Very Amateur Astronomy

It’s been a long road over the summer to get to this point, but I think I finally have this particular nut cracked. This is a single 6 minute frame of the Orion Nebula, unprocessed, washed out by the moon and as it arrived off the camera.

I’m pretty happy with that, it’s not perfect, there is a slight stretching of the stars. But this one one of 10 test images and they were all of equal or better quality. The stretch is along the RA direction which means the likely causes are periodic error and over correction by the guide software. The dec axis is working beautifully despite it’s frustrating backlash.

Experimentation has shown me that the automated aggression control of guidemaster doesn’t function very well on this mount, disabling that option helped enormously.  Adjusting down the RA autoguide movement speed also reduced the stretch. The next testing step will be to do the PEC recording, however Vixen really need to improve PEC on the starbook, I don’t want to have to do it every time I want to do some imaging!

This is addition to my previous post really, as I had taken more images that I had not processed on Tuesday night. First off though I found my missing mosaic video section! This means I’ve been able to complete the image properly now, without the missing chunk. I’ve also left the colour in and applied a small gausian blue to smooth off some of the sharper edges.

The other data I wanted to process was a set of 30 full frame raw format images of the moon taken with the canon 1000D. These were then stacked in registax, tweaked with wavelets and then dumped out to gimp to adjust the colours. In this case I pushed the saturation levels right up, this brings out more of the subtle changes in the moons surface that the human eye really only see’s in shades of grey but the DSLR picks up.  It’s an effect that I really enjoy and not possible with the mosaic above where every individual section has it’s own balance and variation.

We had a clear night!

No joking, it was clear, in England all night on the 30th, and properly cold as well. It’s been so long since we had a good clear night that I had the roof off the shed just as soon as I could get out outside. The only real damper on this otherwise perfect evening was a big bright moon, which of course ruined any chance of me doing any deep space work. But I have a webcam, and I have EOS movie record, so I figured I would make the best of the night and try my hand at some lunar imaging.

The first thing I wanted to try was capturing a few minutes of video from the Canon 1000D using EOS Movie Record. I set that up and left it running whilst I got myself sorted for the night. A few minutes in I noticed a flicker on the screen from the corner of my eye and made a mental note to check it the next day. This is what I found when I reviewed the video :

So that was a bit of pure luck on my part, but it makes for a striking image I think.

Next I setup the webcam on the 250mm reflector to try my hand at lunar imaging. At first I thought I would just stick to one area but after a little while I had started to collect connected sections, and from then on it turned into an attempt at a mosaic. This was my first attempt doing this and unplanned so naturally I missed a bit, which I shall kick myself for later.

• Philips NPC900NC
• Baader UV/IR cut filter
• Orion Optics 250mm reflector

The image is composed of 27 individual frames, each of which came from a 1000 frame avi video, the data was captured with k3ccdtools and stacked in Registax.

In order to build the complete image I first spent a long time in photoshop and came out with something I was happy with. Then out of interest I tried the photostitch tool in Windows live photo gallery. In about three minutes it had aligned, colour corrected and stitched the whole image together far better than I ever could. I was really impressed as about a third of the frames were at an angle compared to the rest due to the scope reversing half way through. So if you have images to align, give it a go, you don’t need to do anything more than select the images and click go, it works out everything else by itself.

The full scale image is some 3360 pixels across and I shall locate some hosting for that soon.

The apparently never ending bad weather has put a real dampener on astronomy related activities over the last month or so, clear nights have been few and far between, and when we get them I always appear to be somewhere else!

The south east of England took a good battering over the weekend from a storm, wind speeds were expected to gust up to 85mph. A good test for the shed as this was the worst weather it has had to face since construction, I’m happy to report that everything stayed intact, though I have spotted some ingress of water around the window, another job for the weekend.

Post storm I spent some more time trying to improve the declination backlash of the Sphinx SX mount, with mixed success. I stripped the mount down as far as possible and spent time trying to understand how it is held together and what needs to be adjusted. I adjusted the worm gear as that had some lateral movement, tightened the ring under the declination clamp as I noticed the shaft had some free play, and readjusted the worm to drive mesh.

