16" Dall-Kirkham rebuild

[Stefan Buda]

This instrument has been my planetary workhorse for over 20 years and is in desperate need of refurbishment and upgrade. I've been wanting to do this for several years now and finally seem to have the time to get stuck into it.

As there are only three types of telescopes, in my opinion, that are good for cutting edge planetary imaging, and one of them being the Dall-Kirkham, I thought there might be some people on the forum that could be interested in learning about this type of telescope. I'll be happy to answer any related questions.

My intension is to convert the Alt-Az mount to Equatorial fork mount and get rid of the field rotator mechanism.
A new electronic focuser will need to be designed and made.
I intend to simplify the primary flotation system.
The active cooling system will be retained as it proved itself to be useful.
The old Bartels drive will be replaced by OnStep controller.
All carbon fibre surfaces need sanding and coating with UV resistant polyurethane.
I may even regrind the mirrors because back in 2000 when I made them I did not have an interferometer and the figuring was done with Coude Mask and Ross Null testing.

[Rod]

This is exciting! I’ve always wanted to build a cassegrain. What Focal ratio is yours?

Which are the other 2 planetary scopes you like? I’m guessing a long focus Newtonian and a Schiespiegler?

Rod.

[Stefan Buda]

Hi Rod,

A Dall-Kirkham shouldn't be made faster than f/15 because of the excessive coma. Mine is f/16 so the focal length is 6500mm.

For planetary imaging aperture rules and I consider 10" as entry level but 12" gets you close to the cutting edge. If you accept that, then you are left with the proven C14s, the large Newtonians and the large D-Ks. The D-K being more of a DIY option as I don't know of any available commercially.

[sfarndell]

Following with interest Stefan. A DK is next level challenging compared to a Newt as you also have to figure the secondary.

[Stefan Buda]

Hi Scott,

I don't think a DK is more difficult to make than a Newt, apart from having to perforate the primary. I suppose what makes the Newt easier is that people buy the secondary off the shelf.
Keep in mind that the DK secondary is spherical and can be tested (figured) against a small test spere that is easily made.

[AlexN]

What an incredible instrument! Having read your previous thread (yes... I went through and read all the pages) about the CDK builds, I know that this scope is going to be an absolute weapon once you're done the refirb! :)

Can't wait to watch your progress on this - and to see the final result!

[Stefan Buda]

Thanks Alex for the encouragement and it's nice to know that you enjoyed going through my "phlog".

Meanwhile the new focuser parts are getting made and in a couple of days I should be able to upload some photos of the assembly.

[Rainmaker]

Subscribed !!! Your workmanship is music for my eyes !

[sharpiel]

I wish you had a YouTube channel we could watch you work.

[croweater]

Love seeing what some of our skilled members are capable of. Looking forward to your next photos Stefan.
Cheers, Richard

[Stefan Buda]

Thanks, guys, for the feedback. Good to know that I'm not wasting my time uploading the images.

Well, the new back end is finished. I could have used a CDK250 focuser but I wanted a longer stroke in order to accommodate various configurations without the need to make many different adaptors.
So basically I modified the CDK focuser design by providing it with an extra long inner sleeve and a longer rack. I also decided to use adjustable teflon pads instead of ball bearings. I have designed many different focusers over the years but never one with teflon pads, and I wanted to see how good that option is. So far it seems promising - it was quite easy to adjust for concentricity, on the lathe, and now it is just a matter of how well it will take the test of time.

[Joshua Bunn]

Very nice focuser Stephan, what do you think of the Teflon pads vs bearings?

[Stefan Buda]

Thanks Joshua, my first impression is good. Of course, there is more friction than with ball bearings but not too bad, although this extra friction is expected to add a bit of hysteresis to the existing backlash in the gearing of the EAF.
Always focusing inward should take care of that.
Also remains to be seen how it will be affected by the large temperature swings on the long run - it gets very hot under the tarp in summer.

