I have just acquired an Astro-Telescopes 102mm achromatic refractor that I intend to mount directly on the CF tube of the 12.5" astrograph to use as a guidescope.
The main resons for choosing this particular refractor are as follows:
1. Economical ($USD 499) but still having good (achromatic) optics - more than adequate to achieve good guiding.
2. Relatively long focal ratio (f/11 - remember the days when these f/ratios were common?). Not only does this give a good focal length for guiding relative to the main scope (1122mm vs 1600mm), but the longer tube permits wider spacing of the tube rings for greater stability.
3. Reasonable aperture (102mm) without being excessively heavy (just over 6kg with rings).
However, although there were good reviews for the optics and it sports a nice tube with a generous dew cap, I expected that it's Achilles heel would be the generic (made in China) crayford focuser (I had seen one review on the web where the owner had problems). Nevertheless, I thought that even without the focuser it was a good deal so I purchased it. Since it will be used for guiding, I intend to add an extension before the focuser such that the camera will be in focus near minimum drawtube extension (for stability). I took some measurements of the focus position to facilitate this (see picture showing the AT102 on a fully adjustable AltAz-Beanbag mount ).
However, I wasn't wrong about the quality of the focuser
Crayford Focuser Repairs
The first thing I found was an annoying ding in the end of the 1.25" adapter. On inspecting the box again, I found a slit right through it that had escaped my notice which lined up suspiciously well with the focuser. Oh well, not the fault of the focuser and just cosmetic ...
When I turned the focuser knobs, I found that it sounded very scratchy and the fine focus knob turned in unison when I turned the coarse focus knob (not correct - the fine focus should turn 10 times faster when you turn the coars knob). I found that some of the grub screws were not tightened, but when I tighted one I found it was cross threaded and turned freely without tightening. On investigation, I found that the base of the threaded hole was not well formed and the grub screw was too short (so did not engage the upper hole/threads). I replaced this with a longer (M4) grub screw that I had and this solved the problem.
The next problem that I found was that the fine focus seemed to "grab" every third of a turn, as if there were 3 detents per rotation. Of course, this should not be, so I began investigating; Gary Hand (HandsOnOptics) said that the most likely cause was a bent pin. This is used as the centre "gear" for the planetary gear system that enables the slow motion (central pin rotates and contacts three large ball bearings which are set in a brass fitting that is attached to the main drive shaft. I found a good article on how to disassemble a similar GSO crayford focuser (see http://backyardvoyager.com/astronomy...ur-gso-focuser), but it didn't go far enough. I needed to completely disassemble the fine focus ball bearing mechanism to get at the centre pin. Also, instaed of a hex nut holding it in, my focuser had a round nut with a slit (similar to a straight screw head), so I had to make my own tool to undo it (see attached pictures). Once I managed to disassemble it completely, I found that the central pin was perfectly straight . It had to be something to do with the pin I reasoned, as the false "detent" matched the rotation cycle perfectly. The ball bearings are held against a groove around the circumference of the pin near one end, so I inspected the groove with an eyepiece, and found a very slight mark/depression within the groove (see the picture to get an idea of how small I am talking about - probably of the order of 10-20 microns deep!!!). This had to be it, as combined with the high mechanical pressure applied, this would mimick a detent I reasoned. I mounted the pin in my lathe (a recent acquisition - I'm no expert by any means!), and smoothed the groove using #320 wetanddry, followed by some metal polish / cloth giving a nice shiny surface (even using an eyepiece). I then re-assembled the focuser and .... voila !!! The false detent problem was gone
Just as I was feeling really pleased with myself I noticed another problem: the focuser drawtube was not square with the focuser body !!??!! I soon found that one of the 4 roller bearings was set crooked and had even badly scored the drawtube where the edge of the roller had dug in. There is an M3 set screw that hlods the roller bearing against an aluminium pad that was not screwed in square to the hole (cross threaded). After a few attempts, I managed to screw it in square; but the screw only engages the hole by about 3mm, hardly enough to ensure a good structural fitting (a screw should engage by 2 x D in soft metal like aluminium - so 6mm would be ideal in this case). I will replace all 4 screws with longer ones later.
