Doing away with the focal reducer

[aparker]

Inspired by Martin's experiment with using a Barlow with his natively short FL scope, tonight I tried something I had been wanting to do for a while - ditching my 0.5x focal reducer and trying some EAA with my C8 at it's native ~2000mm FL.  It took a bit of time for me to learn my mount well enough to get it aligned with enough accuracy that I can reasonably GoTo things with a 10 x 7 arcmin FOV, but having that sorted now, the time was right.  First I did M57 with the FR and without.

M57 at 1000mm:

M57 at 2000mm:

Much better detail in the longer FL view.

[aparker]

I could still have my mount dialed in better.  The hot pixels show about 1 arcmin of drift over that 300 sec of exposure.  

Next I spent a LONG time trying to get this to work with M51.  At 2000mm the galaxy fills the Lodestar FOV beautifully, but unfortunately there are no foreground stars to speak of in frame, and I could NOT get LL to stack.  At one point I tried longer exposures, without stopping to re-run longer darks, and LL just started stacking on the hot pixels.  Sigh.  No M51 for me.

So I went over to M27 and got this 40x10 sec view that captures some nice detail.  Lots of stars here to stack on.

This could even have been better, I think, but at this point the bane of Schmidt-Cassegrain telescopes was upon me - the corrector was dewing up.  Instead of going and trying to find the dew heater, I hung it up for the night.  But I definitely am going to do more of this long FL stuff for small, bright objects.  I could even see adding a Barlow and trying M57 at 4000mm.

[Dom543]

Very nice and interesting!

Have you seen noticeable improvement of image quality or detail between a stack of 10 and a stack of 40?

If so, I would be curious to see the images saved after the 10th and after the 40th stacks.

My experience is with different types of objects, so I don't know how this works with such slow optics.

The results are impressive!

Clear Skies!

--Dom

[Martin Meredith]

I really like your results and shame about M51. I bet it would have looked superb at that FL.

We're maybe hung up on always achieving fast optics but with a really sensitive camera, bright objects and the possibility of long total exposure times (while still enjoying near real-time via stacking), there are lots of advantages to slowing things down. I need a lot more practice in getting things spot on though. As you say, tracking/alignment becomes more critical.

I'm also going to try on some fainter objects like some of the Arps and Hicksons where having a little more resolving power might be beneficial to tease out the form of the interactions/clusters.

Martin

[aparker]

Very nice and interesting!Have you seen noticeable improvement of image quality or detail between a stack of 10 and a stack of 40?If so, I would be curious to see the images saved after the 10th and after the 40th stacks.My experience is with different types of objects, so I don't know how this works with such slow optics.The results are impressive!Clear Skies!--Dom

Dom,

The images absolutely keep improving with more data. M27 was nothing like the picture you see above after only 10 subs. It was much more just a bright region, without all the structural detail (actually a lot like it is visually in a big scope). If my scope had been dew-free at that point, I could have kept going and grabbed even more detail and contrast. I will definitely come back to this object!

Come to think of it, it's funny how intriguing it is to do this even with a familiar thing like the Dumbbell. Even though I've looked at it many times visually, and seen dozens of great astrographs of it, I think I only really LOOKED in detail last night as I was watching the stack build and straining to pick out each new bit of subtle detail that emerged...

[aparker]

I really like your results and shame about M51. I bet it would have looked superb at that FL.

Martin

Thanks Martin.  M51 may need to wait for my someday venture into larger sensors.  A Trius SX-694 binned 2x2 would yield roughly twice the FOV as the lodestar (really 4x; twice in each dimension) which should grab an object like M51 at the same pixel-wise image scale and sensitivity, but with enough "room" around it to get field stars to stack with.  

I worry that lack of field stars may be a frequent problem in looking at the types of faint extragalactic objects that you often work with, when using very long FL.  Objects like planetary nebulae that usually lie in the plane of the Milky Way will always have many, many adjacent stars, even at very narrow/magnified image scale, but looking up and out of the galactic plane where a lot of the most interesting galaxies/clusters are, this may not always be the case.  I suppose experience will tell.

[aparker]

The other thing I'm wondering about is why there are still two hot pixels.  The dark subtraction kills the others (and there will be many more without darks), but not those two.  Maybe a question for Paul.

Martin, does your experience line up with this, or do you usually see all hot pixels eliminated?

[aparker]

I think I'm going to try out NGC6946 next time out.  Should be a good, frame-filling bright spiral galaxy, and its location in Cygnus means there should be adequate field stars to get stacking to work...

[Martin Meredith]

The other thing I'm wondering about is why there are still two hot pixels.  The dark subtraction kills the others (and there will be many more without darks), but not those two.  Maybe a question for Paul.

Martin, does your experience line up with this, or do you usually see all hot pixels eliminated?

What I find is that the darks when they're fresh do a good job but if temps are falling it pays to top them up every hour or so (I'm usually too lazy though). At least with short subs it doesn't take much time out of the observing session. But you may have dead pixels as opposed to temperature-sensitive hot pixels? Do you find they don't get cancelled ever? Maybe we need something like bad pixel mapping to deal with it i.e. the software learns which are your bad pixels and interpolates.

Martin

[Martin Meredith]

I worry that lack of field stars may be a frequent problem in looking at the types of faint extragalactic objects that you often work with, when using very long FL.  Objects like planetary nebulae that usually lie in the plane of the Milky Way will always have many, many adjacent stars, even at very narrow/magnified image scale, but looking up and out of the galactic plane where a lot of the most interesting galaxies/clusters are, this may not always be the case.  I suppose experience will tell.

I just checked some of my deeper images and you're right that it might be a problem for some at 1500-2000mm FL where only 6-10 stars would be visible in short subs.

