Stargazing Documenting the setup of my new telescope

[Andy Resnick]

...it's the NeverEnding story .

Tell me about it...photomicroscopy is the same :)

It seems vaguely misleading that the cost of an OTA (or lens equivalent) represents at most 50% of the cost of a minimally-functioning astrophotography setup.

 

[collinsmark]

After a couple weeks of clouds and rain ("atmospheric rivers" they are being called by the media), I finally had a clear night to get a little more testing done.

One of the first things I need to test with this setup is to get the backfocus spacing more-or-less correct. The critical backfocus spacing is the distance between the field flatterer optics (or in my case the 0.7x focal reducer) and the sensor plane. It's important that the spacing be a certain amount. Too much error in either direction -- too close or too far -- and the stars and anything else will be distorted and blurred around the edges and corners of the frame.

The C14 EdgeHD optical system (including the 0.7x reducer) require a backfocus spacing of 146.05 mm. But there's a bit of wiggle room. If it's off by +/- a half centimeter or so, it's not that big of deal. I'm taking advantage of this wiggle room with the fine focuser. I adjust the telescope's main focus knob for coarse focus, which moves the primary mirror back and forth (but has no effect on backfocus spacing). Then the electronic focuser is used for fine focus, which adjusts the backfocus spacing a little bit, changing the focus as a result. As long as the electronic focuser's travel is centered rougly on or around the optimal backfocus spacing, I should be OK.

So the point of my testing last night was to get the center of the electronic focuser's focus travel in that "sweet spot." I did this by adding or removing various spacers between the electronic focuser and off-axis guider (OAG), then testing.

Figure 1. A few spacer tubes (of various lengths) were inserted between the electronic focuser (the vomit-green colored thing) and the OAG. This was not the final configuration, by the way; it was too much spacing.

Figure 2. Final configuration. I got very acceptable results with just a single spacer tube, shown here. By the way, if you're curious, that knob on the back of the telescope with the orange ring around it is the telescope's main focus knob. That's what I use for coarse focus.

After getting the spacing acceptable, I took a couple screenshots of NINA, while the telescope was pointing to NGC 2301 (Great Bird Cluster).

Figure 3. Screenshot of NGC 2301used for testing backfocus spacing. It's just a 10 second, single exposure, no calibration or processing. [Edit: "autostretch" is applied though, of course.] The vingetting is normal and the results look good. (Recall that this setup using a full-frame sensor and an effective focal length of 2770 mm.)

Figure 4. Same image as Fig.3, but with the Aberation Inspector mosiac turned on. This displays small segments at each edge and corner, and one small sement of the center, for comparison. The stars look acceptably sharp and round all the way to the edge of the frame.

Figure 4 shows very promising results. Now, before you go and say, "those stars look pretty bloated," recall that the focal length is 2770 mm. That, and the seeing was pretty poor at the beginning of last evening when I took the screenshot. And my nextdoor neighbors had their fireplace going, and their chimney is only a few meters away (we're all pretty cramped). All in all, the sharpness of this telescope and the incredibly flat field exceeded my expectations. In my opinion, Figure 4 is really, really good news to me.

So, since the sky was still clear, I decided to run the scope through it's paces for "first light." I decided on M100, for no particular reason besides it being in the right location in the sky. I'll report back with the results in a future post once I check out the data myself.

I did take one more photo of the telescopes this morning taking FLAT frames.

Figure 5. This moring, after the big telescope's "first light." Flats are being taken with artists' "light box light pad" sketch pad things. They're very inexpensive and work surprisingly well as flat panels. The bedsheet hanging in the background is to sheild the telescopes from the other neighbor's porch light, which is always on 24-7 for some godforsaken reason.

To be continued ...

 

[collinsmark]

Speaking of useful astrophotography tools, I would be amiss if I didn't mention "rubber strap wrenches."

Figure 1. Rubber strap wrenches. The tube spacers in the center are shown as an example application.

