Making a CDK astrograph

[Stefan Buda]

Since I finished the Honders, about a month ago, I've been working on a new project - one that is a bit more down to earth than the BRH.
It is going to be a 250mm aperture, f/5.8 CDK. I want to show the people from OO UK how it is done properly.
If there is interest here, I will do a photo blog (should it be called a phlog?) like I did with the Honders?

[Paul Haese]

Ohh yeah, bring it on. Looking forward to it already Stefan.

Are you going with a carbon tube or truss system?

[Atmos]

That would be fascinating to have another phlog ;)

[croweater]

Yes please Stefan :)

[Lognic04]

Awesome, looking forward to seeing the making of a new telescope! :D
If you don't mind me asking, how did you learn to make telescopes, like the mirrors and correctors/lenses?
:)

[RickS]

Big thumbs up from me too :thumbsup:

[multiweb]

+1. Bring it on. :thumbsup:

[Stefan Buda]

Ok, plenty of interest, it seems.

First a bit about the "why", before going on with the "how".
As decent size astrographs go, that have enough focal length to resolve smaller targets, there are not many practical choices, especially if the instrument needs to be portable. Three main types come to mind: The large Newt with Wynne corrector; The RC with field flattener, and the CDK. Of course there are other, more exotic, designs that I did not consider to be really practical, for various reasons.
Although the old Newtonian with a good corrector is theoretically a wonderful astrograph, in practice there is no nice and proper way of getting the focal plane perpendicular to the axis of the focuser. And that kills it for me, and besides, it is such a 20th century thing - like a DSLR.:lol:
The RC with well made optics and mechanics is also excellent, but that is where the problems are buried, because making those highly aspheric surfaces and keeping them aligned is not easy. To make those difficulties manageable, manufacturers keep the f/ratio rather slow.
Then there is the CDK with easy to make secondary and not too highly aspheric primary that can be made very compact and easy to collimate.
So, for me the CDK is the winner.

Some of the specs I'm aiming for:

-Carbon sandwich tube about 500mm long and 300mm OD.
-Should be twice as fast as an RC.
-Only secondary mirror collimation.
-Secondary mirror focusing, no focuser proper.
-Well corrected field of 44mm diameter.
-ASCOM interface for controlling focus, fans and heaters.
-Minimalistic appearance-no "Bob's knobs" anywhere.
-Should eat larger scopes for breakfast.

The mandrel for making the tube is coming along nicely. About 290 hours of 3D printing went into it so far and another 42 to go.

[Stefan Buda]

Logan,

In the age of the "Google University" there is no excuse for not knowing something.

[Troy]

Stefan please don't skip any details on the tube, I'm watching with interest. :eyepop:

[sharpiel]

I'd like you to make me one too please...

[Lognic04]

True :)

[bojan]

Definitelly interested in phlog :)

[rally]

Hi Stefan

Great project, how are you making the corrector lens assembly ?

Rally

[Stefan Buda]

I am keeping tabs on labour and material costs and I will be able to tell you how much you would need to fork out.



Same as I did for the Honders: I cut up a few slabs of optical glass and turn the pieces into lenses using normal lens making techniques - see some pics on my earlier phlog: http://www.iceinspace.com.au/forum/showthread.php?t=151040&page=2

In fact I tried to cut some corners by producing an optical design that used an off the shelf Edmunds PCX lens, but I abandoned the idea because the Edmunds lens is available only for 50mm diameter, which is not enough.
They have a larger version of that particular lens, but it comes only with single layer AR coatings, and that would not satisfy the last point of the above listed requirements.

[Joshua Bunn]

Awesome, watching with interest Stefan.

Cheers... Josh

[Stefan Buda]

Paul,
I think for a ten incher a carbon tube makes more sense than a carbon truss.
Probably from 12" up is a different matter.

[gregbradley]

That's a great project.

I look forward to your results.

One point I would make about CDK designs having used a 17 inch now for several years, it would be good to minimise the vignetting. I believe there are optical compromises between vignetting and F ratio etc but I think less vignetting would be a good thing. It would make it less dependent on good flats.

Greg.

[Stefan Buda]

Hi Greg,
Thanks for the observations regarding vignetting. If you have any other recommendations, please don't hold back.
My design, as it stands now, has no vignetting up to an APS-C sensor size, and less than 20% at the corner of a full size chip. That is supposed to be quite good.

[Lognic04]

Would this type of scope(and this one in particular) be comparable to a Newtonian in image quality, ease of use and cost?

[Kunama]

Watching this thread with much interest !!!!

[Stefan Buda]

The image quality is comparable to a Newtonian with a Wynne type corrector and it should be a winner on the ease of use, but lose on the cost.

[Lognic04]

Cool! You say you will be able to motorise the secondary movement, how about collimation even! :D

[rally]

Stefan,

I asked because the only CDK corrector lens assembly I have ever seen in real life was off a Planewave 20" and it was an extremely deep meniscus doublet that I believe was aspheric - so this seemed to me to be a horribly complex lens assembly that even the experts were getting wrong !
My understanding was that ultimately they changed to a European optician to get it right.

[Stefan Buda]

As far as I know Planewave use a 3 element corrector made of 2 different types of glass, but my guess is that the only reason for that extra complication is to achieve the large field needed for 36x36 sensors.

[Joshua Bunn]

Hi Stefan,

I know that in my planewave CDK 12.5, there are 2 elements, but I can't speak for the larger scopes.

Josh :)

[Stefan Buda]

Hi Josh,

That's interesting, are you sure it is not 2 groups of 3 elements?

I have not tried, but it may be possible to have 2 elements at f/7 and achieve the extra large field. Or maybe they did go aspheric with one of the lens surfaces, as rally suggested earlier in this thread.

[Joshua Bunn]

I am sure :) I have taken them out several times to clean, only 2 elements convex faces towards each other.

[Stefan Buda]

Thanks for that - it is good to know for sure.:)

Meanwhile all the pieces for the mandrel have come together.
After searching far and wide, this morning I purchased some polyester film from Eckersleys, to be used as a barrier layer on the mandrel.

[Joshua Bunn]

No worries Stefan. This page also says it's 2 elements.
http://planewave.com/products-page/telescopes/12-5-inch-cdk-optical-tube-assembly/#.WshqqIARU0M

[gregbradley]

Yes that sounds good. The CDK design is great but the corrector lens perhaps tends to make a bit of bright central area that puts a lot of pressure on getting the flats just right.

Otherwise my CDK is great. The vignetting is still there even at a dark site. Just less so.

Fans blowing across the mirror as well as from the back may help.

The secondary shroud should be baffled. In the case of CDK their shroud is thicker at the base than it is along the outer edge by about 5mm.

The inner tube should be well baffled as should the corrector. Originally my CDK was very prone to flare from bright stars just outside the field of view. The baffling largely fixed that and now its quite good. But my AP Honders has amazing baffling and I have never seen a flare with that. It also seems to help it cut through light pollution so I think baffling is critical in modern scopes. The current set of baffles were made with 3D printing.

Carbon fibre struts with those nice aluminium knuckles ensures rigidity. My CDK holds focus pretty much night after night. I would not trust an aluminium tube, carbon fibre is the go.

A strong focuser is a must as these large format cameras, filter wheel, focusers etc are super heavy.

My CDK has a reducer which I have used at times. The problem with it is lack of backfocus. So only a selfguiding camera would really work with it otherwise its back to a guide scope and the potential problems that makes. Its a shame as the corrector works really well and can really pick up some photons!

So more backfocus at least with the reducer. Without the reducer the amount of backfocus seems fine.

One of the advantages apart from small spot size out to the edges is the fact a CDK is quite easy to collimate being a simple primary mirror.

Another point of engineering importance is the primary mirror mounting. The owner of CDK told me its one thing to make a mirror to diffraction limit on a bench and another to have one to that standard mounted. So the mounting is a critical aspect and the mirror should be supported in multiple places without pinching or stressing.

