Collimation and imaging train tilt

After the CalStar desaster, I decided to try to improve focus, collimation and imaging tilt.

First, I wanted to improve collimation. In order to remove as many sources of errors as possible, I started just with eyepieces.

1. Visual
First using the 40mm eyepiece and then the 4mm, I collimated to scope as good as possible out of focus. Seeing wasn't great - when I tried to use in-focus collimation, I could hardly see any Airy disk.

2. Camera directly on scope
I removed the focal reducer, spacers, filter wheel, OAG. I couldn't get the scope to focus anymore, but that was OK. I used CCDInspector's out-of-focus collimation. I used ζ Cygni - it was very high up and has no other bright stars around it. Used 30 sec exposures to minimize atmospheric and other distortions. And always waited at least 3 images with similar errors.
I started with a 16" collimation error! I turned the camera around to make sure that the error that I'm seeing is purely the scope and not the camera. Moved it slowly down to 4". Then it changed direction. And then it got funny: for the other 2 collimation screws, the star moved away from the screw when I tightened it. But for the third, it moved towards the screw when I tightened it. That certainly explains the difficulties that I had to improve collimation earlier. After I figured this out, I could bring collimation below 1" (then it changed direction from picture to picture).
Also, at this point the image had no tilt at all!

Now I could have slowly added equipment to it.

... but instead, I'd rather focus on Hyperstar imaging. I know now that the camera is completely orthogonal, so I can use it for collimation of the Hyperstar lens. So, reconfigure the entire scope (cables, secondary mirror, hyperstar lens).

Collimating the Hyperstar lens turns out to be a major pain in the ...
I have to use multi-star collimation - because of the wide field of the Hyperstar lens, I couldn't find a brigh enough start that didn't have other stars in its vicinity. Furthermore, to defocus a start enough for out-of focus collimation with CCDInspector is outside of the range of my focuser!
Because of stray light and the moon light, I have to use a shield on the scope to get correct collimation. But that means that I have to take the damn shield off whenever I need to adjust the Hyperstar lens, then put it back on.
This is the initial state:

Tilt and quite a large collimation error.

This is the final result:

Almost no collimation: good!
Almost no image tile: good!
But look at that curvature!!! I have absolutely no idea what to do about that...

CalStar - 3rd night

After last night, I had high hopes to make good progress on imaging. First to finish imaging the Fireworks Galaxy. And then start imaging the Trifid of the North.

Setup went well. Starting taking the remaining images of the Fireworks Galaxy. Then I woke up in time to check if SGPro started imaging the Trifid. Looked good. Went to bed.

Next morning, I checked the images. Fireworks looked good. Luminance of Trifid looked good:

But the RGB all looked like this:


I guess, the battery that powered the imaging equipment and the laptop wasn't fully charged and the camera or the USB hub dropped pixels.

After something major went bad every night, I decided to call it quits and not stay for a fourth night. Also, I was really yearning for a warm bed!

CalStar - 2nd night

I spent the day trying to figure out what is wrong with my scope:

• I checked the flatness of the image - minimum tilt
• Cleaned the OAG prism
• Cleaned the corrector plate (very careful - there was a lot of sand and dust on it)
• Took exact latitude and longitude from the GPS system
• Checked all mechanical connections on the mount

Then at the beginning of the night, I continued to focus on improving the guiding (if I can't figure it out, I could go home).

I used AlignMaster for polar alignment again. I did 3 iterations and in the last iteration, the correction was minimal. Verified collumation - good (below 1").

Mark Scrivener and John Wainwright gave me a couple of good tips:

1. Use the ASCOM driver for guiding - not through the RJ45 cable
2. Disable guiding commands in PHD and just see where the guidestar moves
3. Guide (or test) with the main camera - not the Lodestar. This rules out any issues with the OAG or such.

I calibrated PHD near the equator and ran it without guiding. It showed an almost perfect polar alignment - even after 10 minutes, the trendline looked completely horizontal. And the RA trendline went up very straight and smooth. So, the scope is fine, polar alignment is great. Repeated the same with the lodestar - same result. So, the OAG is fine too.

When I was almost ready to declare defeat, one of my neighbours came over to me and asked what was wrong. I walked him through everything. He then wanted to look at my PHD parameters. And he pointed out that my "Max Duration" was too low (I had it at 200). He recommended to set it to 800...

... and voila! Suddenly guiding errors dropped below 0.2 arsec!!! AAAAAHHHHHHH!

I then started guiding again. Watched it for a little - looked really good.

I woke up at 1am to check if SGPro started imaging the 2nd target - worked!!!

And then in the morning, I found out that at 2am guiding completely stopped!!!!!! NOOOOOOOOOO!!!

But the images from the first target were now MUCH better:

This is much better. Although for a 1x1 binned luminance subframe it's still a little blurry.

CalStar - 1st Night

I arrived at 4pm, setup my tent and my scope - all without a hitch. Yes - all ready to get going once it turned dark.

