Search the Community
Showing results for tags 'effective focal length'.
Reverse-Engineering the Skymax 180 What follows should be of interest to anyone who wants to know any of the following 3 things about this scope. 1. What it looks like inside, its design, how it works, and how to dismantle it and re-assemble it safely (yes I’ve seen that video of someone disastrously taking apart a Skymax 150!). 2. Its real dimensions, including all the major components of the scope including internal mirror separation/s; 3. Its “key numbers” such as effective focal length for various amounts of back-focus or mirror separation, to allow more confident estimation of magnification, fields of view, exit pupil sizes for various configurations. When I went to try to find these details, I couldn’t find anything reliable online. It started by me noticing that the Moon, always very close to 30 arcminutes across, was filling more of the field of view than the official spec of the scope suggested it should. The scope’s advertised focal length is 2700mm, but it seemed that in the set-up I was using it was behaving more like a 3000mm. That set-up was an external Baader Diamond Steeltrak, adding quite a lot to the back-focus compared to the supplied visual back. I did a little searching around, and quickly came across the formula for the effective focal length of a catadioptric: namely EFL = f1.f2/(f1-f2-s) where “s” is the separation between the mirrors. Seeing as the focus-mechanism relies specifically on changing this separation, I had all the excuse I needed to do something I’ve come to enjoy doing: survey the scope and/or take it apart to see how it works. At the same time I also did a focuser-axis-vs-primary-optical-axis test, and discovered that the two axes differed by around 8 arc-minutes; enough to investigate trying to fix it. As it turned out, this was not a problem, it was a design feature (more later)! Anyway, to investigate, I was going to have to see how the primary mirror was affixed to the central tubes and how if at all it might be adjusted, and to do that I was definitely going to have to take the scope apart. Unfortunately as stated I couldn’t find anything online either about how properly to take it apart, in particular the mirror, focusing and collimating end; and neither could I find anything I trusted about its various dimensions. I did find one article whose author calculated back-focus and EFL relationships for a Skymax 180 of almost identical age to mine, and who even went as far as to run its dimensions such as he had them (including corrector-plate refractive index!) through a ray-tracing program; but many of the dimension inputs he used I think must have been assumed or guessed: when I measured and re-measured them I found them very different from his. And the formulae are extremely sensitive to those dimensions being accurate. Disassembly, Adjustments and How It Works The rear assembly comes apart into 2-3 main pieces: the main rear cell, i.e. the cast-metal “back end lump” of the scope you can see from the outside, out of the back of which is poking the focus knob and the visual back; and the inner mechanism comprising the 2 nested baffle tubes, the primary mirror, the focus-mechanism and the collimation-plate. The primary mirror, outer baffle tube and the focus-rod receiving plate are all effectively one bonded-together piece. Although the mirror sits against a flange in the outer baffle tube, it’s heavily bonded on so trying to address any baffle-tube orientation problems is nigh impossible. The inner baffle tube, which sits inside the outer one, becomes at its bottom a stamped flat-metal plate with three threaded holes and a cut-out recess to allow the focusing-rod through. The three holes correspond to the main collimation bolts at the back of the scope, and they are the only things supporting the primary mirror assembly inside the OTA. In other words, those 3 collimation bolts hold the entire mirror and baffle-tube assembly inside the OTA space, and adjusting them points it around inside the OTA. What this means is that the visual back on the OTA IS NOT DIRECTLY CONNECTED TO THE BAFFLE TUBE. There is a gap, and potentially an axis-kink. See my drawing, comparing the Skymax to my Intes M603: What it also means is that blindly removing those 3 collimation bolts as a first step will cause the mirror assembly and baffle tubes to, at best, hang off the focus-rod inside, and at worst, if you’ve already removed the focus receiving bearing from the rear cell, the whole baffle and mirror will be set free to collapse onto the corrector plate and secondary! Therefore: if you really wish to dis-assemble, follow my instructions below! This arrangement is interesting. In my Intes, the visual back is part of the baffle tube which extends all the way through the rear cell: eyepieces, diagonals and cameras attach