First light with my new Esprit 80: M13

[Pompey Monkey]

Initial test showed some elongated stars so I increased the optical spacing to about 55.5 mm: I'm still not sure...

Here is the result of (approx) 22-24 three minute subs for each of R, G and B that has been crudely processed:

Note that I did not focus between the filters - maybe red is a touch off, but not by much.

It was also a bit hazy from time to time...

The optical quality of the scope is very, very obvious from this first image, but there is still some elongation of the stars in the bottom left. I'm not sure if this is a spacing issue of some field tilt, or both.... After my exploits with the Star 71, I am rather sensitive to this kind of thing!

Anyway, it's clearing outside and I want to try out my Ha filter on NGC 7000 that is creeping up in the East

[gingergeek]

Well I like it ! 

[ollypenrice]

Looking at the fullsize I'd say the up-down elongation runs right across the bottom but is slightly worse in the left. The trouble is that making too much of this on a theoretical basis may lead you astray. While I'm inclined to think 'tilt' you never know. I've had chip distance errors which only affected one corner when you'd expect all corners to be affected.

Most of the image is nice and tight. Lose that elongation and you have a winner, I'd say.

Olly

[kendg]

Now I know why I want that scope as my next one.

Great image 

[Pompey Monkey]

6 hours ago, ollypenrice said:

Looking at the fullsize I'd say the up-down elongation runs right across the bottom but is slightly worse in the left. The trouble is that making too much of this on a theoretical basis may lead you astray. While I'm inclined to think 'tilt' you never know. I've had chip distance errors which only affected one corner when you'd expect all corners to be affected.

Most of the image is nice and tight. Lose that elongation and you have a winner, I'd say.

Olly

I'm working on the chip distance step by step. I did the North america nebula last night in Ha. Unfortunately, I mucked up the focus a bit, but the effect can still be seen. This is with an extra 0.6 mm spacing. I'm up to about 56.1 mm optical spacing now. Interestingly, the stars top right are starting to look like the spacing is too large. I don't think this is field rotation as I was guiding on a star very near the centre of the field...

...I need to buy some more shims (and focus better!) to find out.

Slightly cropped NGC 7000:

 

[ollypenrice]

Hi Paul, I'd have said you were still OK at the top and were going in the right direction. It certainly looks better. Just a tad more length might see you clear.

Olly

[alan4908]

Hi Paul

Very impressive images, your Esprit 80 is certainly delivering the goods. I was contemplating an upgrade to my current scope and your new addition certainty looks tempting.... anyway, back the subject of optimal distances:

I ran the two images through CCD inspector and obtained the following results:

M13 image: FWHM: min 1.24px, max: 1.33 Aspect ratio:33, total tilt: 2%, collimation: 2.2px, Curvature: 4.8%. Focus looks spot on.

which to me says that the spacing is not quite correct since the aspect ratio is too high, however, everything else is very impressive eg virtually zero tilt and very low field curvature. 

NGC7000: image: FWHM: min 1.75px, max 2px, Aspect ratio: 23, curvature: 9.8%, total tilt: 5%, collimation: 3.0px. Focus is slightly out.

Assuming that your guiding/PA accuracy was consistent between the two images, by adding the extra 0.6mm you are heading in the correct direction and have reduced the aspect ratio to 23.  Personally, I find that I cannot see issues with aspect ratios if they are below 20. As an additional reference, I find with my set up and 600s exposures, I get an average aspect ratio of about 15.  So, I'd suggest you slightly increase the distance. 

You might want to download a free trial of CCD inspector since I've found it a very useful tool to work out when I've reached the optimal set up.

Alan

 

[Pompey Monkey]

9 hours ago, alan4908 said:

Hi Paul

Very impressive images, your Esprit 80 is certainly delivering the goods. I was contemplating an upgrade to my current scope and your new addition certainty looks tempting.... anyway, back the subject of optimal distances:

I ran the two images through CCD inspector and obtained the following results:

M13 image: FWHM: min 1.24px, max: 1.33 Aspect ratio:33, total tilt: 2%, collimation: 2.2px, Curvature: 4.8%. Focus looks spot on.

which to me says that the spacing is not quite correct since the aspect ratio is too high, however, everything else is very impressive eg virtually zero tilt and very low field curvature. 