Of course the mirrors should be ground to as high a precision as possible and a low expansion type glass or thin type mirror like some make now would be desirable.

Greg.

[Stefan Buda]

Thanks Joshua,

I studied the site and they talk about a pair of lenses which is a bit fuzzy, but I see that their design is f/8, not f/7 as I was thinking, so quite doable with 2 lenses, and they don't even need any aspheric, I'm sure.

But I did not like their misrepresentation of the facts when they are comparing the simulated spot sizes of their CDK with an RC telescope of the same parameters. The CDK is an astrograph while the RC is a telescope.
The RC becomes an astrograph only with the addition of a field corrector.

[Stefan Buda]

Thanks Greg,

I have made two D-K scopes from scratch, and used them successfully (10" and 16"). Also I completely reground the mirrors of an ODK10. That gives me some confidence that I will get it right.

Interesting to hear from the owner of CDK about the importance of making the primary mirror happy in its mounting, because you say that there are fans blowing across the mirror. I made that mistake too with my 16" D-K and it took two years of imaging until I discovered that those fans were making my primary go astigmatic due to uneven cooling. The effect would probably be negligible for a 12" f/8 astrograph, but not acceptable for a 16" planetary scope.

[gregbradley]

Wow, I did not realise that fans could have that effect. I only mentioned the fans blowing across the mirrors as the latest CDKs have that (4 of them) plus the 3 or 4 at the back.

Greg.

[Atmos]

I know a good number of planetary imagers that use a newtonian have boundary layer fans for the primary. Would you expect the same issue?

[Stefan Buda]

In its original configuration, my 16" planetary had 7 fans. Three of them on the back plate in sucking mode, an the other four were removing the boundary layer. I tried the cross blowing ones in various combinations of sucking and blowing but in the end I had to give up and remove them.
The astigmatism problem could only be revealed by the Roddier test.
With Zerodur primary it should not be a problem even for a large planetary scope. The attached image shows the boundary fans still in place.



Yes, I would definitely expect the same problem when the ambient temperature is dropping. However not blowing away the boundary layer may result in poorer images. A planetary scope needs to be designed very carefully regarding thermal issues.

[Lognic04]

:eyepop: That is a MONSTER scope! :D

[multiweb]

That's very interesting. I would never have thought that cooling fans would cause astigmatism by blowing air locally on a glass surface. I guess it makes sense, the bigger the volume of glass the more amplitude in change.
So the challenge is to mount a large chunk of glass so it keeps its position mechanically but still has room to expand and breathe. How do you do that? Cement it with a compound that has the same or very close properties as the glass you're using. Like a crumple zone?

[Stefan Buda]

In this case, with the cross blowing fans, it is simply a matter of uneven cooling that sets up thermal gradients within the mirror, and that leads to asymmetrical figure.
Fans behind the mirror are not as problematic because there is enough turbulence within the small space to even out the temperatures in a symmetrical fashion.

[Rod]

Hi Stefan

Really looking forward to following this project.

I liked the photo of your 16 and I have seen the 10 inch DK at Heathcote. Any chance you could one day start a thread describing these instruments?

Rod

[Stefan Buda]

Hi Rod,

The 10" was made in the late 90's and I don't have much documentation on it, while the 16" is reasonably well documented, I doubt I could find the time for a dedicated thread for it. However I'm happy to dig up photos on request dealing with various details. When I was making it, I thought that it was going to be the first Peltier cooled planetary telescope, but I found out later that Anthony Wesley was working on his own 14" version about the same time.

To get back to the present: I made a first iteration mock-up of the spider/front ring.
I'm going on uncharted territory here, like I did with the Peltier cooled primary, so I have to proceed carefully. The idea is to make this important part of the OTA through 3D printing.
This mock-up study was printed with ABS filament, but the real thing will be printed with carbon fibre reinforced polyester filament, which has a modulus of elasticity about 3 times higher than ABS. Even that may not be enough so I intend to embed some uniaxial carbon fibre as well during the bonding of the parts. Also very thin ribbon cables will be enclosed in the vanes for power and signals. The finished vanes will be about 2mm thick.

[Lognic04]

Wow, very cool that you are 3D printing the spider! Not too long until people start 3d printing whole telescope tubes! :D

[Stefan Buda]

It's been done, just google "3D printed telescope", and you will find a few examples. Except those are not very useful telescopes.
I intend to use 3D printing only in ways that will not compromise performance.

I have fished assembling the mandrel, this morning, and I'm very envious of Elon Musk because he made a bigger one.:eyepop:

[Lognic04]

:rofl::rofl::rofl:

[LewisM]

The French have 3D printed entire houses so entirely possible

[Stefan Buda]

Let's get on with the phlog.
I rigged up a stand for the laminating procedure and the inner skin went on very nicely. The resin gave me about 25 minutes to complete the job, and that turned out to be just enough.
I want to sand the surface a bit before I laminate the Coremat, but I have to do that outside and the weather has turned bad.

[Troy]

Thanks for the info :thumbsup:


From the picture looks like you wrapped the first layer around mandrel and then applied the resin after that. When you completed wrapping the carbon fibre did you use a vacuum bag for it to set?

[Stefan Buda]

It's not very clear from the picture, but first I applied a generous coating of resin onto the mandrel and then I wrapped on the two layers, feeding the cloth under a slight tension. Only after that I brushed on the remaining resin.
The mandrel is very smooth and cylindrical and the cloth conformed very well, without trying to produce bubbles. I wanted to avoid vacuum bagging because the 3D printed skin of the mandrel would distort under the pressure. However, after the resin has gelled, I did wrap the ends of the tube and the seam with 50mm wide masking tape, just to make sure that the carbon conformed well to the mandrel. After about 4hrs I removed the masking tape.

[Stefan Buda]

The weather cleared yesterday afternoon and I was able to do the sanding down of the tube.
After I brought it back inside, I laminated the 3mm Coremat, using about one third of the amount of resin that it should be able to absorb. I painted the resin onto the carbon fibre only, as I wanted the outside of the coremat to remain dry for wrapping. I first wrapped the coremat with baking paper, as a barrier layer, and then with masking tape. After about 5 hours I removed all the wrapping. Today I will saturate the coremat with resin and wait for the next weather clearing.

BTW does anyone have any good internet links pointing to carbon telescope tube making? People seem to be a bit shy about revealing their techniques, or at least I didn't have much luck finding good information.

[Troy]

Thanks Stefan,
yes that's cleared it up. I also thought resin had to go in-between the layers, but that's cool I'm learning all the time.
What brand of resin are you using?

I agree on the little information on the net about making tubes of carbon fibre I searched about three months ago and it was hard to come by.

Cheers

[multiweb]

You could PM David Fitz-Henr. He's a member here and built his own newtonian (http://www.iceinspace.com.au/forum/showthread.php?t=64882&highlight=tube+newtonian).

[Stefan Buda]

The carbon cloth is very good in soaking up the resin, but you would not want to skip more than one layer without applying it.
I use R180 resin with standard hardener. Cost is about $50 per 1200g from Allnex.


Thanks Mark, I had a look, but David bought the tube for his Newtonian.

[Stefan Buda]

Some more pics:
The first one shows the tube after laminating the outer skin with extra resin.
The second one, after it has been sanded back and recoated with more resin.
The third shows it after a second sanding, while I started cutting it to length on the lathe.
All that's left to do is, the 2k polyurethane, UV resistant clear coating.

[multiweb]

How did you machine CF? I remember going through a lot of saw blades shortening my C11 CF tube.

[Stefan Buda]

I used a high speed steel tool and by the time I cut trough the inner carbon layer, it was quite blunt. After re-sharpening it, I took another very small cut to produce a nice clean and square end. So I had to sharpen the tool twice for each end.

[Stefan Buda]

Tube finished.

