Hyperstar lens screwed up my secondary mirror holding

When I wanted to exchange the hyperstar lens with the secondary mirror, I noticed, that the mirror housing was loose and turned easily. That of course, would throw off my collimation each and every time. I contacted Starizona and they are selling a gasket that will keep it in place. Luckily they have a video that walks one through the process of installing it. But I probably will have to remove the entire corrector plate - yikes!

I also contacted Celestron support and asked for advice how to fix this - their response was to send in the entire OTA and they will fix it for me (within the 6 weeks timeframe). Not superhelpful...

After 2 weeks, I received the gasket and went right away to install it. I watched the video a few times and it looked easy enough. But of course it wasn't! :-( After I removed the screws and the retaining ring, I wanted to remove the corrector plate - which looks insanely easy in the video. But it didn't move a bit! After closer inspection, I noticed 4 little screws on the side of the corrector plate. These can be used to center the corrector plate:

So, I loosened all of them a little. And with a little screw driver, I could verify that the corrector plate can now move. But still when I pulled on the secondary mirror housing nothing happened. Upon further inspection, I noticed some rubber that was in three places and seems to connect the corrector plate with the tube:

Turns out, I will have to cut this silicone in order to get the corrector plate out. I did that, after that everything was pretty easy: removed the baffle tube, inserted the gasket in the mirror holder, attached the mirror holder and screwed the baffle tube back in. Then I inserted the corrector plate back. Centered it - that was a little tricky. And finally screwed the retaining ring back on.

Of course, the scope is completely out of collimation now. Will do that at the next clear night.

Accessories for the Hyperstar lens

One problem with the Hyperstar lens is that I can't put the dew cap back on the scope. This makes it harder to take darks, but also for closing it when I'm not using it. I looked around and found solutions for both:

1. A "dark" filter
This is basically a solid 1.25" filter that can be inserted into the filter wheel or the filter drawer. And then I can take darks. That is actually even easier then putting the dust cap on.
This filter will help with taking darks in general. Especially for the @focus routine from TheSkyX, which requires 2 darks to be taken per focus run. And since build 6582, I can name this filter "!Shutter!" and TheSkyX will use this filter automatically to take darks!
Although TheSkyX should actually be able to use the internal shutter of my H694 camera to take darks. But there are several threads on the Software Bisque forum where users report issues with that...

2. An aluminum dew shield with a dust cap
That will not only protect the scope from dust, but also protect the hyperstar lens that's now connected to the corrector plate - I'm still nervous about that. Hopefully, I can still exchange filter drawers with this dew shield in place.

The latter might also be helpful when need to take flats for the hyperstar lens with the flat panel. Although I'm pretty sure that the flat panel will be way too bright for the hyperstar lens. I ordered a 12 Volt dimmer, so that I can adjust the brightness.

Trying out the Hyperstar lens

There are still issues with controlling the filter wheel from TheSkyX. So, I decided to use the Hyperstar lens for now (if I have to exchange filters between exposures anyway).

I removed the camera, filter wheel and OAG from the scope and moved the lodestar back to the guidescope. Then I removed the secondary mirror. But I unscrewed the wrong ring, and the ring that's inside the corrector plate fell in!!! Yikes! Took me almost an hour to figure out how to fix that again. Once that was done, I attached the hyperstar lens and the camera.

Of course, focus was completely out. It took me a while to figure out the new focus range. After that was done, I could do the 2-star alignment, calibration and polar alignment. Unfortunately, it's not as exact in this setup because of the wider FOV. I didn't bother for now to figure out the right parameters for @Focus2, but focused manually with the Bahtinov mask (furthermore, there is a loose connection somewhere in the Robofocus cables and I had to do the fine focusing manually :-(

Once all that was setup, I started to image M42. I could take Ha and Red images, then clouds rolled in :-( According to the weather forecast, I will have to wait several days until I can finish this.