directly to the baffle tube. In the Skymax here, the visual back is part of the rear housing of the OTA, only attached to the baffle tube in a subsidiary fashion via the collimation bolts, and the whole “viewing tunnel” is effectively kinked at the rear of the scope. So collimation on the Skymax is mostly getting the baffle tube to line up with the visual back and eyepiece axis. That certainly helped explain why my eyepiece axis and mirror axes were “out” when I measured them. But if there’s misalignment of the corrector plate, or if the secondary mirror-spot is out of place, then I guess there’s little you can do about it. The native focus mechanism is very simple, far simpler than that of the Intes. It comprises a threaded rod held at one end by a bearing in the main scope back cell, and at the other end attached to a rigid metal plate bonded behind the primary mirror. Turning the knob causes the plate (and hence the mirror and outer baffle tube) to move up and down the tube at a rate of about 0.8mm per knob-turn. Obviously, there must be a tiny amount of clearance between the two tubes, and it’s this clearance that causes the dreaded “mirror slop” when the focus knob is engaged. Next came the front cell which contains the 18-19mm thick corrector lens, and appears to have coatings (green and pink reflections each evident). On the inside surface of the corrector lens is the secondary mirror, an aluminized spot surrounded by a cup-shaped black plastic baffle skirt. This skirt is glued to the secondary mirror, but in my case NOT CENTRALLY! I removed it, thinking in the process that I was reducing the Central Obstruction as well, but soon realised that because the primary mirror hole and retaining-ring were themselves much wider than the widest part of the secondary baffle-skirt, there was no point. Nonetheless having removed it, I cleaned off the residue, re-centred and re-attached it. With the baffle-skirt in place, the remaining exposed secondary mirror is 36.5mm diameter. The main tube itself was a loose fit: larger than the fitting-flange on the front cell, and smaller than the flange on the rear cell in each case by more than I was entirely happy with (it fits inside the lip of the rear cell and outside at the front cell). As such, the tube had to deform in each case slightly when doing up the 4 retaining screws at each end. I plan at some stage to get a carbon tube with a better fit. The dovetail is bolted to the tube, not to the much-more-solid cells, and the tube is only held on to the chunky front and rear cells by those 4 screws each end. Rings for this scope would be a good addition, or a longer dovetail to attach directly to the main cells. What follows is a step-by-step guide to dis-assembling the scope. It describes taking everything apart except the components of the front cell and the corrector plate. Before you start, one very important thing is worth mentioning. As the pictures above show, the baffle-tube arrangement comprises two tubes, an inner one and an outer one. The primary mirror is bonded to the outer one. The outer one is free to slide up and down the inner tube (there is lubrication between them). The main thing which fixes the position of the outer tube, and prevents it from sliding anywhere it likes, is a threaded rod attached to a bulkhead behind the primary, the other end of which pokes through the back of the OTA. The outer tube is held in place by the focus knob, which when turned, activates the thread and moves the outer baffle-tube, and hence the primary mirror, up and down the inner one. Here’s the important bit: if you remove that focus knob, and the OTA is sitting “corrector down” (as instructed in the step 6-8 below) step 8 has you seemingly removing the only thing preventing the outer tube from sliding all the way down, off the inner tube, and onto the corrector plate inside. There is one design feature designed to prevent that from happening: a rubber O-ring sat in a groove towards the end of the inner baffle tube, designed to stop the outer tube from moving any further. BEFORE YOU START, CHECK THAT THAT O-RING IS STILL THERE. Rubber does perish eventually, and if you cannot see that O-ring, I leave it up to you to think of a different way to remove the focus knob to prevent the mishap. Step by Step Dismantling 1. Attach a visual back that extends beyond the focus knob. You’ll be using it to place/balance the rear assembly vertically on a flat surface later. 2. Remove rubber focus knob (it simply pulls straight off). 3. Rotate the protruding brass cylinder clockwise, to gradually expose the threaded rod. 4. Unscrew and set aside the small central Philips screw (and possibly small washer) at the top of the threaded rod (it may be a C-clip instead, in which case remove that). 5. Remove main telescope front plastic “lens cap”. 6. Place scope “front down” on a level surface. 