[Joshua Bunn]

slick ass... whats the wall thickness?

[Kunama]

Beautiful work Stefan, it seems you have a lot of spare time ;) so how about making some UTA rings for me, say 503mm ID, 600mm OD and 12mm thick ....
3k twill should do..... :D:D

[Stefan Buda]

4.7mm



Thanks Matt,
Yes I do have a lot of spare time at the moment. The laser equipment manufacturing industry has just about disappeared in Australia, so work has been very sporadic for me in the last few years. I should probably switch to astro gear full time.
However with your UTA rings I can't see a cost effective way of making them, partly because of the big wastage of carbon cloth that needs to be cut and effectively destroyed in the process. If you can think of a good way of doing it, I could have a go at it.

[Stefan Buda]

I ordered some Borofloat blanks from Germany and looks like I won't be getting them until July. I wanted to make a start on the primary before then, so I dug up an old school Pyrex blank from my collection and spent several days machining it down to 25mm thickness. Actually I overshot the mark by half a millimetre.
I just have to generate the front radius and I can start pushing glass. I already have a thick Pyrex tool that I made for re-grinding an ODK10 primary.
The radius difference is very small and I will fine tune it with a metal ring tool.

[Lognic04]

Cool Stefan, Guessing it should be easier to grind than my 10 inch? :P

[Stefan Buda]

My curve will be a lot deeper than yours (f/3), but I don't mind if we have a race, seeing that we both are at the same stage.;)

[Lognic04]

It's on! ;) :lol::lol:

[Lognic04]

Any news Stefan? :)

[Stefan Buda]

Yes, Logan, I have bad news for you: "Old age and treachery will always beat youth and exuberance."

My curve is done, but you may be able to catch up with the following stages, as I'm working on many fronts.

[Lognic04]

:eyepop: I havent even STARTED hogging, still working on the back!

ok, you win! :P

[Stefan Buda]

I 3D printed the secondary baffle using expensive ($90 for 750g) carbon fibre loaded filament. This stuff produces such a nice matt finish that I'm not sure whether it can be improved with black paint. And it weighs 26g.

[Stefan Buda]

Finished the mounting rails.
I opted for system 20 (2080) extrusions, rather than a solid bar - as I see on many commercial instruments. These extrusions have superior stiffness to weight ratio, compared with solid bars. The end caps are 3D printed ABS.
Another advantage is that the dovetail plate can be moved along the long rail for optimum CG, and thus it does not have to be extra long (and heavy).

[multiweb]

I had no idea you could 3D print CF filaments. So how does it work and holds together? Heat? Glue?

[Stefan Buda]

CF filaments contain only up to 20% short strands of CF and print like any other filament, although slower. The result is nowhere near as strong as laminated CF. I'm waiting for graphene reinforced filaments to hit the market, but that may not happen for a while.

[Stefan Buda]

Making the necessary jigs and fixtures is more work than making the actual parts.
For the primary, I made a new wedge error tester, from an old tooling plate, that allows me to also measure the concentricity of the inner diameter.
For making the dovetail plate, I had to make a fixture for cutting the 60 degree angle of the dovetail.
Next I have to figure out a way of setting up the primary on the lathe to cylindrically grind the inner diameter to the required specs.

[LewisM]

Jaw dropping work!

[Lognic04]

Cool!

[Stefan Buda]

Thanks LewisM, and I'm so glad that the five pointed star is not red. I grew up on the other side of the Iron Curtain.:lol:

[Stefan Buda]

Just need some optics now.

[multiweb]

That looks the par already. Superb. :thumbsup:
Is that a zero magnification scope at this stage?

[Stefan Buda]

:rofl:
It hasn't occurred to me, but you are right (more exactly 1x). Maybe I should call it finished.

[multiweb]

Sounds about right. It has a massive corrected imaging circle and no vignetting. What more do you want? :)

[Lognic04]

Awesome Stefan!!! Wish i knew how to make stuff like that!

[gregbradley]

Wow, this is intriguing stuff. I think we are all somewhat in awe here. You have some super useful skills.

Perhaps a custom making scope business could eventuate here. Are we looking at an Aussie Roland Christen here?

Greg.

[Lognic04]

:lol: I hope so!

[Stefan Buda]

Logan, when I was your age I also wished to be able to make stuff like that.


Thanks Greg, but you are not looking at an Aussie Roland Christen here, although a small production run of Aussie Made astrographs is a possibility if there is demand. We'll see how the numbers add up after the first one is finished.

[gregbradley]

He wasn't always a pro, he started off amateur. Apparently he agonised over the decision to go Pro as you can imagine.

Greg.

[Stefan Buda]

I see, yes, I have been agonising about that from time to time over the years.

[Stefan Buda]

One more piece of the jigsaw puzzle done: The camera rotator.
I opted for M68x1 interface thread.

[Stefan Buda]

Cylindrical grinding of the primary mirror inner diameter went well.
I cemented a hefty steel ring to the flat side of the primary, and that allowed me to set it up very accurately in a four jaw chuck.

Today I finished grit 240 (hand grinding), and I'll keep an eye on Logan so he won't win the race.

[Lognic04]

Nice!

Hehehe :P

[Stefan Buda]

I finished grinding the primary and I'll work on other fronts now to give Logan a chance to catch up.:)

Meanwhile I also made a silicon rubber mould for making the numerous polishing laps needed for the corrector lenses. Small polishing laps are a pain to make without such an aide.

I used plain hardware store silicon rubber that I pumped into the improvised Teflon mould. It would have taken months for the silicone to set all the way through, but I did not have the patience and opened the mould after two weeks. As a result the underside is a bit rough, but the top side squares are smooth and well formed. I haven't used it yet, as the lenses are still at the rough grinding stage.

[Lognic04]

What?! NOOOOO! :P The ASI has been a bit of a distraction :P

Fully finished as in onto polishing or done?

Thats a cool idea to make laps! I assume you use pitch/acculap?

Ive still got time to catch up...

:thumbsup:

[Phil Hart]

hi Stefan

Thanks for looping me into this thread - I've had a good read.

:eyepop:

This is some crazy talent and amazing patience and perseverance you have for instrument making. I am in awe but happy to live vicariously through your efforts.. I'll stick to climbing mountains and trucking eclipse gear round the world ;).

Will be following along your journey now..

Phil

[Stefan Buda]

Thanks Phil, but I can say the same thing about you if I change "instrument making" to astrophotography. :lol:

[Stefan Buda]

The third, and hopefully last, iteration of the spider is coming along very nicely. This time it is printed with the carbon fibre reinforced filament.

It looks like the thickness of the vanes will be no more than 1.5mm - better than my earlier estimate of 2mm. If rigidity turns out to be a problem, I will be able to add extra layers of carbon tape and still not exceed 2mm.

The second attached picture shows the milling of a 16mm wide and 0.25mm deep channel in one of the vanes, needed for the FFC ribbon cable.
The milling took two minutes, but the precision fixture required two hours.

[multiweb]

Wow that's amazingly thin. You reckon it's going to be stiff enough?
Second pic got me confused. Which bit in the first pic is it? Looks like an open profile.

[Stefan Buda]

Mark,
I think the shadows in the first picture confused you. There are 8 distinct pieces there and the second picture shows one of the inner sectors being machined.
Regarding stiffness, remember that people make spiders out of piano wires too and you would not describe a piano wire as being stiff. More pictures will make it all clear, but now I have to wait for the FFC ribbon to arrive before I can start bonding tings together. Also there will be another four 3D printed sectors added to the eight shown in the first picture.

[Stefan Buda]

The ribbon cables have arrived and I completed the first bonding stage of the spider. There will be several more stages, but the attached pictures should give an indication of where this is heading.

[Troy]

Is the ribbon cable for dew removal on the secondary?