The end result is that it seems to be performing well, and the fine focus now has a smooth feel to it
David,
Getting back to your 12.5 - have you received the focuser and corrector yet? From what I understand, the ASA Corrector has a true OD of 3 inches and the focuser also has true ID of 3 inches. Just wondering if you had to do something special to make it fit.
James
David,
Getting back to your 12.5 - have you received the focuser and corrector yet? From what I understand, the ASA Corrector has a true OD of 3 inches and the focuser also has true ID of 3 inches. Just wondering if you had to do something special to make it fit.
James
Well, my understanding is that focuser drawtubes, etc are machined with a small clearance to allow for true sized male fittings. However, I don't like nasty surprises so I asked Optec before I ordered the focuser and here is part of their reply:
"the I.D. of the TCF-S3 drawtube is 3.005 +0.001/-0.002 so we hold extremely tight tolerences. Assuming the corrector is designed to fit into a 3-inch bore, the fit should be snug."
So, about 1/8mm diameter clearance. I have had the Keller corrector and focuser for a while now, and yes I did try the Keller in the focuser and it does fit
Thanks David
In case anyone is considering something like this, I already received confirmation that it does NOT fit in the 3.0" Feather Touch.
James
That's interesting - how was it confirmed? You made me paranoid so I just went down to the observatory and measured the OD of the Keller corrector and the ID of the focuser with my new Digital Caliper (accurate to 0.03mm). The ID of the drawtube was 76.27mm and the OD of the Keller corrector was 76.18mm, leaving about 0.09mm clearance (close to the 0.125mm specified value difference given the caliper accuracy, and also the fact that an OD reading may be too large and ID too small if you are not very careful taking the reading).
The 76.18mm (+/-0.03) is virtually spot on for the theoretical 76.2mm (3") diameter of the Keller corrector. I would be very surprised if FeatherTouch did not allow some clearance. Of course, it could be a QA issue on either side.
One other thought - I have found with my 2" Moonlite Crayford that because I have the pressure adjusted quite high (to try to cope with the weight of my STL-11000M), that my 2" laser collimator will scrape the bottom of the draw tube when I push it in. When I ease off the pressure, it goes in smoothly. So the extra pressure will tend to deform the drawtube very slightly. I think a 3" Feathertouch is a rack and pinion though? So may not be the problem here as it doesn't rely on high pressure on the draw tube.
I emailed Starlight Instruments to check. They said it won't fit in the 3.0" focuser, you have to get the 3.5" version with an adaptor.
The adaptor looks a bit small to me, the TCF-S3 sounds like a better solution.
I have completed the tube counterweight assembly that will be used on the scope to provide fine balance when changing camera / eyepiece, etc. It is based on a similar principle to my 8" prototype, whereby you can slide the counterweight along the axis of the tube, but also slide the whole assembly around the tube circumference (max 120 degrees) to provide lateral balance (eg. if the guidescope, finderscope, RigRunner, etc results in excess weight on one side of the tube). To facilitate rotating the whole assembly around 120 degrees of the tube circumference I have added a clutch on each end (a small teflon "washer" that I turned that is tensioned against the CF rail via a grub screw) - this prevents the whole thing acidentally flopping when the lock screws ( two at each end) are released. Anyway, see attached pictures to get an idea of what I am talking about.
I have completed the majority of the work on the tube saddles. They are basically carbon fibre / epoxy castings that I layed up with dozens of layers of CF. I also tried something new: I made some CF plate and cut it to the shape of the tube saddles and laid it in during the wet lay-up in the mold.
Anyway, I've provided a brief sequence of images below to give everyone an idea of the process involved:
Beginning to rough cut out the mold (made out of mdf board).
Mold rough out completed (using drills and jigsaw)
Mold finished using wood router (seen here mounted on my home made circle cutting jig).
Vacuum bagging the CF plate using compressor and vacuum transducer. Note that this is simply some square flat CF plate; it's not the mold / casting. The plate was later cut to shape with a jigsaw and used as a couple of "layers" in the wet layup in the mold for each saddle.
It was a cold day so I created a warm environment to cure the plate quickly using a home made, low temp oven (ie. hair dryer and plastic cover )
The first saddle cured in the mold and sanded flat using a belt sander.
The first saddle demolded with the second saddle in the mold ready to be sanded prior to demolding.