NGC7000: image: FWHM: min 1.75px, max 2px, Aspect ratio: 23, curvature: 9.8%, total tilt: 5%, collimation: 3.0px. Focus is slightly out.

Assuming that your guiding/PA accuracy was consistent between the two images, by adding the extra 0.6mm you are heading in the correct direction and have reduced the aspect ratio to 23.  Personally, I find that I cannot see issues with aspect ratios if they are below 20. As an additional reference, I find with my set up and 600s exposures, I get an average aspect ratio of about 15.  So, I'd suggest you slightly increase the distance. 

You might want to download a free trial of CCD inspector since I've found it a very useful tool to work out when I've reached the optimal set up.

Alan

 

Thanks Alan - It certainly does deliver much, much, more better images than my previous 'scope  and the build quality is rather superior too. The focuser really does allow for minuscule adjustments and the lock appears to introduce no focus shift that I can tell: The focus problem was, I think, due to my impatience at getting going with the Ha. I really need to wait for those 30 second exposures for the Bahtinov grabber to work properly with the narrowband filters.

I have two more spacers on order from Telescop-Service - a 1.5 mm and a 1.0 mm. Between those and the other sizes I already have stacked in the train, I thin I should get there. It's a bit odd that I'm having to extend so far from the 55 mm in the Skywatcher manual though. Interestingly, my plate solves give a FL of 411 mm too (should be 400 mm, of course). I wonder what the tolerances are in the SW factory?...

I did have the trial of CCD Inspector, but I used it up on the WO! If you would allow me be so cheeky as to keep posting my PNG's here for your amusement, I would be very grateful  It's in the numbers man! Give me the NUMBERS!!! LOL!

[Pompey Monkey]

Well, I added another 0. 5 mm to the flattener distance - it's now at about 56.6 mm. Should I keep going? I'm running out of threads to put spacing rings in... lol!

 

[alan4908]

Hi Paul

An impressive image.. having put the image through CCD inspector it informs me that: the focus is spot on, min FWHM 1.76 px, max FWHM 1.88px, total tilt is 2% and the aspect ratio is 36. So, if your PA and guiding are consistent you have gone too far ! Before trying a shorter spacing distance, I'd suggest you confirm that the PA/guider error is not a factor. For instance, on my own Ha of IC1396 taken with 1800s exposures I obtained aspect ratios of  9, 10, 22 and 47. The image with an aspect ratio of 47 was due to a guide error.  

Alan

[Pompey Monkey]

1 hour ago, alan4908 said:

Hi Paul

An impressive image.. having put the image through CCD inspector it informs me that: the focus is spot on, min FWHM 1.76 px, max FWHM 1.88px, total tilt is 2% and the aspect ratio is 36. So, if your PA and guiding are consistent you have gone too far ! Before trying a shorter spacing distance, I'd suggest you confirm that the PA/guider error is not a factor. For instance, on my own Ha of IC1396 taken with 1800s exposures I obtained aspect ratios of  9, 10, 22 and 47. The image with an aspect ratio of 47 was due to a guide error.  

Alan

Thanks again Alan.

I really should have concentrated on shorter exposures, but my enthusiasm carried me away!

Fortunately, I took three 3 minute luminance exposure of the IRIS with the same spacing. Does this give similar results?

 

Edit: To me, the stars on the lower half (especially the bottom left-hand corner) are still elongated with their major axes pointing toward the centre of the image. This, I believe, is symptomatic of insufficient distance between chip and flattener. The stars on the top right, however, appear rounder than in the IC1396 image which, at 600s exposures, suggests that there is some field rotation going on too.

Edited May 5, 2016 by Pompey Monkey
Observation

[gnomus]

If you smply want to test the optics and eliminate field rotation then why not shoot some short 5-10 second subs with the telescope pointing straight up?  (Sara Wager suggested this to me when I had my Tak FSQ 85 issues.) One benefit of shooting straight up is that you elimnate any issues caused by the focuser tilting in the drawtube.