[Stefan Buda]

A little bit more progress: I 3D printed new ABS grills for the ventilation fans, to replace the old rusty metal ones. The fans are also getting 3D printed skirts to reduce spider habitat and to improve esthetics.

[Stefan Buda]

Time for astro strip tease: The private parts of a dedicated planetary telescope are getting revealed.
Such as the very sexy orange Byers worm gear.
By the way, the whole truss structure with the secondary cage, as seen on the third picture, weighs only 2.6kg, and the big tube only 1kg.

[mental4astro]

Hi Stefan,

If this is a planetary scope, why would you bother with an equatorial set up or a de-rotator? I would have thought that both the exposure times being so short are not a problem and the imaging software takes care of any rotation in the stacking and alignment process. I have no such issues with this and I only use an altaz mount for lunar and planetary imaging. If things can be kept simple and technology gives you an advantage this way, why then go with an eq rig? Or am I missing something?

Alex.

[Steffen]

That’s only for the nearby planets, not the Kepler lot :D

[Stefan Buda]

Alex you are correct regarding the lack of need of a derotator for planetary imaging. Software can take care of that but when I designed this scope that was not the case.
When I think about the simplicity of a system, I also include the software that makes it work. For an off the shelf system who cares what's under the bonnet, but this is a large DIY instrument and I have to rely on open source solutions to drive it. The old Bartels drive is very obsolete and an EQ mount can be driven even by a simple Arduino clock drive if no GOTO is required.
Another thing I didn't like about the Alt-Az was the fact that during lunar imaging it was difficult to know which way were the cardinal directions.
With an EQ mount I can have the camera lined up and it will not change orientation.

[Joshua Bunn]

It would be interesting to know how the movement of the draw tube feels while racking in and out, in terms of resistance, while having enough force to prevent slop but not to much that it binds on the pads.


Is the primary mirror box made from CF composite?

[Stefan Buda]

The movement feels smooth and without stickiness, but the friction is just whatever I think it should be acceptable. The adjustable plugs are M10 x 0.75 and they don't take much to overdo the tightness/friction.
Teflon's high thermal coeficient of expansion is a bit of a worry because I made the pads 3mm thick and I'm not sure what happens when the whole thing gets heated to something like 80 degrees Celsius. I might put it in the oven tomorrow and find out. If it gets sloppy after that, then I'll reduce the thickness of the pads to less than a millimeter.

The whole OTA is made of CF composite but it incorporates aluminium inserts in strategic locations. It also contains laser cut MDF in a few places.

[OzEclipse]

Very interesting project Stefan.

Don't worry about the sort of temperature effects you'll get on the teflon.

Teflon has very good thermal properties. Melting point 327 °C.

In chemical labs, we use beakers and screw top bombs made of teflon to heat liquids that will dissolve pyrex glass and when performing leaches or digestions of some materials. Under no mechanical stress, the teflon beakers and bombs can be used directly upon a lab hotplate.

Teflon is also used as a seal in vacuum systems. It only deforms / extrudes a little when under stress from the vacuum, the seal tension and heated to 140C. It also doesn't fatigue when subjected to large shifts in temperature. I have used it in a cold trap I built for one of our instruments that cycled from liquid nitrogen -196C to heating at +120C. It did this 300C cycle 40 times a day for 20 years. When I decommissioned the instrument, the teflon piece of the trap looked like new.

Under your conditions, I think the teflon will serve you well for the life of the instrument.

Joe

[Stefan Buda]

Thanks Joe,

I know that Teflon is a wonderful material as I've been using it for various things for many years, but it is also a weird material as it has a couple of phase changes where its crystal structure changes reversibly. One of those phase changes happens at about 20 degrees Celsius and I don't have information on what amount of volume change is involved with that.
The focuser is in the oven right now. I'll report back.

[Stefan Buda]

OK, here's the results of the heating test:

I had the focuser in a small oven with a thermometer and after keeping it at slightly above 80 degrees for a couple of hours, I let it cool to room temperature. I found that there was a definite change after the heating. It didn't quite develop a slop, but it wasn't far from. This was at 25 degrees though, above the 20 degrees phase change for Teflon, so I put it in the fridge and cooled it to 12 degrees. This time there was a definite slop. Not much but enough to require a fix.
I think that reducing the pads thickness to less than one millimeter should solve the problem.