I have also cut the slots in the ends since taking these images (to accept the strap ends) and made some bronze cylindrical retainers. I have also picked up the stainless straps that will be used to hold the tube on the saddles - I'll take some more pictures next weekend.
OK, a few more pics this week. I have done a bit more on the tube saddles and made some progress on the secondary spider / hub support. The design for the latter will be based on the offset vanes outlined in Texereau's book as well as a few ideas of my own.
The images are as follows:
I have drilled, tapped and glued in the threaded stainless inserts (used to attach the saddles to the versa-plate) and cut the slots in the saddle ends. I have also sourced the stainless straps from a local supplier. You can also see a length of 12mm stainless rod that I will cut up and thread and pin to the straps, as well as some of the bronze retainers that will provide a method of tensioning the strap to the saddles.
I used some 25mm wooden handle to wrap dozens of layers of fibreglass & carbon cloth to form a cylinder, which will be cut at 90/45 degrees to be used as the spider hub. Here you can see the start of cutting using an old GMC compund saw.
Initial cutting completed - you can see the final part sitting on the saw platform, as well as leftover parts of the layup.
Here you can see the spider hub cleaned up with the wood dowel removed, next to the leftover section. If I muck it up I can use the leftover section to make another one
Here you can see some of the components that will be used to fabricate the spider assembly:
- the 45 degree hub
- 2 elliptical carbon fibre plates that I made (approx 6.7mm thick each). M2 will be attached to one by three dobs of silicone at the appropriate positions, and the other plate will be glued to the 45degree section of the spider hub with stainless threaded inserts to attach to an aluminium collar.
- 32.8mm titanium tube that will be used as the central rotation / translation axis of the assembly. It has an aluminium collar that I will shrink fit on one end that will be used to attach to the 45 degree hub. Note that this particular collar won't be used as I made a couple of mistakes (one was that I bored it out too much which ruined the shrink fit).
- the micro-collimator (on the titanium tube) that I made from carbon fibre / epoxy resin that will be used to provide rotational adjustment (note the 2 push screws with black thumb knobs) as well as positional control (up/down the tube).
- Not shown yet are the stainless steel vanes, 2 aluminium ends that will fit over the titanium tube (like the micro-collimator) and some 8mm titanium rod that will provide anchors for the vanes.
David I am absolutely amazed! First I have never seen carbon laid up like that before and now my wheels are turning. 0.o I have also never seen a sliding weight used like that either. My hats off to you for being unique. My 12.5" truss isn't quite done yet and now I may have to go back over it to see if I can use your methods in my design. Looking forward to following the build....
Thanks Greg. Yes I've developed some of my own ideas and techniques with fibreglass / carbon fibre - which is half the fun of ATM'ing! The thing I love about fibreglass is that you can make complex parts that would cost a lot of money to make out of metal.
I made a little more progress this weekend:
Made the stainless threaded inserts for the 45Degree fibregalss hub (used to attach it to the aluminium collar with M4 screws) and glued them in.
Completed the interference fit of the aluminium collar to the titanium tube.
FYI - an interference fit is made when one part has a smaller diameter opening than the part that it fits over. In my case, the ID of the aluminium collar was 0.075mm smaller than the OD of the titanium tube. I heated the collar to 250C in our oven (expanding it by approx 0.18mm) and put the tube in the freezer for a while. I then clamped the tube upright in a vice and (quickly) took the collar from the oven and placed it on the end of the tube (see image below). After a quick wiggle the collar fell and fully engaged the tube (I had a lip on the collar so it didn't slide all the way down!). Once it cooled the collar tightened very firmly around the tube.
Another few firsts with my lathe (still learning). The attached image shows some counterweights and attachment rods that I've made for the telescope tube. Although I have a dual sliding weight (see previous posts) that will allow variable adjustment, I realised that may not be enough to cope with a large change to the balance (eg. replacing the STL-11000 with an eyepiece for visual use).
The main points:
The threaded rods will be able to be screwed (M8 end) into some threaded stainless inserts in the tube wall.
The counterweights are centrally drilled and tapped to screw on to the M12 ends of the threaded rods (not easy tapping M12 sized holes in stainless steel!!)
The counterweights can then be snugged up on a small flat platform on the tube wall which will make them more or less self-supporting (ie. negligible cantilevering).