[Stefan Buda]

There are 15 lines in the ribbon, rated at 500mA, and I will use 8 for driving the stepper motor. The heater will need no more than 400mA, so I will have enough lines for the temperature sensor and limit/homing switch as well.

[Kunama]

Amazing stuff Stefan :eyepop: , makes my DIY stuff seem rather agricultural :sadeyes:

(Thanks very much once again for the DXF, they are getting cut by CALM Aluminium in Minto at the moment. Going with the bolt option)

[FlashDrive]

Beautiful work ...:)

[Stefan Buda]

Thanks Matt and Col for the encouragement.

Matt, I'm not sure about agricultural, you seem to do a good job. The approach one takes to ATM is largely determined by the available tools, machinery and metrology.


Meanwhile my progress with the CDK has slowed down a bit - one needs to earn a living too.


I finished 3D printing all the components of the spider and front ring. Next I need to make a large precision jig for bonding the spider to the focuser in such a way that everything is perfectly round and concentric, and the vanes are slightly tensioned when the whole assembly gets bonded to the CF tube.

I purchased a bunch of temperature sensors and had to make the housing myself because the commercially available housed sensors are about 50mm long. Mine are 20.

[Lognic04]

Awesome!!! Maybe i need to get a custom secondary spider/holder from you for my 10! ;)

[Stefan Buda]

Logan, there are easier ways for making a secondary spider.
A secondary spider for a Newtonian needs to be sturdier because there is a larger offset for the centre of gravity that needs to be supported.


Meanwhile I 3D printed the focuser cover. After a bit of sanding and a couple of coats of flat black I think it looks acceptable.

[Atmos]

Branded CDK now :P

[Kunama]

I keep pressing the LIKE button but nothing is happening :shrug:

[Stefan Buda]

This is my registered business name since the year 2000, but I never used it on astro gear until now.


Let's hope IIS doesn't turn into a Facebook hybrid. :lol:

[Lognic04]

Nice!!! By spider/holder i meant a secondary holder like you made :)

[Lognic04]

When do you start polishing? :D

[Stefan Buda]

Probably in September when the thermal conditions in my office, where the polishing machine is, will be more suitable. Also the interferometric testing must be done in my shed because it is the only place that I have with a concrete floor, and it would be very difficult now with the low outside temperature - would have to wait forever for the mirror temperature to reach equilibrium every time I measured it.

[Stefan Buda]

The heater element for the secondary is done. 3D printing a soldering fixture made the job nice and easy, and accurate too.
It will deliver 3W from a 12V source. I will try to avoid using PWM. My 100mm Honders guide scope uses a 3W resistor string and recently I got through a night of extreme humidity without any dew on the front lens.

[Atmos]

Resister chain looms pretty cool :thumbsup:
Almost artistic!

[Kunama]

Your attention to detail continues to amaze :eyepop:

[LewisM]

Stefan does SPECTACULAR work.

[Stefan Buda]

Thanks guys for the praise - I always try my best.

So the spider assembly, after a couple more bonding sessions and some cosmetic work, looks like on the attached picture.

The bonding fixture for joining it to the focuser hub is also finished. This will ensure that everything is concentric and symmetrical. This same fixture will also be used for machining the mounting pads that will hold the back plate.

[Stefan Buda]

..and here's the result: Probably the first ever 3D printed astrograph spider.
The four shiny dots near the rim are neodymium magnets for holding the dew shield.

[Lognic04]

AWESOME!!! :eyepop:

[Kunama]

Awesome indeed :thumbsup:

[Stefan Buda]

Beautiful Borofloat blanks have arrived from Germany.

[Lognic04]

Haven't you already done the primary?

[Stefan Buda]

http://www.iceinspace.com.au/forum/showthread.php?t=165877&page=3

[Lognic04]

Ah :)

[Stefan Buda]

The mounting pads that will hold the back plate are done. Because the back plate will not be adjustable, like on the ODK10, I can use 4 mounting points, for extra rigidity.

[Stefan Buda]

The front end is nearly complete. Apart from the secondary mirror, the only thing missing is the PCB that will collect all the wiring. With the help of my son, a professional quality Gerber file has been produced and will be sent for manufacturing.

[Atmos]

This is all coming together beautifully :)

[Lognic04]

Awesome!
How are the optics going?

[Stefan Buda]

I'm fine grinding the corrector lenses at the moment.

The back plate took two setups on the mill and two on the lathe, with about 14 hours work, but I'm very happy with the precision I achieved.

I won't be providing any more updates until early September, as I'm about to go away to wormer places until the end of August - missing half of the great Mars opposition.:(

[LewisM]

I just cannot believe how perfect this is working out! No, I take that back - after receiving some custom bits from Stefan, I CAN believe it.

If only I could afford a bespoke scope like this without my wife surgically removing what I cherish so much :lol:

[sharpiel]

Your Tak? :eyepop:

[multiweb]

This is nearing completion now. Will you flock the inner tube?

[Stefan Buda]

If I turn this thing into a commercial product, perhaps I should offer, as an optional extra, some form of protection gear for men in your predicament. :lol:



Yes, I intend to use self adhesive synthetic velvet, if I can obtain some, as my old supply is running low.

[LewisM]

Touche Les :thumbsup:

[Joshua Bunn]

Stefan, such a work of art. Beautiful!

[Stefan Buda]

Time to revive this thread. I've been back for a full month now, but so much work has piled up while I was away, that I haven't been able to make much progress until now.


The only progress worth mentioning is that I managed to keep one of my small polishing machines going in the evenings and I finished polishing a small batch of corrector lenses. Last year when I tried to get a couple of lenses BBAR coated, the cost turned out to be prohibitive, so this time I'm trying to bring the unit cost down by getting more lenses coated.
The small silicone mould for casting polishing laps worked very well.

The day before I left for my trip, my son sent off the GERBER file of the small PCB, that will live inside the secondary hub, to China for manufacturing. About ten days later my son emailed me that the finished PCBs have arrived. I was stunned.

[Lognic04]

Nice!
Can you show us your grinding/polishing machines? :)

[Stefan Buda]

Not very spectacular but very important progress over the last week:
I dug up the old test plate for the 16" Dall-Kirkham secondary mirror, with the associated tools, and reground it into a test plate for the CDK250 secondary.
I had to change the RC from 950 to 1110, which didn't take long. After the manual grinding I did about 4hrs of machine polishing followed by manual figuring. When the Ronchi lines were looking quite good, I set up the Bath interferometer to find a Strehl of about .8 Further figuring/testing got the Strehl ratio to .939 at which point I called it done.

Also I sent off a bunch of corrector lenses to Longman Optical to have them broad band AR coated. Fingers crossed.

The second picture is for Logan.

[Lognic04]

Cool! Thanks for the pic.

[Stefan Buda]

Good Progress: I made two secondary mirrors, a second Back Plate and Dovetail plate.
I had bought three 120 diameter blanks, so I turned one into a grinding tool and used the rest to make two identical secondary mirrors. Looks quite certain that I will make more than one of these instruments and, in order to bring down the anodising cost, I also duplicated the Back Plate and Dovetail Plate.

Never stops amazing me how long it takes to get an optical surface from about 1/4 lambda to something like 1/10 lambda (on the wave front). The damn law of diminishing returns really tests one's patience.

[Imme]

Looks great....what kind of time are you talking to go 1/4 to 1/10?

[Atmos]

Excellent work Stefan, great to see this project slowly coming together.

[Lognic04]

Wow, you seem to be very fast at this process?!

[multiweb]

Hi Stefan, what are we looking at? Parallel lines? Can you make the difference between 1/4 and 1/10 just looking at the patterns?

[Stefan Buda]

The closer one gets to the desired precision, say 1/10 lambda, the shorter the figuring spells become between testing. The problem is that after a short figuring run, that can get down to a few seconds, one has to wash and dry the optic before placing it on the test plate and then wait at least 15 minutes for the temperatures to equalize for a reliable reading.
After that the optic goes back onto the polishing lap where one also has to wait for the temperatures to equalize. So one hour goes by for a couple of short figuring spells. And many are required unless you have some supernatural ability to predict the behaviour of the polishing lap.



