Thursday, July 31, 2014

Focuser Slip! Replacing my Robofocus with a FLI Atlas focuser

After I installed (and aligned) the new camera, I wanted to calibrate the filter. Especially the focus difference of the individual filters (so that I can use the Luminance filter for autofocusing and then switch to another filter). But when I did that, I noticed that I got very inconsistent results. In particular, it looks as if the focus position got smaller and smaller. But not by 1 or 2 positions (which could be explained by it cooling down during the night). No, by 100 postions or more between runs.

After more investigation, I realized that the focuser was severely slipping under the load. I read a lot about this that the Takahashi focuser together with the Robofocuse focuser could slip - though before I didn't experience this. So, I need a new focuser...

After some research I found two:

1. FLI Atlas focuser
2. Feathertouch focuser

The Atlas focuser is more then twice as expensive. But I also read that the Feathertouch focuser could slip under heavy load. Plus in order to user the Feathertouch focuser I would have to completely remove the stock focuser - and with that the guiderscope platform that I used for my guidescope.

I then tried to figure out how to integrate the FLI Atlas focuser. Most straight forward would be to just move the Takahashi focuser all the way in and put it between the Takahashi focuser and the reducer/flattener. But in order to gain focus, I have to extend the focuser less then an inch:

That would not be enough for the Atlas focuser which is a little less then 2 inches thick.

I tried to figure out how to integrate it otherwise. Two options came up:

1. Remove the entire stock focuser and have a custom adapter from preciseparts build that attaches directly to the optical tube. It would have the advantage that it will make the scope significantly lighter (removing the entire focuser and just replacing it with an empty tube). But it would also have the disadvantage that I would have to reconfigure my guidescope.

2. Remove just the rotator at the end of the focuser. The rotator adds ~1.7 inches, i.e. without it and by moving the stock focuser all the way in, I could integrate the Atlas focuser. And I could keep my guidescope where it is.

I decided to try #2. But removing the rotator was quite the task. I first removed the focuser from the scope:

I then locked the rotator down (I had to severely tighten then rotator screw) and with a strap wrench I could get it off:

But now came the next challenge: the thread at the end of the drawtube was NOT a 92mm thread - but a 98mm! Apparently, Takahashi has a 98->92mm downstep ring. I am trying to order this.

In the meantime, I am using the Flattener which requires a longer focus distance. I put the rotator back on the focuser, put the Atlas focuser on it, then the flattener and then the imaging train. As the Atlas focuser has a fairly short drawtube, I tried to get rough focus with the Takahashi focuser and then locked it down.

I then used SGPro to get focus. I had to set the focuser step size to 1000 (which is the maximum!)- and even with that, I needed more then 20 data points to create a good V-shape curve. As soon as I had everything calibrated, focusing worked really well.

One problem was that it took me a few attempts to connect the focuser and a few times the autofocus routine hanged when it had to move the focuser out. I asked on the SGPro forum about it - will need to get an Ascom debug file to help investigating it. The next night, I chose "ASCOM chooser" and everything worked without a problem. Strange!

But the focuser itself is just awesome - positions super exact, no slip, no sagging!

Wednesday, July 23, 2014

New camera - and alignment issues

I finally received my new FLI camera: a Microline 16070 camera. My experience with the PL 16070 (which FLI gave me on loan to try it out) was awesome, so I couldn't wait to try it out.

I connected it using the new preciseparts adapter (without the OAG!) Checked drivers and everything - all worked well. Then I took my first image:

Looks good - lots of detail.

... but at closer inspection, on the left hand side:

Hmmm, this looks like misalignment or a collimation error.

I decided to check it out piece by piece:

1. Is it the scope or the imaging train?
The TOA-130 scope has a manual rotator in the back. By taking first an image with the elongated stars on the left and then rotating the entire imaging train by 180 degrees. If the elongated stars are still on the left, the issue is with the imaging train, if the elongated stars are now on the right, then the issue is with the telescope (collimation error most likely).

Both exposures show the elongation on the same side. Phew - I really don't want to mess with the optics of my scope!

2. Is it the camera?
I rotated the camera in the barrel of the filter wheel. And the elongated stars stayed on the same side. So, I know that it is the camera. Either the connection to the filter wheel or the camera / chip isn't properly aligned.