I made some knurled thumb nuts to lock the counterweights in place. My very first attempt at knurling - was supposed to be a diamond knurl but ended up with more of a spiral pattern as I didn't get one of the knurling wheels to "bed in" very well. Oh well, still looks OK.
The only thing left to do is drill a small hole in the M12 end of each rod that can be used to tighten it on to the tube (by inserting a small screwdriver or similar through the hole to get some leverage).
You can see a rod / weight / knurled nut screwed together in the image to give an idea of what I am talking about. I went a bit overboard with the weights (3 x 600g and 2 x 300g), but I figured better to over do it a bit rather than get caught short later!
Those weights look real good, how much more do u think u spent on making them vs buying off the shelf ones ?
But nothing beats doing it yourself hey
Well, actually much cheaper to make them myself; basically just the raw material cost of the stainless steel and threaded rod - I haven't actually paid for it all yet but I would guess approx $60. Compared to over $120 for less weights (and not stainless) from a telescope/accessories distributor. ... and yes, nothing beats doing it yourself if possible!!
Some more progress on the secondary assembly this weekend. I'm waiting on a couple of parts that I am having made up (ie. vanes and vane hub plates), but I have assembled the parts I have ready to give an idea of how it will work. I took a long time thinking about the whole
secondary assembly and basically designed it from the ground up. Having used Newtonians in the past (including the one I made recently), I wanted to avoid the problems inherent in typical secondary mirror spider/holders; my main objection being that they are an exercise in
frustration when it comes to collimation!
I wanted a secondary mirror support system that holds the mirror firmly in place with no tendency to move over time, and also provides simple and precise collimation.
In thinking through the design I came up with the following basic criteria:
Since my secondary mirror has a minor axis of 100mm it is thus similar in size to a small Newtonian's main mirror, and that I should keep this in mind in order to break free of the secondary hub paradigm; viz. to avoid collimation points that act on the base of an extended 45 deg hub, where by adjusting the mirror angle it also moves the mirror laterally as well.
I realised that 3 point tip/tilt positioning is not even necessary provided that fine adjustment in rotation is provided. You only require tip/tilt in one axis (being parallel to the minor axis of the secondary mirror); this together with fine control in rotation provides full collimatable functionality.
I wanted to incorporate an offset vane configuration (described in Texereau's "How to Make a Telescope") as it provides maximum stiffness (including resistance to rotation).
I wanted to design out any residual forces that act to pull the mirror out of collimation; ie. any cantilevering forces that the mirror may exert on the assembly and rotational forces (less important) due to the off-centering of the mirror / hub design.
You can see (in the first image) a closeup of my implementation of the tip/tilt supports. Note the assembly is in mirror face down
orientation (though the mirror is not attached; it will be siliconed on to the bottom plate in the image):
There are two non-adjustable support points (seen on the left). Note that contact is made in a line between the two points to avoid stress on the bottom plate in particular.
The adjustment in the third support (on the right) is provided by the nut in combination with a wave spring washer between the plates. This provides approximately +/-2mm of adjustment. This should be more than adequate provided the vane holes, etc are set accurately in place; if more adjustment is needed this can be accomplished by inserting extra washers.
Once collimation is achieved the 2 fixed support point nuts can be tightened, which makes the whole thing extremely rigid.
The other two images show the assembly (less mirror and vanes / vane hubs). Also shown are the two titanium 8mm rods that will be
supported by the vane hubs and hold the vanes:
One of the images shows the end of the 8mm rod held in the push/push screws of the micro-collimator. This will provide fine adjustment in rotation.
Near the end of the main tube you can see a horseshoe shaped stainless steel counterweight that I made yesterday. This will bring the centre of mass of the adjustable sub-assembly (ie. mirror / plates / tube / micro-collimator) in line with the vanes, thus avoiding any tendency of the mirror to tilt due to cantilevering forces. Further, the horseshoe shape will be oriented to negate the weight bias on one side and so avoid any tendency to rotate.
A few pictures of my progress to date. The secondary assembly is now complete (apart from painting and siliconing the secondary mirror on), as are the tube saddles / straps.
First two images show the empty tube mounted with the completed saddles / straps. Note that the white plastic coating on the straps is just protective covering for the stainless steel face side. It will be stripped off.