Thanks Colin.



I hope to become faster with more practice.



Yes the spacing between the interference fringes represent 1 lambda on the wave front regardless of the relative tilt between the two surfaces. In other words they show the size of the air gap between the test plate and the optic. If the two surfaces have the same shape, the fringes are straight.
It is very easy to see a deviation of 1/4 spacing, especially if one spread out the fringes. I kept the fringes rather tight for the pictures as it was easier to photograph them.

[sharpiel]

It's amazing to learn from you Stefan. Thanks.

[multiweb]

So the test plate has to be perfect? Did you make it as well?

[Stefan Buda]

http://www.iceinspace.com.au/forum/showpost.php?p=1399803&postcount=134

[multiweb]

Thanks. I understand now. All this polishing/figuring business fascinates me but I have no idea how it's done hence the silly questions. By the sound of it in your posts although there is a great deal of experience needed it seems to be an iterative process that can be tedious to achieve a high level of figuring, so there's no free lunch. What I find scary it that it's like sculpting. You can't add back if you remove too much and the potential to stuff up must be getting higher and higher as you're nearing your target figure.

[Stefan Buda]

Unfortunately figuring is a bit of a black art that cannot be learned from books. It is very difficult to get a full size polishing lap to do exactly what you want it to do. The polishing pitch is a very viscous liquid that is exposed to various amounts of compression and dragging forces that make its behaviour rather unpredictable. A sub-diameter polishing lap is easier to predict but it is not as good at controlling astigmatism. Figuring is not as bad as sculpting because if you dig a hole somewhere, then you just have to polish away the rest of the surface until the hole disappears - not too hard for spherical surfaces but a nightmare for aspheric ones.

[multiweb]

I reckon it's a skill you must acquire during a lifetime doing it and start young too. Mentoring sounds like a necessary step as well. There are tools to do the job but in the end it's a very organic process and depends on the individual.

[Stefan Buda]

Today I received my corrector lenses back from Longman Optical in Tasmania and I' very happy with the coating job they did. I tried to photograph the residual reflection and it is hard, but you can see on the attached picture just how reflective the uncoated edge of the lenses is, compared to the rest of the surface. It cost around $90 per lens.

Also the primary mirror has reached the final stages of figuring. So far I did it all with the full lap, MOT, and the figure is beautifully smooth.

Tomorrow I'm off to VicSouth so there will be no progress for a few days.

[Stefan Buda]

When I saw the weather forecast, on Friday morning, I aborted the VicSouth trip even though the car was almost fully loaded. Pushing glass instead.

Longman Optical supplied a reflectivity graph for the coatings.

[Stefan Buda]

Yesterday I finished sculpting the primary mirror and that means all the optics are done.
Figuring an f/2.7 mirror is quite an adventure. I did most of it guided by a Ross Null setup and when it became difficult to make more progress, I switched to the Bath interferometer. At that point I was within one lambda wavefront error but it was a long road from there to 0.035 lambda rms.

The attached report is based on only 5 noisy interferograms taken at 72 degree rotation so the mirror is probably better than the report indicates as air currents, vibrations and other artefacts have not been properly averaged out.

[phomer]

Stefan,


Sometimes you just have to ignore the forecast, only Sunday night was a complete writeoff, Saturday was basically clear all night, Friday and Monday had a few hours each.


Paul

[Stefan Buda]

Paul,
I intended to go for two nights only and it looked like the second night was going to have high cloud, so I thought it was not worth the long drive.

[phomer]

Stefan,


I agree, it's a long drive for two nights, but at its best is a very dark site.



Paul

[Stefan Buda]

I spent a couple of days vacuum coating telescope mirrors - including the LOGAN #1.
All the optics for the CDK are now complete.

[Lognic04]

:party2::clap::bowdown:

[Atmos]

The Logan #1 is almost a plug for the Stefan :lol:

Looking good :thumbsup:

[Stefan Buda]

Colin,

You obviously have not followed Logan's project. His first mirror is the same size as my CDK primary and it is not shown on my photo. The smaller mirror is the secondary of the CDK.

[Stefan Buda]

The housing for the corrector lenses is done.

[multiweb]

So first lens goes in, then spacer, second lens and last ring to secure? Is that how it works? Does the orientation or order of the lenses matter?

[Lognic04]

Stefan, the mirror works amazingly!!!
The moon was beautiful at high power and the homonculous nebula was visible too!!! wow!
Thank you so much!

[Stefan Buda]

Yes, that's how it goes. Not only orientation and order matter but also spacing and centration. The key is the spacer, and it must be made to very accurate dimensions, both of it's diameters and the thickness. The lenses are a bit smaller in diameter than the housing but the spacer is not.


Logan,

I'm very happy for you. When you become rich and famous, you can buy me a beer. Providing you reach adulthood by then. :lol:

[multiweb]

Thanks for the explanation and cross section, Stefan. :thumbsup:
Do you provide these little gaps so the aluminium casing doesn't press on the glass with temperature changes?

[Stefan Buda]

That is one reason. Another is to give the lenses the freedom to self centre themselves. For that to work, with such small gaps, the lenses must have very small wedge errors.

[Joshua Bunn]

really enjoying reading these updates and looking at your fine work, Stefan.

[multiweb]

Just one more question(s) Stefan, then I'll leave you alone ;)
Is the rotational aspect important when stacking small lenses like it is for bigger surfaces like mirrors in a folded light path and what is achieved by rotating optical components on their axis? I read it can add up or cancel errors. Is that all it is and does it make a big difference in your opinion or it doesn't really matter? Is that something you test when you do the final assembly with your mirrors?

[Stefan Buda]

In an optical train every element has a tilt error (and other errors too), simply because nothing can be made absolutely perfect, and tilt errors cause astigmatism, therefore by rotating various elements relative to each other, it is possible, sometimes, to minimise astigmatism. But that only works if the effects of individual errors have similar values. For example, tilt errors in a field corrector may be able to compensate to some extent the residual astigmatism of a primary or secondary mirror in a cassegrain system.

To answer your question: It all depends on how accurate the optical components have been manufactured and mounted. It is possible to keep the errors small enough so that rotating elements won't make any difference. That is what I aim for. After I assembled my field corrector, I screwed it into a jig on the lathe and looked at the reflections of a laser beam while it was spinning. The reflected spots, projected onto a surface some distance away, showed little wobble, so I'm confident that it will be ok.

[Stefan Buda]

Don't know how I managed to do ATM without 3D printing in the past.
The primary baffle is done and it would have been a pain of a job without 3D printing/carbon filled filament. The whole thing, including the aluminium ring, weighs only 75g.

[LewisM]

Beautiful job, as always - you'd swear looking at it that it was just a sandblast anodise.

[Stefan Buda]

Thanks Lewis, I like the stuff but it is a bit slow to print - those two cylinders took about 7hrs each. This filament is very abrasive and I should invest into a ruby nozzle as the SS one I'm using is not a good heat conductor.

[Atmos]

Do you think it would be possible to 3D print an OTA structure? Given that it took 7 hours to print those bits it would take a week or 3 to print an OTA! I’m curious as to whether it would have the same structural strength as the way you’ve made the carbon fibre OTA?

[bratislav]

First, most 3D printers won't be able to print something as large as an OTA. Second, carbon charged plastic will be nowhere near as stiff/strong as proper skinned tube. Carbon fibers in a cloth (that is used for lamination) are contiguous and very long; it's their extraordinary resistance to stretching that gives stiffness and thermal stability. Carbon in 3D printed items is a mass of very short and small fibers that are not connected in any way. Better than ordinary PLA or ABS, sure, but you will never see any bows, arrows, fishing rods, skis, surfboards or telescope OTAs (anything that requires stiffness or thermal stability) 3D printed.