[Stefan Buda]

I made the new Teflon pads and modified the SS inserts to match.
After reassembly I repeated the thermal test and found no noticeable change this time.
By the way, each SS insert started life as a standard M8 socket head screw.

[Joshua Bunn]

Great way to make your own grub screws :)

[Stefan Buda]

I also made a new cover for the secondary. The old one wasn't 3D printed and used elastic bands to hold it in place. The elastic needed replacing from time to time as it would deteriorate from the heat under the tarp.

The blue parts are 3D printed with TPU filament and they hook onto the spider vanes. Remains to be seen how well they'll cope with the heat. If they fail, I will replace them with metal strips.

[Stefan Buda]

The finder got a new mounting arrangement. The old one was quite corroded despite being anodized.
Back in the days when I made this OTA, cutting edge planetary imaging was done around a focal length of about 10m because the webcams used back then had larger pixels than today's planetary cameras. That made it very difficult to move from one planet to another as the field of view on the computer monitor was less than 2 arc minutes.
To help with that, I wanted a finder with enough magnification to make this easier but to also be compact. In the end I made my own achromatic doublet of 48mm clear aperture and 250mm focal length. That, combined with a 10mm eyepiece equipped with a custom graticule, worked quite well.
Only one more problem to be solved. On nights with hi humidity, I was getting so much dew on the objective lens that I had to use tissues to wipe it to be able to see anything through it. So a heater needs to be made.

[Stefan Buda]

I had to make a cap for what was until now the AZ-axis. This part allows me to adjust the clamping force on the worm wheel.
Although this was included in the original design, I never made it because the weight of the telescope and the fork provided enough force to engage the worm wheel sufficiently to work well and still allow me to push the scope around in "Dobsonian" mode.
This axis now becomes the RA axis and, without the new tensioning part, the clamping force would be only 61% of what it was before.

[Stefan Buda]

Made an equatorial wedge to suit my location. Nonadjustable for better rigidity. I will be able to fine tune polar alignment using the adjustable pier legs.

[Joshua Bunn]

Quite sturdy Stefan. What kind of adjusters are on your pier legs?

[Stefan Buda]

If you have a look at the very first image I posted in this thread, you will see three large screws that support the whole structure. The M20 screws have locking nuts.

[Stefan Buda]

Picture for Joshua. No welding involved.

[Stefan Buda]

Fork installed.
I will have to hang a big counterweight from the base of the RA axis, to bring the COG closer to the pier. I provided a hole in the base plate of the wedge for that purpose.

[Saturnine]

Hi Stefan
I have been following this thread with some interest, just love atm articles. Just one observation about the screw adjusters for the legs. Would it be better to invert the set screws so that the head is on the ground, not the threaded shank, gives a slightly larger weight bearing surface as well and the lock nut can go either above or below and still be easy to adjust.
Apart from that, great work on the rebuild.

[Stefan Buda]

Hi Jeff,

Thanks for the observations.

I thought that the screws would be easier to adjust from the top as there is no danger of grazing my knuckles on the concrete if I slip up.
The M16 screws, not M20 as I wrote before, can support several tonnes weight and I think this must be the lightest 16" telescope in the world.
I even considered rounding the ends of the screws to give them a bit more of a bite on the concrete, but we don't get too many earth tremors here.
Another reason for having the screws as they are, is to make it harder for the rainwater to get into the threads, which are loaded with grease by the way.
And lastly, I think my toes suffered less damage over the years by not having the screw threads poking out on the top. I'm sure I would've had to implement some toe protectors.

Any comments or suggestions are welcome.

[Stefan Buda]

I extracted the primary assembly from the mirror box.
Lots of sanding and painting to be done.

[sfarndell]

I continue to watch with interest and fascination. Your metalworking skills are outstanding!