If you are ever interested in taking up astrophotography as a hobby, you'll inevitably have to deal with a bunch of tubes and rings and such that screw together. By their very nature, these tubes will bind together making them seemingly impossible to unscrew with one another.

The solution is to make sure you have a pair of these "rubber strap wrenches." They go by other names too, such as "oil filter wrenches." You won't use them everyday, but they will save you from incredible headaches when you need them.

A typical, budding astrophotographer, after experiencing their first nightmare of tube seizing, might be inclined to put some anti-binding/anti-seizing paste on the tube threads so that it never, ever happens again. But now there would be risk of accidental transfer of the paste to the optics, which would be a whole different nightmare. I advise a better solution is to keep a pair of rubber tube wrenches on hand to deal with the binding.

Keep them in your bag of essentials.

 

[collinsmark]

First Light, Acquisition and Processing, Part 1

I ended up get two clear nights in a row. And both with a new moon too! Very good for testing the optics under good conditions. As I mentioned earlier, I pointed the telescope to the "Blowdryer Galaxy," M100, and hoped for the best.

Here is some insight into my processing workflow for galaxies. Let's start with the luminance (L) data.

Figure 1. Single, 60 sec subframe before calibration frames applied. This frame was taken with the IR/UV cut luminance filter (L). The autostretch was applied here so that you can see the galaxy. (Without the autostretch, it would just appear as an almost completely black image.) There were many more like it. I also took many other subframes with the red (R), green (G), and blue (B) filters separately applied, but they are not shown here for brevity. They'll all look like black-and-white images, since I used a monochrome camera.

Figure 2. Here is the same, single, 60 sec subframe, but after DARK and FLAT frame calibration applied. The master DARK and master FLAT frames were created earlier in the workflow, but not shown here for brevity. The autostretch was applied here so that you can see the galaxy. As you can see, the artists' "light box light pad" thing does a pretty good job for producing flat frames.

Figure 3. Here is the L channel after alignment and stacking. A slight crop was applied to remove the edges. Again, autostretch applied here so you can see the image. There are noticeable gradients that were not compensated by the FLATs. Although the nonuniformity of the cheapy flat panel plays a role in these gradients, I wouldn't blame it too much. Most of the gradients are likely caused by my neighborhood's glare and light pollution (porch lights, streetlights, etc).

Figure 4. After PixInsight's brand new "Gradient Removal Tool" (GradientCorrection process). That removed the residual gradients nicely. (Autostrech applied here so you can see it.)

Figure 5. A second crop applied. (Autostretch is also applied here so you can see it.)

Now let's move on to the color data. We'll come back to the L image later in the next post.

Steps take for color data (not shown here for brevity; each color channel looks pretty similar to what's shown above for the L data -- it's basically me doing the same thing three more times, once for each color channel.)

• Dark and Flat frames applied separately to all the red (R), green (G), and blue (B) subframes.
• Each channel was aligned as stacked (separately for each R, G, and B channel; although all images used the same reference frame for alignment).
• The stacked R, G, and B, images were combined using PixInsight's ChannelCombination process to produce a single RGB image.
• A slight crop was applied (same dimensions as the L channel) to remove the edges.

Figure 6. RGB image (no L applied, just yet. That comes later.) (Autostretch applied here so you can see the image.)

Figure 7. PixInsight's new GradientCorrection process applied. I love PixInsights new gradient removal tool. It's pretty easy to use and works well. (Autostretch applied here so you can see the image.)

The next step is to correct the above image for color. Obviously, the relative color balance is not right yet. the first step in this is to use PixInsight's ImageSolver script to platesolve on the image, comparing it to a database of known stars having a known relative brightness and color. (ImageSolver process not shown here, because it doesn't affect how the image looks.)

Once the image has been "solved," giving it a known location in the sky, the "Spectrophotometric Color Calibration" can be performed. In this process you can choose the filters that were used in acquisition, and it will adjust accordingly.