I took the camera out and tightened the barrel screws really slowly constantly rocking the camera back and forth to make sure that it goes in as tight as possible.

... and with that, the elongated stars are gone!!!

Monday, July 21, 2014

Replacing my OAG with a guidescope

I decided to replace my OAG with a guidescope.
  1. When I started using a non-parfocal filter (the light pollution filter), the guide camera got slightly out of focus when switching to a the non-parfocal filter. The star being out of focus wasn't too much of a problem, but if it was a somewhat dim start, the SNR went down really badly which caused guiding issues.
  2. When switching imaging gear (flattener vs. focal reducer, different cameras) I have to refocus the OAG.
At the same time, I was concerned about overall weight of the scope (that I have to carry inside and outside all the time). I asked around for recommendations, and many recommended 50mm scopes - especially the Borg scopes.

I also asked for mounting this guidescope and Astro-Physics recommended to mount the scope directly in the TAO-130 scope (it has a small mounting platform at the end). I ordered the Astro-Physics mount and the scope from Hutech.

When the Borg scope came, I was surprised that it came in pieces and had to be put together:

But assembly was pretty quick:

The next surprise came when I wanted to put the scope into the Astro-Physics bracket. The bracket has an extra ring at the end that secures the draw tube of the focuser to make sure that the focuser is tight. Took me a while to notice that I can remove the thin ring at the end that prevents the draw tube from slipping out:

With this removed, I could pull out the drawtube and put the focuser into the ring of the Astro-Physics bracket:

And finally put the draw tube back in, screw the holder ring back in place, and screw the whole scope together:

The scope has an adapter that allows to screw a lodestar guider into the back (which should reduce sagging and slack):

Astro-Physics recommends to mount short guidescopes onto the mounting plate of the TOA-130 scope (instead of mounting it on top of the scope on a separate plate):

The mounting kits consists of 2 parts: a mounting plate that goes on the scope plate (left) and an adapter for the guidescope bracket (right):

Assembly was fairly easy, first attach the mounting plate:

Then the adapter on top:
One minor challenge with the adapter was to screw it in tightly. The screw shafts in the adapter are quite small and most wrenches didn't fit in there.

And finally, the scope in the bracket on top:

First night out, I started with focusing the guidescope. But to my surprise I couldn't see any stars! Checked everything (optics, cap off??...) And then I noticed pretty large stars when the focuser was moved all the way in. But I couldn't move it in far enough to get focus. Luckily, I had the black extension ring on the scope (see above). After removing that, I could easily get focus. Phew!!

To get perfect focus, I used a small bahtinov mask that I once bought for my camera lenses. It was a bit tricky to slowly move the focuser in and out and then not change anything when I fastend the screws. It helped to focus on a start that's closer to the horizon, so that the focuser doesn't just move out from its own weight. But finally I got this:


And next was to recalibrate PHD2. I first tried to calculate what accuracy I need for my images.

My two cameras have 4.54 micron (SX H694) and 7.4 micron (FLI ML 16070).

With my two setups (flattener - 1000mm focal length or reducer - 763mm focal length), I get the following pixel widths:

(1000mm focal length)
(763mm focal length)
SX H6940.94 arcsec / pixel1.23 arcsec / pixel
FLI ML 160701.53 arcsec / pixel2 arcsec / pixel

The lodestar has a pixel size of 8.3 microns, together with the 50mm focal length of the borg scope this results in a pixel width of 34.239 arcsec / pixel.

    SX H694        FLI ML 16070    
Pixel width [arcsec / pixel]
Maximum error in guider [pixels]

That's quite some accuracy!!!

I started with the default PHD2 parameters which of course weren't great. I started with entering focal length and pixel size to get the right calibration parameters.

Next, I set the minimum move to 0.03 pixels. I then played with the hysteresis and aggressiveness to smooth out the curve and avoid over- or undershooting. Finally, I used PHDLab to understand if I correct unnecessarily (when the movement is still within the allowed range) or don't correct enough.

During this process, I tightened all screws of the guidescope (mounting plate, bracket holder, bracket screws, scope screws, focuser screws). Finally I had everything nice and tight and could easily stay below 0.5 arcsecs - even in my bad seeing conditions!