... weather didn't get much better. So, here is at least a 60sec Ha image (stacked from 21 images, no darks/flats/bias, no post processing):

Hypertuning my CGEM mount

Because there is still drift when I'm imaging, I decided to hypertune my mount - based on Ed Thomas kit and instructions. The kit comes with a DVD from Ed Thomas where he walks one through the process. He is great in explaining bits and some tricks - especially how to disassemble the mount.

I started with taking the mount apart and everything went pretty well. At one of the ball bearings, the rubber cap was slightly misaligned and I could feel resistance when I turned it. I wiggled the rubber cap in place, but to be on the safe side, I ordered a new one.

But then I came to the point where needed to remove the worm gears from the Dec and Ra housing. The rings on the side of the spur gear were impossible to remove. After trying (and almost ruining one) for some time, I decided to leave them in. I could remove the worm gear anyway - hopefully that wouldn't make it difficult to tune it later as I won't be able to move the worm gear left and right.

The next problem was to remove the spur gear from the Dec worm gear. The set screws are apparently almost glued in. And one of them, I stripped! Luckily they were standard 4mm M4x0.7 screws (cone). So, I ordered a new one, waited to make sure that it really fits and then extracted the stripped one. I also ordered a set of allen wrenches that Ed Thomas has in the video. The set that is in his kit is pretty good - but for these glued in screws you need A LOT of torque.

So, I cleaned all the parts and then tuned the two ring gears. From Ed's video, I expected that I will spend an entire day sanding them down. But after an hour, they both fit really well into their housing. Polished and cleaned them (again).

Next, I screwed the motors back in and then started to measure the RA axis. I was amazed how easy it was to get all the ball bearings back into the ring gears after removing all that freaking grease that Celestron put into it. Putting the worm gear back into the housing was fairly easy - except for sliding the spur gear onto the worm gear. It somehow got stuck and I had to carefully hammer it into it. But once that was done, the rest was fairly easy. Then I started with the DEC axis, but could only go so far until I get the worm gear and spur gear back from Ed.

I used the break put all the new knobs from the knob upgrade kit (except the clutch lever knobs).

And finally, I received the disassembled RA worm gear from Ed Thomas. Boy, am I relived that it's pretty much undamaged. Can't wait to continue the tuning ... but it snowed in the Sierra's, so this weekend we are back up to Northstar for some snowboarding ...

Finally, I could reassemble the mount. This was surprisingly easy. The only setback was that the RA axis did not turn easy enough and I had to sand down the ring gear a little bit more (and of course disassemble and degrease everything again). But after that, the axis turned much easier. I then attached the new clutch nobs and put everything together.

Final result:

• I sanded down the ring gears until they are less snug. In fact, now they have a tiny bit of play when I put them into the housing without any lubricant. So, I was hesitant to do more. The don't seem to have any play when I put the lubricant in there.Both axis turn now much easier then before. But the don't spin like they do in Ed Thomas' video.
• Adjusting the motor location and worm gear backlash was quite a challenge (I went forth and back several times). I could not get either axis into a state where a) there is no backlash, and b) the motor sounds "good" (compared to the sound in your video).
• The DEC axis has no backlash at all, the RA axis very little.
• Both motors make a little bit of the "wha-wha" sound. If I loosen the gears, the sound goes away, but then I can feel more backlash.
• When I turn the motors now, there is some kind of faint clicking sound. I looked everywhere but could not find a source for it.
• When I alternate press the up/down or left/right button very fast, I can hear some clicking sound. Which I think comes from the two spur gears having just a little bit of play. But when I move them closer together, the motor locks up.

The mount definitively "feels" much better already. But I'm not sure if I should try to tune this more (and how). So, I checked with Ed Thomas.