I'm following this with interest as I am contemplating an Esprit 120.  (If anyone has a view on the Esprit 120 please feel free to PM me rather than post on Paul's thread.)

Edited May 5, 2016 by gnomus

[alan4908]

2 hours ago, Pompey Monkey said:

Does this give similar results?

Hi Paul

From CCDInspector: the image gives an aspect ratio of 39, min FWHM 1.65px, max 1.81 px, tilt: 6%, the focus is slightly off.  So, basically a similar result.  I agree with the suggestion by gnomus  eg acquire a very short exposure in order to minimize any PA/guider error. 

Alan

[Pompey Monkey]

59 minutes ago, gnomus said:

If you smply want to test the optics and eliminate field rotation then why not shoot some short 5-10 second subs with the telescope pointing straight up?  (Sara Wager suggested this to me when I had my Tak FSQ 85 issues.) One benefit of shooting straight up is that you elimnate any issues caused by the focuser tilting in the drawtube.

I'm following this with interest as I am contemplating an Esprit 120.  (If anyone has a view on the Esprit 120 please feel free to PM me rather than post on Paul's thread.)

 

1 minute ago, alan4908 said:

Hi Paul

From CCDInspector: the image gives an aspect ratio of 39, min FWHM 1.65px, max 1.81 px, tilt: 6%, the focus is slightly off.  So, basically a similar result.  I agree with the suggestion by gnomus  eg acquire a very short exposure in order to minimize any PA/guider error. 

Alan

Good point guys. I will give it a go

As it happens I found out that I can get the evaluation version of CCDInspector running on my Obsy PC (for 30 days only of course) - it was on my processing PC before.

Interestingly this single 2 minute sub shows almost no tilt but 11.8% curvature: the target, M13 was at about 45 degrees elevation at the time. This does, at first glance, point to no real focuser slop and other errors due to PA. However, there are very few stars near the edges, so I'll need to compile a proper dataset  - Watch how the clouds will roll in for the next few weeks...

L_L_2016-05-04_22-36-22_Bin1x1_120s__-20C.fit

[gnomus]

I remember getting an evaluation copy of CCD I when I had my Tak issues.  It may interest you to know that the FSQ 85 that I had (and which Tak Europe said was A1) had a field curvature of 18.7%.  My ED80 was just under 11%, which seems comparable to your result.  I have no idea what any of that means by the way.

There is still some slight issue in your images but, for what it's worth, I think they are just about 'good enough'.  Certainly your Elephant was pretty close.  It is possible that you have some sensor tilt in the camera or camera/wheel combo.  Have you considered that? You could also consider takng some shots with a DSLR to see whether you get the same pattern on two different sensors or rotating the current camera 90 degrees.

[Pompey Monkey]

On 4/30/2016 at 22:58, gingergeek said:

Well I like it ! 

Bleedin' editors crook again

Gnomus said:

I remember getting an evaluation copy of CCD I when I had my Tak issues.  It may interest you to know that the FSQ 85 that I had (and which Tak Europe said was A1) had a field curvature of 18.7%.  My ED80 was just under 11%, which seems comparable to your result.  I have no idea what any of that means by the way.

There is still some slight issue in your images but, for what it's worth, I think they are just about 'good enough'.  Certainly your Elephant was pretty close.  It is possible that you have some sensor tilt in the camera or camera/wheel combo.  Have you considered that? You could also consider takng some shots with a DSLR to see whether you get the same pattern on two different sensors or rotating the current camera 90 degrees.

 

Me neither!

I may be being a little over sensitive but, given the struggle I had with the WO (I completely wasted two months of winter imaging time), I think this is normal. "Good enough"? What is good enough for a £1000 telescope that is supposed to be flat over an imaging circle of 33 mm? The KAF 8300 is only about 26 mm across the diagonal. I had round stars across the field on my 150 PDS with its bargain basement crayford focuser at f4.5. Maybe it is because the stars are smaller that the defects are more noticeable?

Maybe the more I spend, the more critical I'll be - Alas a recipe that will doom me to disappointment for all eternity- I think I'll avoid that path!