The next image shows the inside tube view of the focuser platform. The tube has been reinforced at the attachment points for the focuser, with three holes drilled and tapped for the stainless inserts to be glued (One is shown fully screwed in and glued, another partially screwed in and the third not started.
The next image shows the jig I made to bend each of the vane pairs for the secondary support.
The next 2 images show the assembled secondary mounted in the jig that I used to line up and drill the holes in the tube for the offset vanes (the holes for the vanes are offset since each vane does not line up with the geometric centre of the tube). Note the lugs pinned to the end of each vane; they are designed to maintain absolute stiffness - there are three pins in each vane which prevents any tendency to "swivel" as in many commercial designs.
The last two images show the whole secondary assembly mounted in the tube (the last image shows the geometric offset of the vanes). It feels extremely stiff even when the screws are just finger tight.
Can anyone recommend the best silicone to attach the secondary mirror to the carbon fibre plate? I'm thinking maybe "Roof and Gutter" (neutral cure and good adhesion to most surfaces)?
Last edited by David Fitz-Henr; 03-10-2011 at 06:27 PM.
Nice job on the secondary /spider , David. I would recommend an elliptical ring of silicon blobs at the 70% radius zone, and don't let the silicon cure any thinner than 3mm so use some spacers while it is drying. Silicon has an incredible amount of pull when it cures and is quite capable of warping a secondary mirror in a big way.
There actually isn't a lot of definitive literature on the best spacing / thickness of the silicone blobs to support the secondary. It is more complicated than a primary mirror because a) it is elliptical, and b) the 45 degree cut of the glass means that there is more mass at one end of the elliptical mirror surface than the other. So, a support point (on the back surface) at one end of the ellipse would support more weight than the other end. I don't think PLOP will model this, but I saw reference to another FEM tool that might (though I haven't tried it). I think the saving grace for many secondary mirrors is the fact that they are smaller than primary mirrors and tend to be full thickness as well.
Here is an interesting website that discusses some aspects of mounting a secondary mirror: http://www.cruxis.com/scope/scope110...darymirror.htm
Note however, that this site shows a cellular secondary which doesn't have the classic 45deg cut.
My approach is along the following lines:
To have just three silicone support blobs - any more can introduce additional stresses on the mirror.
Each silicone blob to be no more than 20mm diameter (my secondary has a minor axis of 100mm).
From the Cruxis telescope site (based on his FEA), "Another important conclusion is that it's best to mount the mirror with the edge of the single support closest to the eyepiece".
This seems to make sense intuitively for a classic 45 degree cut in that this mounting point "overhangs" the actual mirror surface, and so the other 2 points (which will necessarily be closer together at the other end of the ellipse) will support a greater load (since this end of the ellipse on the back surface "underhangs" the mirror surface. I'm not really sure why this would be optimal for the cruxis secondary however.
However, I'm not sure which silicone to use?? Websites seem to only quote RTV silicone, but my understanding is that all silicones sold in the local hardware are RTV (Room Temperature Vulcanising)? I had thought a neutral cure would be best, but I see on the Protostar website they mention the acetic acid catalytic type? Dream Telescopes recommend a 1mm thickness, so I would assume that the silicone should have a low modulus (elasticity). Does anyone have an opinion here?
Last edited by David Fitz-Henr; 05-10-2011 at 06:30 PM.
Reason: Remove text diagram as it didn't work (spaces removed automatically)
Looks like you have done some research! My original 8" binocular used silicone squeezed fairly thin to mount all my optics and I eventually found them all to have been seriously warped by the thinly spread silicone causing quite a bit of astigmatism , which is why I suggested a minimum of 3mm.
).
. It had to be something to do with the pin I reasoned, as the false "detent" matched the rotation cycle perfectly. The ball bearings are held against a groove around the circumference of the pin near one end, so I inspected the groove with an eyepiece, and found a very slight mark/depression within the groove (see the picture to get an idea of how small I am talking about - probably of the order of 10-20 microns deep!!!). This had to be it, as combined with the high mechanical pressure applied, this would mimick a detent I reasoned. I mounted the pin in my lathe (a recent acquisition - I'm no expert by any means!), and smoothed the groove using #320 wetanddry, followed by some metal polish / cloth giving a nice shiny surface (even using an eyepiece). I then re-assembled the focuser and .... voila !!! The false detent problem was gone 
)