When I connected the Mach1 scope, I couldn't remember which location number was CalStar - it would be awesome if I could assign names... Next surprise: the Mach1 mount can't run off the same battery as the rest of my equipment! The motors sounded horrible and slews stopped half way there. So, I needed to use one battery for the mount. And the other battery for the rest of the equipment. Had to improvise to connect the second splitter to the second battery. And I hope I can recharge both batteries enough tomorrow.

Next, I did a rough polar alignment with the RAPAS, star sync and then wanted to do a more accurate polar alignment with PEMPro. But it complained that my camera was rotated in a bad way and that it couldn't do it. No idea where that came from.

So, I used AlignMaster instead. Not sure how exact this was as I needed exact longitude and latitude of the place.

Next, I wanted to work on collimation with CCDInspector. I could improve it, but at some point the results didn't make sense again. I think I got it somewhere below 1" - good!

And finally PHD...
... complete desaster! I have no idea what was going on with that. I could barely get the accuracy below 0.6" and very erratic. In San Jose in my backyard, I could get it below 0.4". At some point - it was almost midnight by now- I got too tired and too irritated to continue trying things out. I setup the scope to start imaging anyway. Maybe guiding would improve later that night.

I got up in the middle of the night when SGPro had to slew to the next object. And PHD could not resume - didn't find a guide start. I checked everything. Finally I took new darks for the autoguider (this time from a REALLY dark side) - and that fixed it! Guiding by now was a little better - but not by a lot.

Next morning, I checked the images - they were all very bad.

This is a 1x1 binned luminance subframe :-(

Clearly the guiding is not suitable. Need to figure out what to do about PHD. I compared my guiding graphs from last night to guiding graphs at home - they don't look that much worse. But much more erratic.

Now I had to use my solar panels for the first time. Needed to cut some cables and make some connectors and got one battery to recharge. I hope that the other (the one that powered the mount) doesn't need recharging.

Oh, and: IT WAS REALLY COLD! I do remember that I was camping once in England when it froze and snowed. But last night felt worse. I guess after 20+ years I need a new sleeping bag :-S

New Focuser

Somehow the Robofocus controllbox broke - can't use it manually anymore, can't use it remotely anymore...

I was pretty annoyed by Robofocus (too many cables, fragile cables, the bulky motor construction in the back of my scope, backlash) anyway. So, I used this as an opportunity to switch to a MicroTouch controlled Feather Touch focuser. The other advantage of this is that I have a great manual focuser if the motor or motor control fails.

I ordered it from Starizona and it arrived some 10 days later (if you order Starizona, don't try to use their order status to understand the status of your order - it doesn't get updated!)

Installation was a breeze. First I installed the Feather Touch focuser, then I removed the turning nobs and installed the MicroTouch focuser on top. Took me a little over an hour! And now I have just a little knob back there and not the whole Robofocuser arm!

The red circle shows the focuser + motor. Compare this to the clunky Robofocus arm in this post.

Next, I wanted to install the MicroTouch control software and ASCOM driver. Installing the MicroTouch control software was easy. But I could not use the ASCOM driver - I always got this error message when I tried to select the MicroTouch focuser in the ASCOM chooser:

"Incompatible Driver (microtouch.focuser)

This driver is not registered for COM (can't find ProgID), please reinstall."

I tried various combinations:

• Installing the MicroTouch control program first or later
• Installing the driver as admin or not
• Disabling User Account Control

Nothing. Mailed Dean from Starizona and received the following instructions:

Make sure you run the installation MicroTouchUSB2 setup by right clicking and running as admin.  Then run the MicroTouch the application not the setup itself by right clicking and run as admin (required first time only).

Did that - and now I could use it from SGPro!!! Yei!

But not from TheSkyX :-( I could select not the MicroTouch focuser in the ASCOM control box, but couldn't click "OK" (it was disabled). When I selected "Settings" - I always got an exception.

Searched the TheSkyX forum and found this post: http://www.bisque.com/sc/forums/p/18147/76294.aspx#76294. I downloaded the zip file mentioned which contained a native driver for the MicroTouch focuser.

After I installed it, I had a "Starizona" entry in the focuer chooser dialog. But when I selected it, I received another error message.

... strangely enough, when I went back to the Ascom focuser and chose MicroTouch focuser in it's chooser, it worked (the only strange thing was that I got a popup asking me if I have a USB or USB2 controller - and that popup box opened BEHIND TheSkyX). Well, I take it!!!

Next, I played with the automatic focusing routing in SGPro. I got really weird results: my graph always looked very flat with some ups and downs. Took me a while to realize that the step size is set by default to 1, i.e. there is VERY little focuser movement compared to the Robofocus (where I had set the step size to 8!). The MicroTouch focuser has 60000 steps. I.e. with a step size of 1, I won't be able to move the focus all the way in and out - but I pretty much never need that anyway. So, I cold keep this step size and allow for very small adjustments (could be handy when I image with the hyperstar lens which has a very small critical focus zone) or I could set it to a higher value and allow for wider range.

And then I wanted to train the temperature compensation. From the manual, it looked VERY easy: focus, start training, wait until the temperature drops, focus again, stop training. Now I have a compensation ratio that the MicroTouch controller would apply automatically when the temperature changes.