[Atmos]

Had a feeling that would be the answer :lol:

[Stefan Buda]

Bratislav already answered your question so I'll just add a couple more comments.

Carbon filled filaments have a loading of max 20% fibre - a lot less than laminated CF.

"Delta" 3D printers would be perfect for the job as they can print very tall objects.

Once graphene becomes cheap, it may be possible to produce printer filaments based on super plastics, such as PEEK, that could possibly have the necessary characteristics for structural parts.

Already there are 3D printers that can imbed contiguous fibres in the print, but the cost is prohibitive for OTAs.

[Stefan Buda]

At long last I got the two backplates engraved so next week I can take them to the anodizers. I used a friend's CNC mill and I gave serial number 000 to the prototype.

[multiweb]

Looks great. First of many?

[Stefan Buda]

First of two, for now. We'll see.

[Stefan Buda]

As I was making the hub for holding the primary mirror, I realized that I can do a test of the finished mirror for eccentricity and co-axiality before taking it off the lathe.
It spun without any perceptible wobble so I'm happy with it.

[multiweb]

How's the glass mounted? Is there an aluminium bush running through it with a retaining ring. Do you use cork as spacers?

[Stefan Buda]

Yes, there is a flanged bush going through the central hole and there is another flange at the front. The front flange has a thread for attaching the baffle. The mirror's location is defined by its back face and its internal diameter. The front flange also has a silicone rubber O-ring that pushes the mirror gently into contact with the bare aluminium flange on its back side. I may add a very thin shock absorbing layer, but definitely not something as rough and agricultural as cork.

[strongmanmike]

So glad the chuck didn't accidentally come loose....:question:.....:scared:

Mike

[multiweb]

Just like in the case of the machined grooves in your corrector lenses arrangement how do you work out the gaps or compression zones dimensions needed in your mirror mounting when you have glass in direct contact with aluminium? When temperature changes, too much and it gets lose, too tight and the glass will be pinched.

[Stefan Buda]

I made the diameter of the hub 0.05mm smaller than the hole in the mirror, at 20 degrees C. That should be good enough up to at least 80 degrees ambient. You can double check the math and let me know if I made a mistake - the diameter is 73mm and the alloy is 6061. I also have an O-ring on the hub to stop the mirror coming loose when the temperature drops. The amount of compression on the O-rings is based on nothing more than educated guesswork and it can be adjusted later, if needed. I made the axial gap also 0.05, a bit more than it is needed because the front flange is conical rather than spherical, so it does not conform very well to the front of the mirror.

[multiweb]

Thanks for the explanation Stefan. Makes sense. I now realise precision machining and knowing exactly how much play you have goes a long way with potential thermal issues down the road. Packing with cork is a bandaid.

[keller60]

I have been reading this in the background with great interest. The attention to detail has been remarkable and am looking very much forward to seeing the first light results :thumbsup:

[Stefan Buda]

That's correct but don't get me wrong about cork. I think it has its place in ATM but for this particular job I don't think that it would be reliable enough.
If you remember, at the beginning of the thread, I said that there will be no collimation adjustment for the primary mirror, therefore its position must be very well defined/constrained.



Thanks. The plan was to have some sort of first light by the end of the year and it looks like it may happen.

[multiweb]

Understood.

[Stefan Buda]

The backplate has been anodised very nicely at Elecrtromold and I was able to drill the electrical contacts, post anodising. I will use the backplate as common electrical ground for the fans and the heaters in order to simplify the wiring behind the primary mirror.
The heaters will be 3D printed with conductive filament. I have experimented with two different brands and the one from Jaycar turned out to be suitable, but it is so brittle that I can only print with it on a very hot day.
Apart from the heating elements and the dew shield; all mechanical parts are done and I will start assembly over the coming days.

[Retrograde]

Wow - Looking forward to seeing pics of it all coming together.

Although I haven't commented up until now I've really enjoyed following this build thread (and of course your Riccardi Honders build thread as well).
Great to see someone building more complicated optical configurations with such craftsmanship.

[multiweb]

Looks very good. :thumbsup: Is that bore (threaded?) conical in the middle?

[Lognic04]

Very cool!

[Stefan Buda]

Thanks Pete.


The bore is threaded on the other side. The conical bit is just a stray light trap.

[Stefan Buda]

The tricky operation of bonding the back plate retainer pads is done.
I had to throw into battle my Taylor Hobson micro alignment telescope that I bought years ago at a Thales auction and never used until now.

It required the making of a precision adaptor that is bolted to the back plate.
I removed the back plate and put it back a couple of times, after the epoxy had set, and the repeatability is very good. The line of sight always hits the centre of the secondary hub within 0.001" .

[multiweb]

I was watching the video about the various applications of the micro alignment telescope and read the brochure on the Taylor Hobson website. What did you exactly use it for and what was the procedure? Squaring the back plate? Centering the secondary?

[Stefan Buda]

I used it to make sure that the axis of the back plate was pointing exactly to the centre of the secondary hub while the epoxy bonded the retainers into their exact locations.

[multiweb]

Was that with the center spotted secondary in place or did you have a very small centered marker at the front to achieve that level of precision?

[Stefan Buda]

The shaft of the stepper motor has a tiny dimple in the centre and I aimed at that.

[Stefan Buda]

I did another test today, with excellent result. I wanted to see just how rigid the backplate was with the four attachment points, so I mounted the tube on my EQ6 and did a horizon to horizon sweep. This is a severely overdone test, considering the weight and length of the alignment telescope, which is about 5Kg. It is like attaching a 10Kg camera at the right spacing. Well, the line of sight did not deviate more than 0.002", and that is better than what I was hoping to achieve.

Next I will test the rigidity of the spider assembly, and for that I will have to make an accurate crosshair target.

[multiweb]

The stepper motor at the front that is part of the mechanism to focus the secondary?

[Stefan Buda]

Yes.

[sharpiel]

Impressive workmanship down to minutiae. Love your OCD attention to detail.

[Stefan Buda]

:lol: Engineering without OCD approach is just agricultural machinery level.
That is why people spend a whole year taming their new astrographs. Not all brands, of course.

[Paul Haese]

You're a legend Stefan and yes your attention to detail is significantly better that on my AG12. Your anodising is done correctly for one thing. Not mention I doubt my focuser is actually square to the optical line. I really enjoy watching you put a scope together, knowing you have made the components. Truly excellent work, I'd love to own one of your scopes.

[Stefan Buda]

Thanks Paul,
Although the corrected Newtonian is as good optically as any RC or CDK, mechanically suffers from the fact that there is no easy way of aligning the mechanical axis of the focuser with the optical axis during construction.

Meanwhile I took advantage of the heatwave and printed the heater elements. Also I tested for secondary flexure by using a dummy secondary mirror made of scrap metal onto which I attached a small flat mirror. I used a crosshair target in front of the alignment telescope and focused on its reflection in the dummy secondary. There was no significant flexure going horizon to horizon.

[Stefan Buda]

Last day of the year and I'm ready for first light. Sort of.

Actually I did hit a problem with the focuser: The image shift during direction reversal is a bit worse than what I was hoping for. I hooked up a MyfocuserPro2 board to the stepper motor and used a test program to drive it back and fort while watching the crosshair target reflection with the Taylor Hobson. I was getting an image shift of about 0.25mm and I could not fix it, so I'm taking the front end back to the drawing board. I will make a new mechanism with 4 guide pins instead of 3, and if that still comes short on performance, then I'll move onto using miniature linear ball bearings.

Does anyone of you have any experience with commercial secondary focusers? I'm curious whether they produce any image shift on direction reversal. Zero shift seems like a difficult thing to achieve without making the mechanism much bulkier. I'd like to avoid that.