All this work didn’t quite have the hoped for jump in improvement, guidemaster still reports a declination backlash of 10000 milliseconds! Ten seconds is frankly awful, and I was pretty disappointed that all my hard work resulted in such a minor improvement.

Round about this time is when I noticed guidemaster and ASCOM giving me some odd errors about starbook firmware versions, now this is a brand new starbook received as a replacement for my broken one. One of the first things I had done upon turning it on for the first time was to check the firmware was at the most recent release, and indeed to said v21. However, 21 is not the same as 2.1, a silly mistake on my part and great advert of good practice in logically naming your software releases.

So I updated the firmware and that certainly seems to have improved matters, the declination movement is far more positive now hich should allow me to tighten down the worm / gear mesh a little more.

Slow progress then, but there is movement!

October so far has been a month of frustration, with no end of problems both equipment and software related to wade through. I think there is a light at the end of the tunnel now, and most issues seem to have found a fix. Lets rewind to the start of the month however, and the next clear night after my triangulum image.

It had been my hope that after two nights of testing and work that I might finally have had all of my various issues with guiding sorted out, and so I went out thinking that this night would be an error free night of excellent imaging. Almost immediately however my carefully tested guiding software threw a tantrum and started giving very poor results on nearly every image, odd shaped stars and nasty drifts all over the place. This was not good, the problem appears to be in the declination axis and guidedog didn’t seem to be handling it.

This triggered another night of testing and adjusting, including another session with WCS to get the alignment as perfect as possible, I think I will have to register that software. I ran through every free autoguiding software solution I could locate.

• PHD Guiding – So much promise but it hates modified webcams even at 5fps.
• Guidedog – Lovely software, easy to use, but it doesn’t appear to handle dec backlash.
• Proguider – Looks good, but I couldn’t work it out, needs better documentation.
• Guidemaster – Bingo!

I eventually tried out Guidemaster, and found software that not only does some excellent calibration routines to work out declination backlash, movement speeds and over useful information, but also manages such useful things as automatic aggression adjustment and automatic setting of single direction declination adjust. That last one is great, declination will normally only drift off one way, so if it only corrects one way then it will always be loaded on the gear and backlash won’t be an issue.

Unlike PHD, Guidemaster has no problems with dropped frames from the SC modified webcam, a real issue with PHD to me. I also liked the calibration routine and the ability to adjust the calibration movement time on both RA and Dec individually, really handy if you are suffering from a lot of backlash.

Once calibrated I ran several five minute test exposures and was massively impressed with the result, it appeared to be guiding far more accurately that I had managed with the other programs, I have to say that it gets my recommendations hands down. It did also however, highlight the issue with the declination axis that had caused Guidedog so many problems. Over 5.5 seconds of backlash and the calibration routine needed 9 seconds of declination travel before it was able to register the movement.

Over the weekend I found the cause, when setting up for an image the declination axis stopped moving entirely. This was a major concern and I immediately stripped the mount down hoping that it wasn’t anything major. It turns out that it was the worm gear assembly, which sits in a captive bolt which is adjusted to restrict the linear movement of the gear whilst allowing it to rotate. For some reason the captive bolt had seized onto the worm shaft, now everytime the declination axis was rotated it was slowly screwing the bolt further into the assembly and eventually clamping itself so tight it could no longer move. Luckily no damage was done, I was able to release the captive bolt assembly and took the time to clean and re-grease it.

Unfortunately that was the end of clear skies for now, I’m really hoping that it was the end of my problems as well, time will tell.

A weekend of mixed results, though with two excellent nights I had hoped for more, however I tripped over myself a bit and made some mistakes plus I had some interesting software problems. Perhaps ‘annoying’ would be a better description than ‘interesting’.

The plan was to pick a target and use both nights to do a proper study, my experience with imaging andromeda has taught me to spend more time and gather more images. More frames means more details and a better signal to noise ratio. However before I could do anything I needed to get everything setup and working again.

First thing was to setup the guide scope, previously I have used the Vixen A70LF for this purpose but decided in this case to try out the William Optics ZS66SD instead. I also had to get the mount properly aligned. All of this took a while and I was running a bit behind schedule by the time I started imaging.