7. Partially, maybe ½ a turn, unscrew the three main (bigger) collimation bolts at the back. BUT ONLY UNTIL THEY LOSE THEIR TIGHTNESS – DO NOT COMPLETELY UNSCREW THEM YET! THE MIRROR IS ESSENTIALLY HANGING OFF THESE SCREWS INSIDE THE OTA! Do not touch the smaller recessed “locking screws” – keeping them in place allows you to restore the position of the primary at roughly the same orientation when it comes to re-assembly. 8. Unscrew the 3 screws on the flange around the focus-knob, and set them and the flange-plate aside. 9. Unscrew, by hand, and set aside the exposed brass focusing assembly (ACW) all the way off (perhaps 20+ turns!). Also notice and remove a black rubber washer underneath the assembly. Be careful of the evil black grease! 10. The threaded rod, covered in black grease, will now be poking up through the hole in the back plate. 11. Unscrew and set aside the 4 screws on the side of the OTA holding the rear cell to the main OTA around its outside. From here you need to be very careful with your movements to avoid things getting knocked and toppling over. 12. Lift the whole rear assembly out of the main tube and carefully place it on the level surface “focuser down” (i.e. collimation screws and visual back at the bottom). 13. With a small/short hex socket-insert, with your fingers from underneath, carefully unscrew/twiddle off all the way the 3 main collimation bolts that you loosened earlier. Once done, the primary mirror assembly and baffle tubes are now only resting on the OTA’s rear cell. 14. Whilst supporting the rear cell on your flat surface, and holding the baffle tube, gently lift the baffle tube (and primary mirror and focus-rod) away from the main rear cell. Be careful not to lose 3 hidden fat little washers between the plate at the bottom of the baffle-tube and the rear cell: they can stick to the underside of the plate and fall off later if you don’t pay notice them. 15. Remove the O ring around and near the top of the inner baffle tube. This O ring prevents the outer tube from sliding all the way off if somehow the focus-rod isn’t holding it. It’s tricky to remove with all the grease lubricating the two tubes: try not to damage it. 16. You can now separate the two tubes by pulling the outer tube off the inner: the outer tube with its mirror, focus-plate and (evil-black-greased) focus rod; and the inner, with its flat plate at the bottom. 17. Further dis-assembly of the primary mirror and its components is not possible, as you will see that the primary mirror is extremely heavily bonded in place on its tube. Re-Assembly 1. Place the rear section on the hard surface, resting on the Visual Back 2. Place the 3 fat doughnut black rubber washers on the 3 larger collimation-holes (they act as crude tensioning-springs for the collimation-bolts) 3. Carefully place the collimation-plate/inner baffle-tube so that its 3 holes match the main collimation holes, and such that the focus-knob cut-out matches the focus-hole in the rear cell 4. From underneath, using a short suitable hex-insert, screw in the 3 main collimation bolts most of the way, but do not tighten 5. If necessary, re-grease the lower part of the baffle tube where the outer will slide over it 6. Gently and carefully lower the primary mirror / outer baffle-tube onto the inner one 7. Re-fit the rubber O-ring into its slot 8. Remove excess grease from the upper end of the baffle tubes 9. Bring the 2 halves of the scope back together again: carefully lower the rear assembly back into the main tube, and re-attach the 4 screws. The scope should now be on its front, with the visual back “up”. 10. The black-grease threaded rod should now be poking up through the focuser-hole. Replace the black flat rubber washer into that recess 11. Screw on the focuser assembly, “thicker bit” first, clockwise onto the threaded rod until it’s all the way into the recess, and a couple of turns more. 12. Screw on the small locking screw onto the end of the threaded rod (or replace the C-clip if that’s what it is) 13. Replace the flange-plate and secure it with its 3 screws 14. Push the rubber knob back on 15. Screw the (larger) collimation-bolts back until they are reasonably firm: having not touched the smaller locking-screws, this last action returns the primary mirror assembly to close to the orientation it was before you started. Dimensions and Measurements A. 45.4 mm: Front Rim to centre of Corrector Plate B. 18.6 mm: Thickness of Corrector at Centre C. 42.5 mm: Depth of Secondary Baffle Skirt D. 421.5 mm: Front Rim to Front Rim of Rear Cell E. 357.5 +- Nx0.787 mm: Secondary Mirror to Centre of Primary Mirror where N = no. full turns ACW from where supplied visual back & 2” diagonal come to focus (more back-focus => smaller separation => bigger focal length) F. 83 mm: Front of Rear Cell to Flat Back of Rear Cell (not including Visual Back attachment!) G. 