[Stefan Buda]

I achieved visual first light just hours before the end of the year.
The position of the focal plane is spot on, collimation is dead easy with the two thumb screws and the optics show no aberrations with a 10mm eyepiece.


Because I anticipated problems with the focusing mechanism, I did not glue in the front end assembly and that will allow me to make a new redesigned one. Meanwhile I'm converting the existing one to a 4 guide pin version to see if I can make it work well.

[multiweb]

It's looking good Stefan. :thumbsup: So when are you going to stick a camera in there?

[Stefan Buda]

Where the eyepiece is in the last picture. Almost too much back focal length without a classic focuser. If I fail at getting the secondary focuser working well, then I can always fall back onto using some sort of Crayford type. I would hate to do that, because of all the potential alignment problems. Just because a Crayford type focuser works well, it does not mean that it has no co-axiality/co-linearity and orthogonality problems. Those are difficult errors to deal with in a Crayford..

[multiweb]

Ha yeah I had figured that one out. :lol: I was wondering when. Can't wait to see the fov and image scale on this thing.

[Stefan Buda]

:rofl: Yes, my eyesight must be deteriorating. Looks like it may be a while until I stick a camera in it. I reworked the focuser to a 4 pin version and although it works better in some ways, it did not eliminate image shift.
Back to the drawing board for a while.

[Joshua Bunn]

Lovely piece of work Stefan. I've enjoyed following this.

[Stefan Buda]

Thanks Joshua.

I've been experimenting with more modifications of the secondary focuser. I changed the spring-loaded nut mechanism to a floating nut solution that decoupled it from the mirror carrier plate. It works beautifully but the tilt hysteresis remained the same, unfortunately. But it occurred to me that there is a software solution to this problem. Any loaded motion system suffers from backlash and hysteresis but a proper anti-backlash routine can deal with both problems. So on the short run I will leave the front end as it is and I will start testing the optics on the stars. For that I will have to 3D print a temporary box for the full version of the myFocuserPro2 controller. Eventually I intend to make my own controller that also deals with thermal management.

Initially I won't be using autofocus software so I made a Bahtinov mask for manual focusing.

[Stefan Buda]

Just about ready for testing on the sky - with a temporary myFocuserPro2 controller.

[sharpiel]

"SN: 000"

Love it!

[Stefan Buda]

Thanks Les.

I did some testing last night despite the not very good conditions.

Collimation took only a few minutes, with my ZWO ASI120 watching the defocused doughnut of Canopus and adjusting the two thumb screws. The long baseline between the adjusting screws and the 0.5mm pitch of the screws makes the tiniest adjustments very easy.
The attached image is from a 1min AVI using 100ms exposures, stacked with Autostakkert and stretched in AstraImage. I'm quite happy with the optics and tonight I will try to use a bigger sensor.

[multiweb]

Very nice trap! It's alive! :thumbsup:

[Stefan Buda]

Thanks Mark. :lol:

I sent Bratislav some Canopus doughnuts to be put through his Roddier analysis software and he has just sent me the results. The optical train came it at over 0.92 Strehl and I'm happy with that. There is a small amount of residual astigmatism that we haven't been able trace. It could be thermals related or a small amount of mirror pinch. Rotating all the individual optics, did not change the orientation so it has been turned off in the attached report.
Damn good this test. Clearly shows the low zone near the edge of the primary, that I knew about (not significant), and the slightly turned up edge.

[bratislav]

Yup, it also easily picked up the "by design" residual high order spherical at mere 13nm! (undercorrected primary plus two lens corrector leave some minor residual aberrations uncorrected - this is unavoidable consequence of all CDK's).
It always surprises me to see that people still go to inordinary lengths using all sorts of lyrical adjectives and convoluted descriptions to describe how good or how poor their optics are. Roddier spits out all the gory details, in numbers, not bullspeak. And it takes just about the same time to perform as a decent (eyepiece) star test would.
Yes, it does take a bit of a learning curve as well as decent preparation, but it still amazes me to see this test not being more widely embraced by amateurs.
Perhaps it leaves too little to wax lyrical (well, none) - it sounds much more boring to talk about secondary quadfoil in nanometers rather than of "diamond specks of dust on a black velvet" and "needle pinpricks of light". :P
Oh well.

[Stefan Buda]

Thanks Bratislav for helping me with this. One day I will have to tackle that Roddier learning curve.

Last night I tried to produce an image with a larger sensor, to asses the off axis field correction. I had a few problems that were not OTA related. I had installed a guide scope, but forgot to add extra counter weight to the mount. So the mount was being driven down hill and it was quite windy too, therefore my tracking was quite bad. Not to mention that polar alignment was not very good either. In the end I managed to capture a bunch of 10sec frames of IC2391 with a few calibration frames. My old QHY8 camera is not responding to gain and offset changes since I started using a different laptop - another issue to sort out.
Anyway, the wind died down completely by the time I packed up, and the data collected was good enough to convince me that I a have nice evenly corrected field corner to corner.

[multiweb]

Yep! Looks excellent! :thumbsup:

[Atmos]

It’s fantastic to see this coming along so well Stefan!

If you do want to test out a larger sensor than the QHY8 I have an ASI094 you could borrow; Full Frame 36.3mp sensor.

[darkage]

Awesome project and amazing build quality. Dare I say how much did this end up costing you ?

[Stefan Buda]

Thanks Colin for the offer, but first I want to make a new front end with hefty linear bearings. I managed to come up with a design that stays within the original physical envelope.


Thanks Gareth,
It cost less than an equivalent size Newtonian built with off the shelf components, plus a few hundred hours of labour.

[Troy]

A good looking scope that performs well. Congrats :thumbsup:

[Stefan Buda]

Thanks Troy.

I'm about one month behind with this project, but finally ready to take it to a dark sky location.
I made a completely new/redesigned front end (focuser + spider assembly) and it seems to work incredibly well. Visually I can't detect any image shift when I reverse focusing direction. Maybe the camera will see a couple of pixels but it won't be an issue. It would be very interesting to hear from Diego about his super duper secondary focuser in this regard.
I also made the dew shield and rigged up a guide scope.

[Stefan Buda]

I did take the scope to a dark sky location for testing but I had to evacuate the place, because a total fire ban was issued for the area, before I had a chance to capture a good image set. However I did a number of tests and I'm very happy with the new front end. The focuser has a backlash of only 4 or 5 steps and the image shift is so small that I had to turn off the drive and capture star trails, while pushing the focusing buttons back and forth, to make it show up as a discontinuity on the trails.

The only issue I had was the inadequate guiding. PHD was producing an oscillation index of half a pixel, which at the CDK focal plane contributed about two extra pixels to the star sizes. The guide scope is less than a third the focal length of the main OTA. So I decided to knock up an OAG for the next test and try to improve the PHD settings in a way that will stop it from trying to chase the seeing. It will be interesting to see if my old EQ6 mount can deliver.

On another front I entered all my notes into Excel tables and worked out a very detailed costing. It turns out that the OTA, including dovetail plate and camera adaptors, contains exactly 280 parts, down to the last screw.
Calculating with a rather low hourly rate, the price worked out to $8805 including GST but without the focuser control box and dew shield. The controller will take the price to around the $9000 mark.

I would like to hear some feedback on what you guys think about commercial viability at this price. Keep in mind that this is a very user friendly, compact and capable instrument without collimation/field flattener/focuser nightmares.

[AstroApprentice]

Hi Stefan,
The scope looks amazing! Could you please list the full specs, including optical parameters, dimensions, weight etc?

[glend]

Stefan, that figure is your cost right? So what profit bargain are you going to put on your materials, effort, etc? Given usual retail markup, I suspect you would need to charge $15k minimum.