Figure 8. PixInsight's SpectrophotometricColorCalibration process applied. (Autostretch applied here so that you can see the image.)

Figure 9. Now that the gradients have been removed and the colors have been calibrated, the RGB image was cropped a second time, using the same parameters/dimensions that were used for the L image. (Autostretch applied here so that you can see the image.)

Note that the above image is still pretty noisy and doesn't contain the data from the L image. That's where we'll start next time in Part 2.

To be continued ...

 

[collinsmark]

First Light, Acquisition and Processing, Part 2

Normally in this part of the galaxy workflow, I would add in the hydrogen-alpha (Hα) data into the mix. However, adding in the Hα data is quite a complicated bit of work, and quite frankly, I didn't gather enough Hα data in the limited time I had to make it interesting. I might come back to it once I've had more clear nights to acquire the data. But for this "first-light" test here, I'm skipping it, and forging ahead with just the L, R, G, and B data alone.

So the next thing I did is apply an actual histogram stretch to L and RGB images described in the last post. I then combined them using PixInsight's LRGBCombination process.

Figure 1. Luminance (L) and RGB images combined into a single LRGB image. The advantage of adding the luminance data to the color data is primarily for a natural method of noise reduction that doesn't rely on any complicated signal processing routines. Acquiring L data is really quick way to gather lots of photons, allowing the Central Limit Theorem to take care of the noise naturally, and without having to gather all the photons for the R, G, and B channels separately (which would take 3x the amount of integration time). Obviously, some color data is necessary, which is why we spent all that time on the RGB image. Acquiring luminance data gives you a lot of "bang for the buck," in terms of the precious time the telescope is pointed at a broadband target. (This only applies to broadband targets, by the way. Luminance data doesn't do much good if you're shooting narrowband targets.)

Now that all the data has been incorporated into the image, it's time to move on to the signal processing.

Figure 2. Sharpening and noise reduction routines have been implented -- a little, not too much -- using RC Astro's Blur eXterminator and Noise eXternimator plugins. Strengths of these processes have been reduced below their default values. We're going to break it up, applying a little bit here, and a little bit more toward the end of our processing.

The next step is to enhance the detail in the core of the galaxy. To do this I used Herbert Walter's "GAME" plugin (https://www.skypixels.at) to create a mask around the core of the galaxy's core. I then used PixInsight's CurvesTransformation to reduce the brightness of the mask to about 1/3rd. Applying that reduced mask to the LRGB image (protecting everything except the galaxy's core), I used PixInsight's "HDRMultiscaleTransform" process to the image, enhancing the details in the core.

Figure 3. HDR Multiscale Transform applied to the galaxy's core, enhancing the details in the core.

For the next step I need to remove the stars for a bit (don't worry, we'll put them back in a second). For that I used RC-Astro's Star eXternimator.

Figure 4. Stars temporarily removed.

At this point I created another mask, this time using PixInsight's RangeSelection mask tool, protecting all the dark areas. Only the bright areas around the galaxies can be affected after applying the mask.

Figure 5. A touch of PixInsight's "Local Histogram Equalization" applied with the RangeSelection mask applied to the image. This brings out some of the detail in the galaxy's spiral arms.

Before I put the stars back, I did a little work on them (not shown). I noticed what looked like a bit of chromatic abberation, but only on the brightest of stars. Well, that's a shame. It might have happened because I used a frame from the luminance data as my reference image for alignment. Maybe I should have used a frame from the G data. Oh, well. Live and learn. Next time, I guess.

Rather than start all over, I did a little creative masking with the stars, and reduced the color saturation of the very brightest stars. That got rid of the chromatic abberations pretty well.

While I was at it, I reduce the stars a little bit using PixInsight's MorphologicalTransformation before putting the stars back.

Figure 6. Stars are back. You can put the stars back into the image using PixInsight's PixelMath.