[Stefan Buda]

Thanks Scott. Here are some more metal work pics for you.

I will try to not use the side support levers any more. All the planetary imaging I do is above 45 degrees altitude, so I might be able to get away with just the central support and the 9 point flotation approach.

I'll retain the heat pump which worked very well.

[Joshua Bunn]

Quite an interesting mirror support there, Stefan.
Were the lateral supports on a cam adjustment to get the right spacing to the mirror edge?
I'm trying to understand the layout of the last 3 pictures, in the middle photo, are the heat pumps not installed between the plates, just doest look like there is room...
Also, are the 3 sets of 3 supports, independent from the other supports, and do the collimation adjusters act on those supports or the mirror support plate as a whole?
A thought on the m16 leg adjusters, I think a point would be better than flay, on the ground. A rotating flat bolt has a tendency to wonder when turning it, which may alter PA in an undesirable way. :)

[Stefan Buda]

As my friend Diego says, there are two ways of making things: adjustable or right. I prefer the later, so the side supports were not made adjustable. The lead weights are able to swing several millimeters in any direction and that is equivalent to less than half a millimeter movement at the top - the fulcrum of the levers being close to the top. The mechanical advantage is about 8, so each 150g of lead weight can apply a force of up to 1.2kg on the mirror. Because the top part of each lever was bonded to the mirror, the weights were not able to swing but they all worked regardless of orientation or location around the mirror.
Regarding the heat pump, there are details that are not visible on the photos and that makes it a bit hard to understand. The big black disc is the cold plate of the pump and the three big radiators are the hot side where the heat is released into the air stream before getting vented out the back.
Between the cold plate and the radiators, you can see the main structural element that I call the wobble plate. It is made from two laser cut SS discs, with many aluminium spacers in between and a SS hub. On the top of the wobble plate, but hidden by the cold plate, we have the 9-point flotation support. The flotation is made so the there is a 1mm gap between the mirror and the cold plate and it is not adjustable.
The mirror box has a large sturdy SS hub with a spherical rim a the top end, which fits into a cylindrical section of the wobble plate hub.
Each of the three SS pins, at the bottom side of the wobble plate, go through the back of the mirror box and terminate in lockable differential screw mechanisms. The differential screw adjusters can tilt the wobble plate a couple of degrees in any direction, as well as move it along the optical axis for adjusting the mirror spacing.
Regarding the pier leg adjusters, you are right about rounded being better than flat, but this telescope is too big to be moved anywhere, so unlikely that it will need frequent tweaking.

[Joshua Bunn]

So its a kind of astatic mirror support...?
Thanks for the explanation on the supports. Are the 9 points of floatation, fixed in the same horizontal plane, so each 3 do not pivot relative to the other 2?
Looks like a hefty mirror, whats its mass?

[Stefan Buda]

Astatic is correct.
The supports are the classic pivoting triangles variety but the devil is in the details, as they say.
The cold plate has several layers. The main layer is a 3mm thick laser cut disc that is provided with 9 holes for the support pins to protrude. The holes are 10mm diameter, but the pins are only 6 and it is the second layer of the cold plate that stops the support triangles from rotating out of position. The second layer of the cold plate is a 1mm thick cork sheet that has 9 perforations of about 6.5mm fot the support pins.

The mirror is 38mm thick at the edge and started life as a classic full thickness (1 to 5 ratio) Pyrex blank that I sliced into two discs...but don't start me on that story.

[Stefan Buda]

I opted for a hammered silver finish to reduce radiative cooling.

[Stefan Buda]

I 3D printed a couple of custom plugs (grommets) for the wires that go to the Peltier devices and the temperature sensor. I used rubbery TPU filament.

[Joshua Bunn]

Thanks for clarifying the backplate configuration, Stefan.

[Stefan Buda]

I think that the best use of 3D printing in ATM is for making baffles, using carbon filled filaments.