... but the temperature did not change during training! Well, it went a little bit up and down, but it did not change significantly. And I started the training at 9pm and waited until midnight - and it was definitively cooler then. 

First narrowband image

I tried to capture the veil nebula from our backyard, but the noise level was just very bad. So, I tried to take the same images with my Ha, SII and OIII narrowband filters. There are various formulas out there how to combine these, e.g:

R: 0.8*Ha + 0.2*SII
G: OIII
B: 0.85*OIII + 0.15*Ha

another formula that I found:

R: 0.4*Ha + 0.6*SII
G: 0.4*OIII + 0.3*Ha + 0.3*SII
B: OIII

Guess, I'll have to play with these.

I wonder if I should create the 3 color channels in CCDStack with the respective weights and combine them there. Or if I should just stack the individual images and then combine them in Photoshop. I'll try CCDStack first. An email to the CCDStack support mailing list pointed me to File Math or Combine with Pixel Math or Weights.

First, I calibrated all frames. Because I took them over several nights, I used a different strategy:

1. Calibrate all images per night and store the calibrated frames
2. Rotate (if necessary), align, correct and stack all subframes
3. Align all 3 resulting frames and store the aligned frames
4. Now, I can load Individual aligned frames into CCDStack to combine them. E.g. for the first hubble palette, I load the aligned Ha frame and the aligned SII frame. 
5. With pixel math, I multiply the Ha frame by 0.8 and the SII frame by 0.2
6. Combine-Sum the resulting frames for my red image
7. Repeat for the other colors
8. Finally combine the 3 frames to a color image.

Here are the resulting images:

I also tried to just use the individual images for individual colors: OIII=Red, Ha=Green, SII=Blue:

I like the last one the best, it shows lots of fine details in the nebula.

I processed the other image in the same way:

So far so good...

Now, I have to do some more processing:

• there is still a lot of background noise
• need to calibrate both images to make them blend better together
• and then I need to stitch them together

Another imaging automation program: Sequence Generator

By chance I read about Sequence Generator Pro. It's another imaging automation program - for WAY less then the programs that I looked into before. One of the most interesting features for me is that it integrates with PHD guiding. I was always struggling with TSX guiding. The other interesting "feature" is its price: $99!

I downloaded SGPro, installed it, read the helpfile and watched some of the tutorials. It has a very different concept then CCDAP:

• It's based on profiles which describe the system setup (scope, camera, guiding, plate solver, filter wheel...) This is actually very good for as I will image with at least 2 configurations (hyperstar and focal reducer).
• It doesn't integrate with TheSkyX which means that object selection isn't through a planetary software, but the "Frame and Mosaic Wizard" is actually really good.

It's definitively somewhat rough around the edges - many things have to be manually configured or installed. It took me a week to configure and setup everything:

• The frame wizard relies on an external service (which was down for a day). But the good news is that it caches all downloaded data and images, so I can use those later again.
• Automatic focussing: it's a very straightforward process. It took me a while to find the right step size though. I ended up running the focus routine with a step size of 2 and checked how many points were in the critical focus zone (the plateau of the curve).
  I then had to play with each filter to figure out which exposure time to use. I increased the exposure time step-by-step until the focus curve was smooth.
  I had to set "stop guiding during focus routine" - because I'm using an OAG, autoguiding is naturally severely affected if I change the focus too much. The last remaining issue is, that SGPro instructs PHD to select a guide star before the focus routine, then it stops guiding, runs the focus routine and restarts guiding with the previously selected guide star. But now, the guide star isn't in the same location anymore and PHD has to do some crazy guiding to get it back into center. I asked on the mailing list. Jared (one of the developers - they are both crazy active on the mainsequencesoftware mailing list) didn't have an idea how to accommodate for this, but will think about it.
• I used this as an opportunity to start using PHD2 - which works very well. To avoid guiding on hot pixels, I have to use darks. A great advantage of PHD2 over the previous version is that it can re-use darks for subframes and doesn't try to take a new one. I had to set the threshold after which SGPro starts taking images to 0.5 - that's already pretty low for my backyard conditions.
• SGPro uses plate solving for every slew - which is great. Makes it very reliable. Initially, I tried to use Elbrus, but it didn't always work. I then switched to astrometry.net which works great. But now I am relying on an external service - which is bad when I take the scope out. Fortunately, there is a local astrometry.net setup that I can use.
• It was strange that I had to set the "auto meridian flip" bit - I had expected that it is set by default.
• SGPro reads the configured ASCOM devices and offers these directly. That avoids the popup when connecting where you have to choose the ASCOM device. Nice!
• Naming of the images is great!
• I tried mosaics, meridian flips during imaging, multiple targets - all works very well.
• The built-in Image Grader is great for a quick analysis of acquired images.

The only missing thing (for me) are dusk/dawn flats. It was very convenient with CCDAP to just setup the scope, rough align it and then start the CCDAP flat routine. It would wait until it's dark/bright enough and run through all the filters, dither and adjust exposure times. I wish SGPro would implement this as well.