All that's left is some final touch-ups using CurvesTransformation, and a final, small round of sharpening and noise reduction using RC-Astro's Blur eXterminator and Noise eXterminator again (as before, not too much; just a little).

Figure 7. Final image for this "First-Light" test.

There's a lot of detail in the final image -- more than PF will allow me to display by embedding the image in a post. If you'd like to see the image in its full resolution, here is a link to it:
http://www.shadycrypt.com/PF/BlowDryer2024_FirstLight_Final.jpg

=====================

Blowdryer Galaxy (a.k.a., M100, NGC 4321, Mirror Galaxy), imaged from my back patio, March 2024. M100 can be found in the constellation Coma Berenices.

Equipment:
Celestron C14 EdgeHD telescope
SkyWatcher EQ8-R Pro mount
Celestron 0.7x Focal reducer (for C14 EdgeHD)
Off-axis guider (OAG) with guide camera
Baader LRGB filter set
ZWO ASI6200MM-Pro Main Camera

Software:
N.I.N.A.
PHD2 Guiding
PixInsight with
o RC-Astro Plugins
o SkyPixels "GAME" plugin

Acquisition/Integration:
Location: San Diego, USA
Bortle Class 7 (maybe 8 ) skies
All subframes binned 2×2
Stacked using drizzle algorithm
L: 207×60 sec = 3.45 hrs
R: 119×60 sec = 1.98 hrs
G: 107×60 sec = 1.78 hrs
B: 115×60 sec = 1.92 hrs
Total integration time: 9.13 hours

 

[Devin-M]

…a nice, modest rig…

 

[DennisN]

…a nice, modest rig…

It's just the bare-bones.

 

[collinsmark]

I've got more data!

Here's M100 (a.k.a., NGC 4321, "The Blowdryer Galaxy," "The Mirror Galaxy") with about 37 hours of data and a lot more care with post-processing. (It's the same target as before, just with more data and more care.)

Not only did I gather more L, R, G, B data, but I also gathered and integrated some Hα data giving a "punch" to the nebulous regions in the galaxy.

Figure 1. M100 with about 37 hours of data from the new telescope.

Also, here's a 100% crop of the image showing more detail.

Figure 2. 100% crop.

Equipment (dew mitigation accessories not listed, since they're not part of the optical/imaging train):
Celestron C14 EdgeHD telescope*
SkyWatcher EQ8-R Pro mount*
Celestron 0.7x Focal reducer* (for C14 EdgeHD)
Off-axis guider (OAG) with guide camera
Baader LRGB filter set*
Antila 3nm Hα filter*
ZWO ASI6200MM-Pro Main Camera

*first light

Software:
N.I.N.A.
PHD2 Guiding
PixInsight with
o RC-Astro Plugins
o SkyPixels "GAME" plugin

Acquisition/Integration:
Location: San Diego, USA
Bortle Class 7 (maybe 8 ) skies
All subframes binned 2×2
Stacked using drizzle algorithm
L: 532×60 sec = 8.67 hrs
R: 439×60 sec = 7.27 hrs
G: 426×60 sec = 7.10 hrs
B: 472×60 sec = 7.87 hrs
Hα: 16×300 sec + 30×600 sec = 6.33 hours
Total integration time: 37.43 hours

For comparison, here's an image of M100 from the Hubble Space Telescope (HST):

Figure 3. Image from HST (not an image from my telescope!) for comparison (Source: https://science.nasa.gov/mission/hu...night-sky/hubble-messier-catalog/messier-100/ ).