When I designed the 16" DK I decided not to use a secondary baffle, in order to keep the secondary obstruction as small as possible. As a result it was impossible to have a full cutoff of the sky background, which would be bad for an astrograph but not too bad for a planetary telescope.
Now, as an improvement, I 3D printed a couple of additions to the primary baffle for further reducing stray light reaching the camera. As a result, the fully illuminated field will be reduced to about 10mm diameter - more than enough for lunar and planetary imaging.

[Stefan Buda]

With the heatwave gone, I was finally able to test the primary mirror with my Bath interferometer.
When I made this mirror, I relied on zonal measurements with a Coude mask and Ross Null testing, during figuring. I did not have an interferometer back then, so I was curious to see how close I got to the required conic constant and if there were any significant zonal errors.
I wasted half a day trying to get sufficiently stable interference fringes until I realized that the laptop was blowing a stream of warm air into the light path. After that I managed to capture a bunch of usable interferograms. The optical design calls for the primary to have a conic of -0.7298 and my test today came in at -0.731, which is a lot closer than what I expected.

[Rod]

Congratulations Stefan.

That’s an impressive result. How useful did you find the Ross null test?

Rod.

[Stefan Buda]

Thanks Rod!

The Ross Null test is just about as useful as the Ronchi test would be if the mirror was spherical. In other words, it becomes difficult to interpret when you approach the required precision of better than 1/4 wave.

[Satchmo]

At this point you can move to a knife edge , which will allow you to refocus the apparent shape to pick the best work line to work towards a plane wavefont in the null , which is not really a strategy available with a ronchi grating , which just displays the progression of relative slopes from centre to edge .

[Stefan Buda]

Yes, but the knife edge test is much harder to combine with a Ross null lens.
The Ross null is very sensitive to misalignment and the light source offset needs to be kept very small, especially for fast mirrors.
In fact, it is mainly the residual misalignment, that shows up as astigmatism, is what mainly limits the usefulness of the test in the final stages of figuring.
However, the Ross null is very useful for evaluating the smoothness of the surface.

[Stefan Buda]

I had to make small modifications to the gearboxes because I'm getting rid of the old unipolar stepper motors and replacing them with bipolar NEMA 17s.

The SMA panel jack is for the index pulse. It is connected to a microswitch, which is activated by a notch on the larger timing belt gear.

[Stefan Buda]

I made a new power distribution box. The three switches are for the active cooling, the fans and the finder's heater (yet to be made).
The status LED for the active cooling is a colour changing one and it was a pain to get its current limiting resistor right. The cooler should never be on during observing so the changing colours should be a remainder to turn off the Peltiers.
Today I received some very fine Nichrome wire from Temu for making the finder's heating ring.
Also have ordered a new dual thermometer with humidity sensor.

[Stefan Buda]

I added a 10kg counterweight to the equatorial wedge to bring the COG closer to the pier. The aluminium cylinder has room for another 20kg of lead.

[strongmanmike]

Very old school "Stellafane (https://stellafane.org/convention/historic/photos4.html)" like bit of kit, robust and capable looking and even the finder looks to be capable....of holding bird netting..? :question::lol:

I loved dreaming about having many of the masterpieces in the photos from that star party in the 80's

Mike

[Stefan Buda]

Don't let looks fool you, this OTA contains hardly any metal, but if you refer to the mount, then yes, this rebuild looks like a step backward - going from AltAz to EQ. Also, this conversion has turned it into a bit of a Frankenscope even though it was never designed to be good looking. However, this is a special purpose scope and I think the EQ mount is a better option.
Regarding the bird netting, I did not notice the illusion until you pointed it out.:lol:

[strongmanmike]

This epic bit of work inspired my friend and I to construct a massive over engineered Hale 200"-esque Serrurier truss fork mounted 18" F5 Astrograph back in the 80's, even had the massive pillow block bearings....that we never finished, got about 3/4 done, before discovering wine, women and song... sad :(....it really did resemble the Hale 200" except on a fork mount...so more like the 120" Lick perhaps?