In theory one could piggyback a wide field scope/cam to do the alignment and then apply it to the longer FL image but maybe that is complicating matters! Or maybe not if the wide field setup also doubles as an e-finder as in some of Nytecam's work. All it would take would be for LL to continually output alignment parameters for some other smallish piece of code to rotation/translate/scale the main image -- easily achieved.

The e-finder concept is also potentially a practical way to use EAA with a big dob on an equatorial platform…. fast stacking at f4.5 with 18" of aperture could be rather interesting if it could be made to work.

Martin

[Dom543]

I don't quite see why you are worried about having too few stars.

As far as guiding goes, people routinely to off-axis guiding with SCTs. OAG relies the same stars that are visible in the FOV of the main tube.

Regarding the reliance of stacking on stars, with H-alpha filters LL often uses less than 10 stars and works just fine.

I may be missing some other reasons you need more stars.

--Dom

[MarsG76]

That is a fantastic result... Well done, the M27 and M57 look great. There is definitely a lot more detail in your 2000mm image compared to the one imaged at 1000mm.

I too have started imaging in F10 on my 8" SCT ever since I managed to get my PA, tracking and autoguiding where I'm able to do beyond 30 minute subs with no drift.

I use a modded DSLR for all of my DSO imaging and I find that using low gain and long sub exposures, upto 30 minutes subs, makes a big difference in capturing not only the faint stars and nebulosity but also more detail can be extracted in processing when stacking the long subs.

The arcs you have made by the hot pixels when stacking, if you're using a guide scope for guiding, I think this might be caused by mirror flop.

I'm looking forward to see your result when you start imaging at 4000mm.

[Martin Meredith]

I don't quite see why you are worried about having too few stars.

As far as guiding goes, people routinely to off-axis guiding with SCTs. OAG relies the same stars that are visible in the FOV of the main tube.

Regarding the reliance of stacking on stars, with H-alpha filters LL often uses less than 10 stars and works just fine.

I may be missing some other reasons you need more stars.

--Dom

Hi Dom

LL requires a minimum of 5 stars to perform stacking. Checking some of my deeper images and looking at halving the FOV there are going to be some marginal cases (although this could be traded by exposing for longer in some cases to reveal a few more). 

Here's a single sub (15s I think) of Hickson 70 where I'm just showing the estimated FOV for a Barlowed version of the same shot. (I.e. this wasn't Barlowed).

For a Barlowed shot I'm guessing I'd struggle to pick up all of these stars in the same exposure time. This example might be fine, but there will be others where it could be an issue. But let's see how it pans out in practice. This is at 1.6m FL so the issue will be greater at 2m and beyond.

Martin

[Dom543]

Dom,

The images absolutely keep improving with more data. M27 was nothing like the picture you see above after only 10 subs. It was much more just a bright region, without all the structural detail (actually a lot like it is visually in a big scope). If my scope had been dew-free at that point, I could have kept going and grabbed even more detail and contrast. I will definitely come back to this object!

Come to think of it, it's funny how intriguing it is to do this even with a familiar thing like the Dumbbell. Even though I've looked at it many times visually, and seen dozens of great astrographs of it, I think I only really LOOKED in detail last night as I was watching the stack build and straining to pick out each new bit of subtle detail that emerged...

Alex,

What you are saying sounds logical. Astrophotographers also recommend as many frames for stacking as possible to improve data and detail. The reason for my question was that I don't see noticeable improvement of image quality after the fifth frame stacked. It may have to do with the settings I use. I usually have contrast at 0.5.

It would be helpful, if you could please post a screen grab showing your settings panel.

Thank you,

--Dom

[aparker]

I don't quite see why you are worried about having too few stars.As far as guiding goes, people routinely to off-axis guiding with SCTs. OAG relies the same stars that are visible in the FOV of the main tube.Regarding the reliance of stacking on stars, with H-alpha filters LL often uses less than 10 stars and works just fine.I may be missing some other reasons you need more stars.--Dom

Dom, M51 is probably a rare case. If you look at a photo of it, and imagine a rectangular field just capturing the two galaxies, there are not any bright field stars in view. Using up to 30 sec exposures, I was unable to get LL to recognize enough stars to stack successfully. Obviously many other DSO have enough foreground stars that this is not a concern (for example my shot of M27 above).

[Martin Meredith]

Alex, as you can see on the Barlow thread I managed to get M51 to stack at 1600mm FL but your intuition that stacking could be an issue at long FLs in certain fields is absolutely spot on. In perhaps 1/3 of the cases the number of available stars was on the borderline for LL. This is in the context of short subs (15s in my case). For M51 there are a bunch of what I take to be foreground stars superimposed on the galaxies that I imagine LL is using for stacking purposes. In some of my shots I had to reposition the target to gain the extra stars. 

Martin

[aparker]

Alex, as you can see on the Barlow thread I managed to get M51 to stack at 1600mm FL but your intuition that stacking could be an issue at long FLs in certain fields is absolutely spot on. In perhaps 1/3 of the cases the number of available stars was on the borderline for LL. This is in the context of short subs (15s in my case). For M51 there are a bunch of what I take to be foreground stars superimposed on the galaxies that I imagine LL is using for stacking purposes. In some of my shots I had to reposition the target to gain the extra stars. 

Martin

Hi Martin,

Looking at your result leads me to suspect that I was already suffering from dewing when I was trying to get M51 to work.  My scope is mounted way up on my balcony "pier" so I can't really see the front end of it - I only realized the dew was getting bad when my laptop started getting damp...  In my individual subs I only saw the fuzzy swaths of light that are the spiral arms - no stars that I can recall.  I need to go back to this with the dew heater in place.  Possibly Wednesday if the forecast holds.