Here's my speculation as to why M100 is called "The Blowdryer Galaxy" (taken from my post in "Our Beautiful Universe" PF thread https://www.physicsforums.com/threads/our-beautiful-universe-photos-and-videos.800540/post-7081032)

I speculate that there was an amateur astronomer out one night observing M100, perhaps with a group of guests, and the poor sap's corrector plate fogged up from dew. Naturally, the observer scrambled inside to grab a hairdryer and extension cord to warm up the sky-facing optics. I mean, we've all done it. Everyone present thereafter started calling M100 "The Blowdryer Galaxy." This time though, the name stuck and slowly spread to others. That's just my speculation: I have no solid evidence or source as to how this object got its nickname, but that's my guess. If anybody can find a credible reference as to how this galaxy got its "Blowdryer" nickname, let me know.​

Fighting dew is a real thing. Every amateur astronomer must deal with this, whether it's for an astrophotography setup, scientific setup, or even visual observing with eyepieces. It's always a battle with dew.

When buying a first telescope, the seller usually doesn't emphasize how important dew mitigation is. But the buyer will find out the hard way once their objective fogs up after only a few hours of observing. At the very least invest in a dew shield if your telescope doesn't already have one attached (most refractors come with a dew shield built in, most Schmit-Cassegrains [SCTs] do not. Maybe it's not such a big of deal with Newtonians.) If your scope has a built in dew shield, make sure it's extended when the telescope is in use. Consider getting a dew strap or two along with a dew controller. Don't forget about the finderscope -- that can fog up too.

Of course, as a last resort, there's always the hairdryer. But if you find yourself in that situation, it's already too late.

About half the peripherals by weight on the new telescope involve dew mitigation.

Figure 4. New telescope's dew mitigation. The "Dew Heater Controller" is part of the power distribution box (Pegasus Astro Ultimate Power Box V2) that does other things too (12 V power distribution and USB hub).

 

[Vanadium 50]

That's really pretty.

It looks (and should look) quite blue. How accurate are the colors?

 

[collinsmark]

That's really pretty.

It looks (and should look) quite blue. How accurate are the colors?

That's an important question worthy of discussion. The answer isn't simple.

If I had to give a short answer it's this: the color saturation that I used in this image is higher than what would normally be used in a terrestrial photograph. And the color saturation that I used in my photo is substantially less than what was used by whoever processed the Hubble (HST) photograph (shown in the above post for comparison).

The full answer is complicated and nuanced. As a general rule-of-thumb, when it comes to making pretty pictures in astrophotography, color detail is fair-game. The person processing the photo has full creative license to choose the color mapping (i.e., which filters map to which color channel), color saturation, white balance, etc., all without sacrificing scientific merit. It's perfectly acceptable to be creative with colors when processing astrophotos.

It's also acceptable to attempt to remove optical imperfection artifacts that were caused by Earth's atmosphere or optical imperfections in the telescope. Examples are removal of halos caused by narrowband filters, gradients caused by anything that causes artificial gradients (Moon, light pollution, vingetting of optics, etc.) and diffration artifacts. Global sharpening is also acceptable. Sometimes the detail is still in the data, but just needs a little coaxing out with sharpening algorithms. That's totally fine.

Nonlinear editing should be used sparingly, but sometimes acceptable such as dodging a galaxy's core, and/or applying HDR masking techniques to bring out the detail in what would otherwise be blown highlights, so long as the detail is present in the original data (i.e., not already blown-out in the subs).

What's not acceptable (more than highly frowned upon) is to add luminosity details and structures that don't exist in the original data at all. That's not cool.

But when it comes to color detail, almost anything is fair game, so long as some level of consistency is maintained across the image. It's fine for an astrophotographer to get creative with color.

Take the Hubble (HST) image (linked to in the above post) as an example. It's extremely blue and also highly color saturated. This was done solely by the choices made by the astrophotographer processing the data. It's way more blue than what our naked eyes would see if the galaxy was bright enough. But it's all totally fine, since its just color choices which are fair game.

For my image of M100, I waffled back and forth for days, trying to get the colors to look right. I probably spent more time on this one than any other target I've ever worked on to get the color balance the way I wanted it. The image posted here was what I decided on in the end. But given more time I probably would have waffled some more.