Mike

[Stefan Buda]

Thanks Mike for shearing your early ATM (mis)adventure.
That Stellafane scope looks quite a bit out of proportions though - a 12"-er should be a lot more compact, whatever the type of mount.
However, I remember 1982 very well, as it is the year of my escape across the Iron Curtain.
I built my first scope about two years later - just in time for Halley's comet.

[Rod]

Hi Stefan

I’m enjoying reading about your progress. I like the shroud/tube the at sits inside the truss - looks very neat. Is it made of carbon fibre too? How important is it to baffling? Or is it mainly to keep the dust away?

Thanks

Rod.

[Stefan Buda]

Thanks Rod!

The shroud/tube is a single layer of carbon fibre laminated onto a 1m length of Formatube wrapped in aluminium foil as resin barrier. The Formatube was removed destructively. It weighs exactly 1kg and its role is to help with laminar air flow in the light path as the venting fans draw air throug it.

[Stefan Buda]

The heating ring for the finder is done and installed.
It has 55 ohms resistance and that dissipates about 2.6 watts of heating power at 12V.

[Stefan Buda]

Last night I achieved first light of sorts.
After fine tuning the polar alignment, using a broomstick as a crowbar for nudging the pier supports, I went on to tweak the collimation and finally I even did a bit of lunar imaging.
It wasn't smooth sailing though. The mount in this configuration is much more prone to vibrations and resonances. I was driving the RA axis with an Arduino Uno/DRV8825 and getting vibrations that were interfering with the imaging.
Once I changed from 1/16 micro stepping to 1/32 and adjusted the current draw, things were ok, but I will need to do some vibration dampening work.

[JA]

Stefan, you went to the moon last night:thumbsup::thumbsup:

Best
JA

[Joshua Bunn]

Thanks fir the updates, Stefan. I'm enjoying following. Crisp capture of the moon!

[Andy01]

Very impressive first light Stefan, I imagine the results will be spectacular once you've sorted out the remaining niggles. :)

[Stefan Buda]

I'm afraid, this it is the only way for me to space travel. I was born 50 years too soon, otherwise I would be lining up to get on Musk's Starship.:lol:



Thanks Josh, it was a difficult capture as I was only able to correct the RA drift and the Moon drifts in both axes. I have ordered a cheap CNC shield that I will write some code for, and I will be able to make DEC corrections too. On the longer run, I will use a OnStep controller, just at the moment I don't feel like sorting out the multitude of versions that exist.



Thanks Andy, the main niggles will be with the recoating of the primary mirror as my vacuum chamber can fit max 10" mirrors.

[Rod]

Looks really impressive Stefan. Despite your concern about remaining niggles, the Moon image looks terrific. I’m looking forward to seeing more images from it.

Rod

[Stefan Buda]

Thanks Rod,

I posted another image, from the same night, in the solar system section.

I also found a mistake in my Arduino code that was contributing to the drift in the tracking.

[Stefan Buda]

I received the CNC shield V3 that I had ordered, and I can start writing code for it.
Amazingly it cost $10, including 4 stepper controllers.

The old hand paddle worked so well for over 20 years that I will retain it for the Arduino based controller. I only need to change the wiring.

[iborg]

Hi Stefan


This this is Philip, with the binoculars. Thank you for help on the day and for showing me your scope.


As I said, looking very good indeed!


I expect you are well aware of this, but, just in case, you should set the current limit for the stepper motor on the driver.


A link showing this is here (https://www.pololu.com/product/1182), about 3/4 down the page.


Have fun


Philip

[Joshua Bunn]

Nice Stefan,


Is this a manual machine to CNC conversion?

[Stefan Buda]

Thanks Philip, yes I have been using the DRV8825 drives for a while on my DIY engraver, with the same CNC shield. With the small NEMA17 motors I set the current limit just a bit above 1A.



Josh, this CNC shield is open source, and it is used for all sorts of DIY engravers and 3D printers. The firmware for CNC control is available from the GRBL website. I am writing my own firmware for the telescope control.