Next time it is clear, I will try rotating camera and see what happens. In the meantime there is cold beer in the fridge. Cheers!

[gnomus]

I hear you.  I dropped the best part of 3K on the Tak plus bits, in the hope that I could take 'poor optics' out of the equation.   I was dismayed when it didn't work.  Once I returned the Tak I got a WO 71 that was awful.  It has left me somewhat sceptical about the claims made by manufacturers and retailers.

However, you can't get perfect.  Your Elephant sub looked fantastic.  No-one is going to be peeping into the corners of that image at 100%.  And even if they did they would be hard pressed to see very much amiss.  

Edited May 5, 2016 by gnomus

[Pompey Monkey]

3 minutes ago, gnomus said:

I hear you.  I dropped the best part of 3K on the Tak plus bits, in the hope that I could take 'poor optics' out of the equation.   I was dismayed when it didn't work.  Once I returned the Tak I got a WO 71 that was awful.  It has left me somewhat sceptical about the claims made by manufacturers and retailers.

However, you can't get perfect.  Your Elephant sub looked fantastic.  No-one is going to be peeping into the corners of that image at 100%.  

Thanks gnomus

"Sub"? 21 600 second exposures dark and bias calibrated and stacked! (just no flats yet)

[gnomus]

1 minute ago, Pompey Monkey said:

Thanks gnomus

"Sub"? 21 600 second exposures dark and bias calibrated and stacked! (just no flats yet)

Oops - sorry if I misunderstood.  Terrific stack nonetheless.  I don't know if you can expect much better than that - that is what I meant by 'good enough'.

If you have a moment, go to Astrobin and search for images taken through an expensive scope of your choosing (one beginning with 'T' if you like).  I think you will find many more corner defects than you have in your stack - and most of these images are not posted at 100%. 

[Pompey Monkey]

Well, I finally declare myself "happy"

I added another 1 mm to the train and I managed to get out for an hour tonight between clouds: The elongated stars now appeared on the other side of the image! Not quite what I was expecting fro the received wisdom (I was expecting them to go elongated at at right angles to what they actually did).

So I then reduced the spacing to only 0.5 mm over what I had for the Trunk. I took 20 second images of a stars almost overhead (Alioth, I think) to eliminate guiding and field-rotation errors.

I present a stack of 6 of theses 20 second exposures here. Not quite perfect, but good enough for me - Thanks for the realism injection, Gnomus.

Incidentally, without going into details, what CCDInspector and what my eyes tell me don't necessarily correlate!

 

Edited May 7, 2016 by Pompey Monkey

[gnomus]

I think I'd accept that.  It's great when the 'testing' finally comes to an end.

[Pompey Monkey]

10 minutes ago, gnomus said:

I think I'd accept that.  It's great when the 'testing' finally comes to an end.

I was talking to my (far too grown up) son today about this issue: To paraphrase, he said to me, "What does it matter if a few stars are a bit out? Just remember what amazing astronomical sights the process is showing you, that you'd never get to see with your eyes alone."

Kids. Who needs them? Meh

 

[Pompey Monkey]

Well I don't know what's going on anymore!

Last night I started some Ha of M101 just before astro-darkness. I took great pains to get the focus right but on inspecting the first subs I got this:

Eggy stars again!

Edited May 15, 2016 by Pompey Monkey

[Pompey Monkey]

So I did some 3 minute L subs of the galaxy and got this:

Yep, eggy stars.

[Pompey Monkey]

So I rotated the camera and filter wheel and took another 3 minute sub (this is of Alioth, eta Ursa Major, very close to M101). The eggyness moves, so any tilt must be external to the camera.

All of the last three images were taken with the scope pointing to within a few degrees of the azimuth, so that should rule out focuser slop. My PA (not that it should matter at these exposure lengths) is good to 0.12 minutes, thanks to a Polemaster, so field rotation is not the issue.

The flattener to chip spacing is the same as for the North America nebula above which, according to Alan, gave the best numbers in CCD inspector, although I confess to not really understanding these yet.

So what is causing the issue?

Time to contact the vendor, methinks. I'm glad I bought it from FLO