If M100 was bright enough to see the colors clearly with the naked eye, it probably would look more washed out, and maybe less blue and more beige than what's displayed in my final result. But if I chose colors such that the whole image to look washed out and beige, the viewer would miss out on some of the subtle color variations that actually do exist in the galaxy.

 

[Vanadium 50]

I agree that there is a fine line between "getting the most out of your image" and "making it up yourself."

I guess so long as it is bluer than M22 or M87, I am OK with this level of processing.

It should be one of the bluest galaxies around. Its star formation rate is crazy high.

 

[Syandan21]

Considering the situation are the suit and mask a little bit overkill?

What if you made a fish tank glovebox container to limit the dust while you work on it?

https://www.labcompare.com/General-...e/?compare=783,16856425,784,782,785&catid=135

Very interesting! Did you study anything space related in college or is this just a hobby? Just curious.

 

[jedishrfu]

For me its a casual hobby that I play with when the weather is cooler and Orion is easily visible in the night sky ala Fall and Winter nights.

@collinsmark is a more gifted, serious and either wealthy or seriously in-debt astronomer as you see from his cool setup. Mine is a simple Orion 4" Casegranian reflector on a camera tripod. where I am the clock drive and no photos are taken.

 

[Vanadium 50]

either wealthy or seriously in-debt astronomer

With a very understanding spouse.

 

[jedishrfu]

With a very understanding spouse.

Reminds me of the late great Jackie Gleason:

To the moon, Alice!

 

[collinsmark]

Collimation screw, side project:

So I decided to give Bob's Knobs a try. Here's a summary on how that went.

Background:

Collimation is important on any Cassegrain telescope system. Collimation is the alignment between the primary mirror, secondary mirror, and final image location. Ideally, the light path should form a "column," such that the light path has cylindrical symmetry (after accounting for reflections by reflecting elements).

Sometimes adjustments are necessary. Even high quality telescopes will need adjustments from time to time.

The collimation screws on the C14 EdgeHD are adjusted via Philips head screws on the secondary mirror assembly. (See Figs 1, 2.)

Figure 1. Front of the telescope with the collimation screws hidden behind the Fastar logo thingy that partially shields the screws from dew and whatnot.

Figure 2. Fastar logo thingy twisted to expose the collimation screws. They are Phillips head screws.

Potential problem:

What I don't like about Celestron's solution is:
o You have to use a tool to adjust the collimation. This makes collimation adjustments more time consuming, since you must grab the tool for each and every small adjustement. If you ever forget to bring the tool while at a remote location, you can forget out collimation adjustements.
o Said tool might slip and fall onto the Schmidt plate. (Eek!)
o Phillips head screws means yoy have to apply pressure to the tool just to get it to work, increasing the chance that it might slip and fall onto the Schmidt plate. At least with the Meade LX200-ACF, the collimation screws were hex wrench heads. But Phillips heads?! that's just asking for trouble.

Potential solution: Bobs Knobs.

Bobs Knobs makes and sells collimation screw replacemnts that are knurled knobs that you can adjust with your fingers. I.e., no tools needed, once they're installed. I decided to give them a try.

=== Caution to anyone replacing their collimation screws ====
Btw, if you ever change your collimation screws, ensure you replace the screws one at a time! Uncrewing all collimation screws at the same time could cause the secondary mirror to fall off!
=================

Figure 3. Collimation screws are now Bobs Knobs.

Replacement went well, and I got everything re-collimated. They work as advertised. Adjsutment is easy, since they stick out quite far, allowing the user to get a grip on the knobs from any direction. But there's a problem.

The problem with the Bob's Knobs:

They stick out too far! For an illustation of this, see Figs 4, 5.

Figure 4. Bobs Knobs stick out quite far, past the lip of the optical tube assembly (OTA).

Figure 5. Demostration that they stick out past the lip. Sorry about the bad focus here, I was using my cell phone camera.