• The ring gears are probably OK. As long as the axis turn easy enough that I can balance the scope properly.
• The CGEM mount will always have a little bit of backlash - it's just not manufactured better.
• The "wha-wha" sound is not bad for the mount. I basically have to choose between less backlash and less sound...
• The clicking noise when the motors turn is weird - will have to investigate that.
• But the clicking noise when I change directions is normal for the CGEM mount. The spur gears will always have a little bit of play.

---

I could not figure out where the clicking comes from. It's most pronounced when slewing really fast on the dec axis. I decided to give it a try. And my first experience is just awesome:

• It's easier to balance the scope - way easier
• My guiding results are much better. The correction that PHD needs to make are WAY smaller (and I haven't even done backlash compensation or PEC training)

Very hopeful!

My summary:

• Yes, Hypertuning is tricky and cost some (actually not a lot) $$$
• In my case it took a while and was quite a lot of work (and stress) - but it was totally worth it. Not just the end result of having a much better mount, but also the better understanding on how my mount works and how I can tune it.
• Finally, I'm somewhat disappointed how hard Celestron makesit to do any of this work with their mounts by (quite literally) glueing them together. I'd be thrilled if Celestron would sell the mount in pieces and then lets us put it together.

And finally: Ed Thomas rocks! The CD's that he ships are really good and explain a lot. But then he gave me a lot of help and guidance through the process.

First image with new setup: M44

As my very first image, I chose M44 - I figured that an open cluster is a good start object. Because I still couldn't figure out how to control the filter wheel from TheSkyX, I had to move it between series manually :-(

I took LRGB images. All images have a lot of noise - I wonder if it is because of our light pollution here. And I needed new darks, bias, flats and dark flats. I ended up with:
10 Luminescence 60 sec
5 RGB 240 sec
20 Darks
20 Flats
20 Dark Flats
20 Bias

Taking the Flats turned out to be quite the challenge. The flat panel that I use is too bright for the color filters and I had to use such a short exposure, that the vignetting did not come out. I ended up waiting for dawn and took them then. Also, they had lots and lots of artifacts:

But on further inspection (and after taking a second set of flats and processing them) it turned out that my scope (the camera and the corrector plate) are really this dirty...

I also used this as an opportunity to try out CCDStack. It's much more manual then Deep Sky Stacker, but I think it's worth the effort. I'll have a month or so before I have to decide if I want to buy it.

After a day of learning and playing, I finally had my first LRGB combined image. I did some basic processing in Photoshop and ended up with this:

Overall, I'm pretty happy with this, the star colors are good (but not too saturated). The only problem I had was that the background was very noisy. If you zoom in, you can see that the stars have a pretty strong halo, but that it then drops off. I wish I could have done this better.

This was also the first time, that I used the OAG instead of the guidescope. And I had the same drift during guiding that I had before. The good news is, that the guidescope seems to be OK. The not so good news is that there is this drift. I hope that this will improve when I hypertune my mount (everything arrived yesterday!) Unfortunately, I won't have time to do it this weekend and next weekend.

Setting up the filter wheel and the off-axis guider

I finally received my filter wheel from Starlight Xpress!!! So, I could set that up and also finally use the off-axis guider from Starlight Xpress.

First, I reconfigured the scope: removed the 4 spacers (for correct backfocus) and replaced it with the filter wheel / OAG. And now, I can use the reticle eyepiece in the guidescope. And I had to do a bunch of recabling (and need a shorter USB cable for the filter wheel).

Then, I tested the filter wheel with the test software from Starlight Xpress - works!

Next, I connected the filter wheel to TheSkyX. Assigned names to the 7 filters. But then I couldn't control the filter wheel from the camera plug-in. It just didn't move at all. I checked the Bisque forums and a couple of folks mentioned issues with the Starlight Xpress filter wheel when they used a USB hub. So, I connected the filter wheel directly to the laptop - but same result. The weird thing is, when I connect TheSkyX to the filter wheel and use the Starlight Xpress program to move the filter wheel, TheSkyX shows status "Moving". I.e. there is some connectivity to the filter wheel.

