a telescope would be mounted (see picture).

 

Mike, I have the synscan version, does it also build up a model or is that just the stargo version?

[AngryDonkey]

On 11/06/2019 at 23:59, Palantir said:

Mike, I have the synscan version, does it also build up a model or is that just the stargo version?

Sorry, not sure but I think if you run it with EQMod then it will be possible (and many other things). 

[Palantir]

Thanks Mike, do you have the star go or synscan version?

[AngryDonkey]

I've got the stargo.