I can tell when I put the lens cap on the telescope, that it teeter-totters on the Bobs Knobs. That's not a big deal if I'm just leaving the scope outside for the night, covering it up with the TeleGizmos cover. But it is a huge deal if I ever need to tear down the setup.

Here's the thing. When I remove the OTA, it needs to rest on the ground (at least temporarily) facing down, with the lens cap touching the ground. But with the knobs sticking out, that means the full weight of the OTA will be placed on the Bob's Knobs. The Bobs's Knobs are attached via the secondary mirror assembly. And the secondary Mirror Assembly is held in place solely by the Schmidt plate. That's going to damage the Schmidt plate for sure.

Can you image the horror of carefully, very carfully, slowly lowering the OTA onto a solid surface, only to have the Schmidt plate shatter on the spot?

I mean, good god, Bob. What were you thinking here?!

I suppose it would be fine for a permanant setup that never required maintenance. But that's not for me.

If I ever have the gumption to take the scope out the dark site, or if I ever have to pack the scope up for some sort of patio maintenance, and I forget to switch out the collimation screws, that's a guaranteed disaster.

Conclusion:

I will not be keeping this Bob's Knobs configuration (sorry, Bob). While Bob's Knobs might work fine for some people on some scopes, it's not for me on my telescope.

In the mean time, I'm going to attempt engineering up an alternate solution with McMaster-Carr. If that doesn't work out, I'll go back the Celestron's factory default screws.

 

[Tom.G]

How about machining the heads to be thinner? Also, I can't tell for sure but, it looks like there is a shoulder under the heads that could also be machined off.

"Machining" can be nothing more than some tedious work with a file.

Worst case, if there is a shoulder to remove, you may need to shorten the threaded portion to avoid interference somewhere.

Please let us know what your solution is.

Cheers,
Tom

 

[collinsmark]

How about machining the heads to be thinner? Also, I can't tell for sure but, it looks like there is a shoulder under the heads that could also be machined off.

"Machining" can be nothing more than some tedious work with a file.

Worst case, if there is a shoulder to remove, you may need to shorten the threaded portion to avoid interference somewhere.

Right. I've given that idea some thought, but I'm not sure I trust my skill with such tools.

Yes, the "too tall" culprit involves some plastic, cylindrical spacers/shoulders. I suppose they could be sawed down to half their length, and also machining the ends of the screws to compensate. Then, if the large, knurled knobs start interfering with the twisting Fastar logo cover thingy, I would have to machine those too, making the diameter smaller. But again, I'm not sure I'm up to the that task. (I'm also concerned about corrosion; If any of these screws easily corrodes, I'll be in a world of hurt.)

I went a different route and ordered some parts from McMaster-Carr. I'm hoping those should work better (fingers crossed).

I'll try keep you posted with pictures and a parts list once everything's ready to swap.

 

[Vanadium 50]

Have you talked with Bob? He may be interested in improving his product. In another area I have dealt with some highly specialized cottage industries and they have universally welcomed feedback to help improve their product.

 

[Vanadium 50]

PS When can we see a picture of Phobos?

 

[collinsmark]

Collimation screw, side project (part II):

After a couple weeks of overcast nights, I finally was able to replace the collimation screws with the new solution and re-collimate.

As a refresher, the reason I'm doing this is because the Bob's Knobs screws stick out too far, risking certain damage to the Schmidt plate if I ever remove the optical tube assembly (OTA) from the mount without swapping the collimation screws back to the original ones (see Fig. 1).

Figure 1. Bob's Knobs collimation screws stick out too far.