37 mm: Exposed Diameter of Secondary Mirror H. 58 mm: Secondary Baffle Skirt Width (at wide end) I. 42.5 mm: Depth of Secondary Skirt J. 63 mm: Diameter of Primary Mirror Retaining Ring (i.e. wider than secondary skirt!) K. 200 mm: Primary Mirror Diameter (i.e. oversized) Other quantities not shown on diagram: F1: 472 mm: Primary Mirror Focal Length (measured as half the centre of curvature, itself measured from its reflecting a point source F2: 127.92 mm implied Secondary Mirror Focal length (to force formula to give 2700 mm with supplied back and diagonal and all other measured dimensions. Very difficult to measure). also: - 0.787 mm pitch of focus-knob (movement in primary for one full turn of native focus-knob) - 665.3 mm Circumference of lip of front cell (=> diameter 211.8 mm) - 670 mm Circumference of inside of OTA tube at front (=> diameter 213.2 mm) - 1.60 mm OTA tube thickness - 682 mm circumference of INSIDE lip of rear cell (=> diameter 217.1 mm) - 681 mm Circumference of outside of OTA tube at rear (=> diameter 216.8 mm) The way the main (steel) tube fits on to the front and rear cells is worth taking note, if you plan to fit a carbon tube, for instance. The tube slots OVER the cell at the front, but INSIDE the cell at the back. The difference between the two meeting-face diameters is 217.1mm less 211.8mm, i.e. 5.3mm. Which means that should you wish to upgrade to a, say, carbon tube, it would need to have at most 2.65mm wall-thickness. The steel thickness on this scope is 1.6mm, which is accommodated by the fact that it’s flexible, and on this scope at least needs to flex to fit. The native focus mechanism is, as shown, a knob on the back which moves the primary mirror up and down the tube. The secondary mirror is fixed in place as an aluminized spot on the back of the front corrector lens. Focusing via that knob changes the separation between the mirrors, changing the focus-point and back-focus and changing the system’s Effective Focal Length, according to the formula EFL = - f1.f2/(f1 – f2 – s), s being the mirror separation. If I could simply find out what the individual focal lengths of the two mirrors were, and what the mirror separation was for given positions of the focus knob for a given back-focus amounts, I could calculate what focal length a given arrangement engenders, and therefore construct a more accurate mag / exitpupil / FoV ready-reckoner to stick to the side of the scope. I started by roughly estimating what these numbers might be just by “eyeballing” the scope, doing some crude measuring and putting together a simple spreadsheet. I hoped that would be close enough and I wouldn’t have to dis-assemble. For instance f1, based on reflecting a point source back to itself from the mirror seemed about 450mm, f2 around 90mm and judging by where the main mirror looked positioned, the mirror separation looked something like 370mm. Then I noticed something about the EFL formula. f1 x f2 is going to be a reasonably large number, in this case 40,500. f1-f2-s is going to be quite a small number, here -10, suggesting an estimated EFL of 4050mm. Hmmm. Big number divided by small number is going to be very dependent on the small number, especially if that small number is a difference in measurements. It doesn’t take much leeway in those numbers for that denominator to be, say, 0mm for example and the calculated EFL to become infinite. Or even negative! Clearly, more precise dimensions rule here. “Rough estimates” weren’t going to cut it: I could come up with whatever results I wanted just by slightly varying the key dimensions. I was going to have to make accurate measurements. Two of these accurate measurements initially presented a challenge: the precise thickness of the corrector lens (the mirrored spot is inside the tube and at the centre of a highly curved and rather thick glass plate); and the precise position of the centre of the primary mirror (it’s recessed into the rear cell of the scope, has a big hole where its centre should be and you can’t see the thread which moves it). I needed to be ingenious about each of these. My digital micrometer saved the day. The back end of it can be used as a depth gauge, and using it I was able to accurately estimate the thickness of the corrector plate and, because the primary mirror is recessed into the rear cell, the distance below the rim of the edge of the primary (I also needed to take account of the sagitta of the primary). The distance between the front and rear cells was trivially measured from outside the fully-assembled tube. And finally, using the pitch of the threaded focusing-rod, I was able to determine the mirror separations for any position of the focus-knob. Effective Focal Length and Back-Focus All the above having been done, I was now in a position to estimate EFL for various back-focus settings, and for my various actual set-ups. These