[Stefan Buda]

Aperture 250mm
Focal Length 1450mm
f:5.8
Fully Corrected Flat Field Diameter 42mm
Secondary Obstruction 50%
Quartz overcoated mirrors
Broadband AR coated lenses
Back Focal Length (From Field Rotator to Focal Plane) 133mm
OTA Diameter 299mm
OTA Length (without Field Rotator) 496mm
OTA Weight (including Rails, Field Rotator, Dovetail Plate and Adaptors) 10Kg
Built in primary and secondary heaters


I did not apply any mark-up for design and development cost recovery because I would have made this instrument anyway for my own use, and I wanted to make the price as affordable as possible. If I start making them for sale, then I will have to find ways of making them more efficiently in order to recover those initial costs.
Just to clarify: The Price I worked out includes both the labour and the cost of materials.
Also no retail pricing applies because I do not intend to make them available for retail. At least initially I intend to make no more than a couple per year as I have commitments in the laser industry.

[Paul Haese]

Hi Stefan,

If you built a 12" or 14" I would be certainly interested. I expect the price would be reflective of the size but I am interested.

[Stefan Buda]

Hi Paul,

One of the reasons for 10" aperture is that my machinery is limiting me to that size. A larger one would require outsourcing parts, with all sorts of quality control and cost issues.
Also I think that there is a bit of a gap in the market at the moment for portable 10" astrographs.
This one travels nicely on the back seat with room left for an Esky.
So, I'm afraid I can't commit myself to designing a larger one at this stage, but who knows what the future will bring.

[AstroApprentice]

Hi Stefan,
Just wondering if you could comment on how you think your design compares to other CDKs of same aperture? I note yours has faster f ratio than these two:
http://www.hubbleoptics.com/cdk.html#_
http://agoptical.com/10-imaging-dall-kirkham/

Also, What’s the central obstruction %?
Thanks.


[QUOTE=Stefan Buda;1416138]Aperture 250mm
Focal Length 1450mm
f:5.8
Fully Corrected Flat Field Diameter 42mm
Back Focal Length (From Field Rotator to Focal Plane) 133mm
OTA Diameter 299mm
OTA Length (without Field Rotator) 496mm
OTA Weight (including Rails, Field Rotator, Dovetail Plate and Adaptors) 10Kg

[Stefan Buda]

[QUOTE=AstroApprentice;1416188]Hi Stefan,
Just wondering if you could comment on how you think your design compares to other CDKs of same aperture? I note yours has faster f ratio than these two:
http://www.hubbleoptics.com/cdk.html#_
http://agoptical.com/10-imaging-dall-kirkham/

Also, What’s the central obstruction %?
Thanks.

I added a few more things to the specifications list.

My CDK design is 1.9 times faster than the Hubble one and 1.33 times faster than the AG Optical.
Mine is significantly lighter and more compact than those.
Mine has larger central obstruction than the Hubble and smaller than the AG.
My design is limited to 42mm diagonal sensors, the others were designed to work with larger ones.
Mine has built in heaters while for those it is optional extra.
Mine uses a secondary focuser, allowing a very clean, rigid and precise imaging train.
Mine comes with a precision camera rotator while the others have it as optional extra.
I can't comment about the mechanical build quality for those, but mine is as good as it gets.
Last but not least my price seems to be better too and you are dealing with a local manufacturer that is eager to help you out if you get stuck.

[strongmanmike]

[QUOTE=Stefan Buda;1416235]

Holy cow, now that is a compelling and convincing sales pitch :prey2:

Mike

[multiweb]

Agree. These should sell like hot cakes. :thumbsup:

[Stefan Buda]

Thanks Mike and Mark for the feedback.

[bratislav]

You should add that :

- primary is precision edge ground with zero wedge, and will never require collimation after it is factory assembled
- no tools of any kind are required for secondary collimation, just fingertips
- both primary and secondary are heated and can be kept dew free, greatly enhancing the longevity of coatings
- secondary focusing combined with ultra rigid back plate means no flexure/tilt whatsosever, even with heaviest cameras, off axis guiders, adaptive optics devices and filter wheels

and most importantly:
- guaranteed optical quality, with every unit tested via intereferometry as well as Roddier on real sky. No ifs, buts or excuses of any kind

[Atmos]

For the quality of what you’re making it’s a very good price Stefan.

[John K]

Having seen the scope Stefan has built, I can categorically say it’s a very impressive instrument and way above anything commercially available. Everything is over engineered.

Stefan’s passion for making impressive optical instruments is second to none and he recently completely refurbished and rebuilt a second hand scope for me to take the instrument to another level.

To top it off, Stefan is a great guy who will go out of his way to help others in the amateur astronomy scene and above and beyond just making something for them.

John K.

[Stefan Buda]

Thanks Bratislav,
I didn't want to make the reply too long. :lol:
Yes, basically it is possible to have the primary removed and then put back without upsetting collimation.
The mirror coatings are not quite as reflective as the multilayer dielectric coatings claimed by the other brands, but the protected aluminium coatings, in mine, can be redone while for the others it's the end of the road if they deteriorate.



Thanks Colin for the feedback.



Thanks John for the kind words.

[traveller]

I will second what John said. Stefan is a all round nice guy with an designer's eye for detail and an engineer's demand for precision.
I've been following this thread and it's an impressive piece of work!

[Stefan Buda]

Thanks Bo for the kind words.

I left the scope set up last night and woke up early this morning so I snapped some lo-res moon images.

The attached image is a single 8ms capture with the old QHY8 OSC.

[Stefan Buda]

I tested my improvised OAG last night and it works, but I will need to fiddle with the PHD settings to get it better. Even as it is now, I get better guiding than with the guide scope.
By the time I sorted some camera problems, it got late and even the moon was up, so I just grabbed 10x60sec subs of Eta Car and some darks to match.
Should be good from a dark site with more subs and calibration frames.

[Stefan Buda]

Went to the Messier star party last night and I wanted to do a 3 panel mosaic of Eta Car, but unfortunately I was not able to solve my guiding problems. One of the axes misbehaved all night. Tomorrow I will perform autopsy on the old EQ6. I bought it second hand many years ago and I always used it with short focal length instruments until now.
The attached image is a roughly processed bunch of 59 minutes of not very good subs - due to poor guiding. Full frame of QHY8 resized to 40%.

[DavidU]

Remarkable work Stef, my hat is off to you.:thumbsup:

[traveller]

Very promising results Stefan,
Any chance of you coming to Snake Valley this weekend for some fine tuning of your CDK?
Bo

[Stefan Buda]

Thanks Dave.


Hi Bo,
I ripped apart my mount this morning and made some improvements to the mechanism. The DEC axis idler gear was able to move sideways because the shaft was smaller than the hole in the gear. I wanted to make a better fitting shaft but I discovered that the hole in the motor plate, for the shaft, was not drilled squarely. I ended up drilling a larger hole, making a new shaft, and increasing the hole in the idler gear. To get to the point, if I can get the mount working satisfactorily, I will zip up to Heathcote again to have a go at producing an image. At this stage it looks like it will be Wednesday, if the forecast holds. Unfortunately Snake valley is a bit uncertain weather-wise and I don't want to risk wasting time at this stage. One day I would like to make it to Snake Valley though.

[Atmos]

Guiding issues aside it looks like an excellent performer Stefan :thumbsup:

[Stefan Buda]

I'm very happy with its optical and mechanical performance, but with my present camera I won't be able to get the max resolution out of it. For that I would need a mono camera with smaller pixels and probably a better mount.

Meanwhile I made another CF tube and managed to avoid some of the cosmetic issues I had with the first one.

[Stefan Buda]

I should have posted a 100% crop from the last image to better show the resolution. With a well matched pixel size mono camera and good tracking/guiding it should have no problem exploiting the full resolution that seeing allows.

At the moment I'm in the process of selecting and ordering components for the controller box so that I can start designing the PCB.

There will be no updates for a while as I'm also taking two weeks off for a late holiday.

[sharpiel]

Enjoy yourselves.