From anybody interested, here's the specs for Celestron's collimation screws (specifically for EdgeHD scopes, shown here):

Table 1. Collimation screw specs for EdgeHD optical tubes:

(Source: https://www.celestron.com/blogs/kno...on-current-production-celestron-optical-tubes)

I've got the C14 Edge HD, so that means I need M3×0.5, with 18 mm length. I found the following parts from McMaster-Carr:

• Plastic-Head Thumb Screws, Knurled, M3 x 0.5 mm Thread, 20 mm Long (Head color: Black)
• Delrin® Acetal Plastic Washer, Water- and Steam-Resistant, M3 Screw Size, 3.2 mm ID, 8 mm OD (Thickness 0.7-0.9 mm)

Figure 2. My McMaster-Carr order.

The washers are meant to reduce the shaft length from 20 mm to something closer to 18 mm. I figured two or three washers per screw should do the trick (I went with 2 in the end). I bought plenty of washers to be on the safe side.

While it's not clear from the image and description, the screw threads are chemical-resistant 18-8 stainless steel. (More on that later; boy, I hope that means they're also corrosion resistant.)

As a gentle reminder, when replacing your collimation screws, always replace them one-at-a-time and not all at once (else your secondary mirror might fall off). See Figure 3.

Figure 3. First of three screws replaced. Top two screws are the Bob's Knobs, and the bottom screw is my solution.

Now that I had one of the Bob's Knobs screws out for comparison, I photographed it on a random book alongside one of the original Phillips-head screws, and one of the screws from my new solution. See Figure 4.

Figure 4. Collimation screw comparison. Screws were placed on some random book along with calipers for scale. Top screw is my new solution. Middle screw is one of the original Phillips-head screws. Bottom screw is one of Bob's Knobs.

Below (Figs. 5-7) are images after their full installation.

Figures 5-7. New screw solution installed.

Re-collimation went well. The new solution is almost as easy as it was with Bob's Knobs. But now there's no increased risk of Schmidt plate destruction/obliteration if I remove the OTA from the mount.

The only thing I worry about is corrosion. The original Phillips-head screws and the Bob's Knobs had threads of black oxide (or maybe black anodizing?). Where the screws in my solution are chemically resistant stainless steel.

Which is more resistant to corrosion? This isn't my area of expertise. Is switching to stainless an upgrade or a downgrade in this regard?

I've tried to look online for answers, and from what I can gather, I might be OK. But I haven't been able to find what I consider a definitive answer, and corrosion resistance isn't my field of expertise.

Besides the corrosion questions though, things look quite promising.

 

[Vanadium 50]

Which is more resistant to corrosion?

You'll find out.

Stainless is not one thing. It's a whole family of alloys. Some are better in this regard than others.

 

[collinsmark]

You'll find out.

Stainless is not one thing. It's a whole family of alloys. Some are better in this regard than others.

Right. Specifically, the new screws are 18-8 stainless steel. Most of what I find online is 18-8 corrosion resistance is "excellent." So I think I should be OK. I hope.

 

[collinsmark]

Oh, and I did check the original Philips-head screws against a magnet, and they were attracted to the magnet (indicating the original screws were steel, and not aluminum). So with that, I'm hoping I won't have any dissimilar metal problems with the new screws, beyond what would be with the original. I hope.

 

[russ_watters]

Figure 5. Demostration that they stick out past the lip. Sorry about the bad focus here, I was using my cell phone camera.

I can tell when I put the lens cap on the telescope, that it teeter-totters on the Bobs Knobs. That's not a big deal if I'm just leaving the scope outside for the night, covering it up with the TeleGizmos cover. But it is a huge deal if I ever need to tear down the setup.

Conclusion:
I will not be keeping this Bob's Knobs configuration (sorry, Bob). While Bob's Knobs might work fine for some people on some scopes, it's not for me on my telescope.

At this point I can't remember if my C11 came with knobs or if I retrofitted it, but either way I agree they are a huge benefit for collimation. For me it's about fumbling around in the dark to find the knob (instead of fumbling around with a screwdriver). Potential solution: add a stand-off insert to the lens cap by epoxying --anything to it. Pieces of wood, nuts, whatever. You only need a few mm.