Then I wanted to figure out the difference in focus. I used the Bahtinov mask for this and subframed stars in to get a very accurate focus. The 4 Baader filter (LRGB) have the same focus, while the Ha filter (from Orion) and the OIII and SII filter (from Astronomik) have an offset of +36. I could enter the offset into the Ascom driver, but not sure if and how that would be used from TheSkyX. It would certainly be helpful if TheSkyX automatically adjusts the focus when changing filters.

As the final step, I needed to focus the lodestar. But all lodestar images have a very strong glare in TheSkyX:

I don't see that effect in PHD or the lodestar program (from Starlight Xpress). Though in TheSkyX I see WAY more stars then in the other 2 programs. I asked on the Bisque forum about that.

And throughout the entire night, I had several connection problems. At some point the cables that run down the mount (USB, power) got loose when the mount slewed into an extreme position (I extend those cables a little bit). The serial ports needed some wiggling before they wanted to connect. And I had a power outage again - I can't seem to figure out where this comes from (I already replaced the 12V adapter and also the splitter).

...

I received a couple of answers:

Apparently, the image from the OAG is normal... I can try to use one of the stars - although there isn't much of a contrast. I could also try to take flats (and maybe darks and bias) to reduce the image properly. But wouldn't I need to take new flats every time I change the OAG (move it in or out - turn it).

Interestingly: when I started autoguiding with an image like the above, I got this picture in the subframe:

It seems as if subframing helps with the vignetting. So, that might work!

With this, I tried autoguiding in TheSkyX. Now, with the OAG, the movement should be much more sensitive. In x-direction, the default of 10seconds for the relay worked. But in y-direction I had to go higher then 100 seconds for the relay. In the +y direction it worked with much smaller values. But in the -y direction I needed the high value. I wonder if this is because of backlash in the mount. And if yes, if it will improve once I hypertune my mount (maybe this weekend?)
I started guiding with TheSkyX and could finally get it to work. But I had an image shift - like I had when I tried to guide with the guidescope. But the shift was very slow, good enough to try some imaging.

With regards to the filter wheel, I was (of course) asked to update to the latest version of TheSkyX - done.
I was also told, that TheSkyX only moves the filter wheel when I am taking an image (i.e. not when I just select a different filter in the drop down).
And finally, that I need the "driver in the ASCOM category that's installed along with the application."
I tried this, and the filter wheel indeed turns before taking an image. But after a few images, it timed out with the 209 error (as described in the Bisque forum).
I could not figure out what was meant with the ASCOM driver, will have to reply to the Bisque forum.

The distorted lodestar images are because of binning??!!

Tonight, I wanted to figure out why the lodestar images are so distorted.

First, I attached the lodestar to the main scope. There the image was perfect. So, it seems as if it is the guidescope. I moved it back, and ... no elongated stars at all! Strange. So, I focused the scope as good as possible (with the small Bahtinov mask).

Well, I then tried to work on autoguiding with TheSkyX. At some point, I had the elongated stars again. It took me a while to figure out, that this only happens when I'm 4x4 binning. Here are some images:

No binning:	2x2:
3x3:	4x4:

The effect is easily noticeable. I wonder if this is a blooming effect?

Well, after that was cleared up, I tried to get autoguiding to work. But still no luck. I posted the guiding log file, the lodestar image and the autoguiding image to the Software Bisque forum - hopefully I will get some help.

And I also realized that I need to build a shield for the guidescope. Here is an image where some light reflects off the main scope:

And here is the same image with all lights turned off and some black paper on the scope:

And that was it.

...

Few days later: Yes, this is because of binning. Apparently, the lodestar only supports 2x2 binning. And if you choose a higher binning value, it only bins in one direction, i.e. 3x2 and 4x2 - that explains why everything looks so elongated. I can actually set the max binning in the ASCOM driver configuration. After I did that, TheSkyX only had 1x1 and 2x2 binning in the dropdown. Riddle solved!