This was the point I ran into problem number one, though it took me a while to work out the cause of the issue. Basically the guiding was terrible, lots of shift in various directions causing a bunch of nasty issues. there were a few causes for this :

Guide Scope Setup

I had used the ZS66 with the webcam, but had ran into problems getting the webcam to reach focus. Without thinking about it I used the 0.6 focal reducer to make life easier, without doing the math. This made the field of view of the guide scope way wider than that of the imaging camera. Not good, the guide software had problems making the scope move enough to register a change with out the adjustment speed being pushed right up and that meant it was constantly over running. I fixed this by using a diagonal to increase the focus length in order for the webcam to focus without the reducer, but I still think the zs66 is to wide field for the purpose.

Software Problems

I had originally used PHD Guiding for control, but I kept having problems with it even with the issues above resolved, and on the second night when I changed back to the Vixen scope for guiding. I found PHD was constantly dropping frames from the webcam, sometimes several in a row which was resulting in poor guiding. Even when it did seem to be guiding correctly the images coming back were of very poor quality with unexplained shifts (with the ZS66 anyway may be the field of view again). On the second night I tried Guidedog instead. Whilst the guiding was not perfect, it was certainly much better than I had got out of PHD.

So lots of testing, setup, changes, adjustments and general mucking about resulting in me only getting just over 2 hours 20 minutes of data back. Not quite the 8 hours I had been hoping for. The result is not one of my best :

Triangulum Galaxy

• Orion Optics 250mm F4.7 Reflector
• Vixen Sphinx Mount
• Canon EOS 1000D @ iso 800
• 29 x 5 minute exposures (plus darks & flats)

Not my best work, but it will be improved with the addition of more data on the next clear night.

I had another go at the ISS last night, this time with the 2x barlow in place in an attempt to achieve higher magnification :

• Exposure 1/1500
• Gain 50
• Brightness 60

I think that’s gone a bit above the limit of the webcam. I can’t increase the exposure any further without getting blurring, which means I have to increase the gain which in turn produces to much noise. Still it has been a very production week for imaging for me.

The next experiment now I have the UV/IR cut filter is to look at using EOS movie record to capture the AVI through the barlow and see what magnification I get with that combination. The better chip and better control should allow for a better image, but I think any targets will be to small. I may need to keep an eye out for a higher magnification barlow for this option.

I also want to try the webcam, through the barlow but with my 0.6 reducer attached as well. I may well be getting carried away there though. The ideal would be an imaging source camera, add yet another item to the shopping list!

Continuing my exploration of webcam astrophotography I have had a couple of goes at capturing the ISS as it has been very bright for us in south east England over the last few nights. The best efforts came from last night.

Some good detail captured there, so I’m happy with that! The equipment and settings used are as follows :

• Orion Optics Europa 250mm Reflector
• Manually guided on the Vixen Sphinx mount
• Philips SPC900NC webcam (Exp 1/2500 – Gain 40 – Bright 60)

I captured around 2000 frames of which about 50 contained the target, any sequential runs of four or more were exported and stacked in Registax. I applied a gentle wavelets filter to bring out the detail, then a very soft gaussian blur to clean them up.

Tonight I may go completely mad and try it with the 2x barlow!

Last night I spent a bit more time working on the Andromeda image in Pixinsight to sort out the colour, boost the saturation and remove a bit more noise from the image. I’m now finished I think, and I have to admit to be very pleased with the result.

Here is a close up of the core cropped from the full size original image. You can see detail right down to the centre.

My next challenge will be to improve my knowledge and skills of the software by reprocessing some of my older images. I can’t wait to see what lies hidden behind my previous dodgy versions!

I’m never happy with my processing, and I’m always looking for a better way. On the strength of various threads on the Stargazers Lounge forum I decided to have a go with the Pixinsight astronomy image processing suite, following the excellent video guides from Harry Page. Here is the result :

You can see I have managed to create a significantly more subtle image, and retained detail right down to the core. I’ve got to say I am hugely impressed with pixinsight, everything just seemed to come together so much easier!