estimates are predicated on the assumption that with the supplied diagonal and Visual Back, the focal length is actually as indicated on the scope’s plaque, i.e. 2700mm. The only quantity I couldn’t easily measure, the focal length of the secondary, was the “degree of freedom” I could change that allowed me to “fix” the EFL to 2700 for a certain set-up. At some stage I’ll try to actually measure it, but this will do for now. Below is a chart showing Effective Focal Length of the Skymax180 against the likely range of back-focus behind the OTA: The geometry means that actually the EFL is a linear function of back-focus, which surprised me (see formula below). Incidentally the 440.5 is the distance from the secondary to the back of the OTA. For the Skymax 150, for example, the formulae would be the same but with a different value of “440.5” (I’ll update this post with Skymax 150 values and dimensions when I get back to London). A couple of extra data points that I didn’t include on the chart are those for the “end-stops” of the native focuser travel, i.e. 0 turns ACW and 29 turns ACW: 0 turns => -40mm backfocus (i.e. inside the rear cell!) => 1952mm EFL => 375mm mirror separation 29 turns => 1113mm backfocus => 6200mm EFL => 354mm mirror separation For those who prefer formulae to calculate these things: Effective Focal Length EFL = (BF+440.5).F1/F2 + F1 all quantities in mm; BF measured from rear of OTA Mirror separation s = F1 – F2 + F1.F2/EFL In the use of these formulae for estimating the various quantities, I’ve ignored the effect of the corrector plate. The only exception was the measurement of the Primary’s focal length, which I measured with the corrector plate removed. I don’t think it matters too much, and hopefully is partially compensated-for by my back-solving of the focal length of the Secondary to achieve 2700mm in OEM configuration. (Re-)Collimation I now had knowledge of what was mechanically going on behind the mirror and what the collimation bolts did! As mentioned, those 3 bolts are the only things attaching the primary mirror support assembly to the rest of the scope, and they basically point the mirror around the inside of the tube. The secondary is fixed, the visual back attachment is fixed, the only thing that you can change is the orientation of the primary inside the tube. Thus collimation involves aligning the primary’s axis as well as possible with the visual back’s axis and the centre of the secondary. If either of these are out of place, it’s an exercise in compromise. A popularly suggested method for aligning Maks and SCTs etc is the “hall-of-mirrors” method. But I’ve found it unsatisfactory: it can tell you if you’re reasonably close, but if you aren’t it doesn’t tell you what to adjust to get it right. I’ve found that star-test collimation is much more intuitive and sensitive. Point at something like a 2nd-mag star, ideally Polaris because it stays still, on a night of not too bad seeing. Using at least a 10mm eyepiece, and very slightly defocusing, you'll notice a set of concentric(ish) rings around a small hole with a point in the middle. Very likely though, the doughnut you see will be "squashed" towards one edge. Establish which collimation bolt best corresponds to that squashed position, by putting your hand over one side in front of the scope and seeing where the gap appears in the view: the bolt closest to that gap, or the one opposite, is the one to move first. Adjust that collimation bolt, and the squashiness will either improve or dis-improve. One proviso: as you turn the collimation, the star will move out of view, so have your controller handy so you can keep it in view during the process, to avoid spending 10 minutes re-discovering Polaris at high magnification (been there, done that). Keep going through that process until the ring-pattern is as symmetric as you can make it. I find that it comes right quite suddenly at the end. You could go one step further to do super-fine collimation (I often don't bother) by going to super-higher magnification, getting to best focus, and "symmetricizing" the Airy disc, but to do that you need almost perfect seeing which is rare. Whereas symmetricizing the doughnut can be done on more ordinary nights and gets you very close. Once there, it should hold reasonably well for the future. People call catadioptric collimation a "dark art", and one of the reasons I think is that the various internal designs are often very different, and collimation is doing different things "inside". And they never tell you what's inside, so you are effectively adjusting something by blind trial and error hoping that whatever it is lines up. If you’ve got this far, well done! And thanks. I hope this will prove useful to anybody else with a reason to want to know how this scope and its siblings (Skymax 150, 127 etc) actually work inside. Cheers, Magnus