PIER-14 Prism done new flats done - existing jobs what to do restart or new or leave etc

Hi Steve,

Option 1 seems best for the M81/M82 job targeting the IFN. We are deep into a long job (75% done) so we don't want to restart.

None of my jobs have started yet, so I don't have any existing collected data to worry about.

M78 we have only got 15 subs by my count, so you don't lose much by restarting and it saves the complexity of doing two different calibrations. But it is not my job, we will have to ask the owner what they want to do.

Similarly only 12 subs so far for LBN878.

M66 I would be inclined to restart. It is only a 2 hour job, but again, let's ask the originator what they want to do.

I am just looking forward to some of the targets we have chosen recently getting going so option 3 is attractive to me, but I don't stand to lose any data ;-)

old_eyes

As I explained I am easy any any decision made however the only reason I made the post was the fact that the prism was causing issues and complex gradients etc

We have the flats so the data is not lost so I see no harm in starting again on all plans and if the data pre prism removal is ok then add it into the mix if you choose. Most data has not been started at present anyways so its a 90% painless fresh start

So my opinion would be to keep the existing data and simply start again but I will acquiesce to yourselves if choose otherwise

Generally I agree. The exception is the long M81/M82 job. We are 75% done out of 20h data collection. Finishing off with the new configuration rather than restarting seems logical, I can't tell if the data collected so far is fit for purpose, only the originator can answer that. If we restart, I would get some of the other jobs underway first.

old_eyes

Old_eyes

In this case the originators are myself and Pete, we simply put some jobs in to get the pier up and running, we dont mind what you do but if I am honest the only issue is the prism

lets for now say the data is good, however the complex gradient although removal in post processing is a PITA

We could

1, just put moon down on the remaining data that needs to be collected that gives it the greatest signas to noise ratio (in case users struggle with the prism)
2, just collect as normal
3, start afresh but make a long term moon down project collected over a few months only when moon is out the way

Come back to me with your thoughts

re the other data sets, they are a none issue really as they are not far enough in to be a concern  if we restart:)

Steve

What is the quality of the 75% of the data already collected? Only you and Pete can judge whether it can be saved.

If you feel the need to restart, I am fine with that. I would just ask that we push it into the background a bit so that some of the more recently submitted jobs get done as well.

Thats was what I meant with making M81 moon down and a lower priority object, more of a long term project sort of thing over the next few months

I am sure the existing data on M81is fine but if I may be a little more candid, I am worried about users struggling with the prism so I am trying to nudge you in the direction of starting again and if you can work the the "prism" data on top as well then take it as a Brucy bonus

I am one for nipping it in the bud rather than users potentially moaning about the processing 'prism" hells afterwards because we did not draw a line under it when we could have easily at this present juncture

Again please come back to me as we have some clear skies coming up

Steve

Steve,

As I said earlier, none of my jobs started before the prism issue was fixed, so I am not directly affected. I am quite prepared to abandon what has been done so far and start again.

If that is what you want me to say then yes. My vote is to restart the queue and give up on partial jobs with the prism problem. I just did not want to decide for other subscribers.

I agree, which is why I asked on a thread so I had more opinions. soon enough the clear skies would have force a decision anyways. So as I don’t tend to be a dilly dally sort of person executive decision made. We will restart all jobs and if the pre prism data proves to be useable and great then, it’s bonus time

M81 I will lower the priority and put to moon down and tweak some of the priorities on other jobs. We may have clear skies this evening and despite a near full moon I would like to get the prismless system tested and start getting data.

Steve

Good idea to test prismless system. Maybe one of the narrowband jobs on the other side of the sky from the moon to maximise the chance of a good test?

Something like that or we may just point at M42 in HA as its bright so I can use short subs, so less susceptible to the lunar influence

Not 100% sure until it gets dark and we have a better handle on it anyways

Steve

That may be a better idea. As you say, nice and bright and still about 54 degrees away from the Moon.

It will probably turn out to be cloudy anyway!

Dont jinx it

;-)

To be fair we have a blocking high over us pushing the westerley north and over the top of us so I do beleive we are a go

Still I have fingers and toes crossed as we have had the focuser on P3 worked on so we are testing that also!

Just a few test shots in HA on M42 I feel, not so good a night

we will run 20s and 300s images for as long as we can and share tomorrow so you can stack and check re prism removal. I feel the data will be a little cabbage to say the least lol

We do what we can!

Phil is just zipping up a mix of 20s and 300s test M42 subs so they will be available to download this afternoon

Although it was clear last night a combination of full Moon and hi haze has made the data a little pants to say the least but it's good enough for the purpose of post prsim checking however.

If you could let me know what you think please once stacked etc ?

Steve

Morning guys, 

Only just logged in and saw this thread. Good news on the prism, and I'm happy with whatever is done. I agree that with most of M82 imaged it's probably best to complete it. 

With the data received so far I've only managed a monochrome version of M31 after a fair bit of cropping, since none of the handful of blue subs were usable. This however has been far easier to do as there were no meridian flips. 
On the contrary Simeis 147 is too challenging due to the problems caused by the flips. The only thing I can do is to crop some off the top and bottom. 

It's possible/probable that I'm suffering from lockdown fever combined with a very high mileage brain, but I'm struggling to understand how that prism caused the effect I'm seeing. I did mention this shortly after viewing the data but it was dismissed then. I'll try to explain it again with an example. 

When imaging east and capturing an equilateral triangle -shaped star formation, the single star is at the top of the sensor and the prism obstruction is on the bottom. In this example they are opposite to one another. Now what should happen when the scope is flipped upon crossing the meridian? The sensor is now rotated 180 degrees, so the single star is now on the bottom. Star alignment will rotate those frames 180 degrees to match the ones from the east. If the obstruction rotated with the entire optical train, as should be if the prism is being blamed, then that too should be inverted and be at the top. The star alignment rotation would then also place the obstruction at the bottom and align. This is not what I am getting. The obstruction is in my case always appears on the bottom, so after every frame that needs rotating in alignment, it ends up with it at the top. The result, two shadows to deal with. This is clearly illogical. If it wasn't then during calibration a single dust bunny would also end up appearing in 2 opposite places. 

I'm suspecting that somewhere in the processing I may be doing something wrong and perhaps not selecting an option that would cure that. When I imaged at home I only ever imaged in the West due to trees, so have no prior experience. 

Sorry for the length of this post but it's not an easy thing for me to explain. Please help an old man who I suspect is starting to lose it. 

Cheers, 

Ray 

Hi Steve, 

Just read your post after finally posting mine. Great news and I'll definitely be straight on it. 

Cheers, 

Ray 

Short version Ray

Said removal of prism should be the saviour of your pier flipping sanity

Please also test the post prism data set to make sure your also ok as one would presume it will also have a pier flip in the middle ?

Steve