Jump to content

Stargazers Lounge Uses Cookies

Like most websites, SGL uses cookies in order to deliver a secure, personalised service, to provide social media functions and to analyse our traffic. Continued use of SGL indicates your acceptance of our cookie policy.


  • Content Count

  • Joined

  • Last visited

Everything posted by Ajohn

  1. This one thinks it's best for some one to find out themselves. Many do after a while John -
  2. The need for the shutter depends on the lens - just as it does when a magnifying glass is used to set fire to something. I would be cautious personally. John -
  3. Might be of interest. While browsing ebay I noticed that there are some 1 mod gear hobs available at rather low prices for this sort of thing from China in HSS. I mod is pretty close to 0.125 circular pitch. It's actually 0.1237. I would have thought a bit of lapping would easily take care of that. For metric people there are also some 1.5mm cp ones from the same source but I feel the teeth formed would be a little weak. 3mm would be a lot better. John -
  4. I nearly bought one of his Maksutov designs. He was helping some one not far from B'ham to make one. Obtained the glass etc and also figured the corrector for him. Going on the size I suspect most would have fitted a 2in focuser but not Mr Dall - 1 1/4. One of the interesting aspects was the thickness of the corrector. Designs according to Maksutov tend to have thick ones as well as that can give coma free images. It was up for sale unfinished shortly after Dall died. I thought it was too risky at the price asked and the seller wouldn't accept what I offered. The corrector would have to be coated and the mirror spot added. I also managed to get onto a university site that showed calculations on just how accurately the corrector had to be held - that was the aspect that put me off most. Actually it also caused me to send a mac newton I bought back due to the alignment it arrived in. Russian and it had clearly been thrown together. John -
  5. I just found a free ray tracing program. The actual download links are right at the bottom of the page. http://www.busack-medial.de/download.htm It comes with some example designs and doesn't look too complicated to use. It looks like it will also pre optimise designs but that will probably be more complicated to use. The doc's also contain a story of how optical designs are "stolen" by people at times. I'm managing to run it on Linux via Wine with few oddities so far. One thing it will allow people to see is how awful the off axis performance of Newtonians are especially as the F ratio goes down. John -
  6. F11 80mm is a far more sensible option for the technology these sorts of achromatic scopes use. In real terms the old 3in F15 refractor gives a true indication of where they need to be if the usual magnification range for a given aperture is to be expected. Go to 4in and they need to be longer and so on. I believe there were some 6in designs that could still be F15 but gave stronger colour halo's around stars. Optical systems don't scale up well as any problems scale up as well. The F ratio remains the same so the diffraction spot does as well. Conversely they improve if scaled down. That is why using an aperture stop can improve things. There may also be edge defects but the same effect would be noticeable even if there wasn't. The only fix is more complex designs within limits and the use of different glasses - ED's etc. John -
  7. I think the achromat has been shown the wrong way round Peter. Despite that it will be virtually impossible to align the images from the refractor and the newtonian, or even keep them aligned for that matter. The focal lengths would also have to be EXACTLY the same other wise the image scale would be different. Neglecting these problems light intensity would be increased. Dall came up with some form of low obstruction newtonian but I've never found any details. I understand that the obstruction was so low it was near refractor standards as far as results were concerned. John -
  8. Ajohn

    All Sky Camera

    I thought this might still be about - it's pretty unique https://ia600809.us.archive.org/25/items/FerrantiELineTransistorApplications/Ferranti-ELineTransistorApplications.pdf Of interest to budding analogue fiends who want to use bi polar. It's unusual in that the sums they suggest actually work. Some of the ZTX E-Line parts are also unique most especially the higher powered ones but I have no idea if they are still about. John -
  9. Ajohn

    All Sky Camera

    The matching comparator is the LM393, data sheet here www.solarbotics.net/library/datasheets/LM393.pdf but it looks like the LM2903 is cheaper and the same style part. The op amp variant of that is more expensive than the 358 for some reason. No opp amp prices I can find on these but the same circuits can be used with an op amp. There's no harm in using others providing they are single rail types. With these I would generally aim for currents a hundred uA and up through the networks but they can be used at lower values. That is where they score over bipolar circuitry. Most small signal transistors are designed to run at 2 to 5 ma. A few much lower. John -
  10. For planetary and narrow field Dall Kirkam's are very attractive. Several of the commercial cassegrains are reckonned to be DK's really. They have far more off axis coma than a cassegrain or even a newtonian of the same F ratio but planets don't exactly need wide field views. On axis they are just as good as a newtonian. The 2ndry is spherical so much easier to make and is also a lot less inclined to go out of alignment. The primary is a part formed parabola so less figuring is needed making it a little easier to make a fast mirror. The catch really is that the design depends on the rad of the 2ndry being correct for the design. Some people make a concave one 1st and then use that to check the actual mirror using fringes - never tried it but a green led should probably give decent ones, blue would be a bit more accurate. The mirror main mirror is an ellipse so can nulled at it's 2 focal points (conjugates in Mike's program) I've seen comments that it's so difficult to set the test up that it's quicker to use Foucault. One design I just looked at would put the far point at around 350ins and that was for an F2.8 main mirror - needs a long room. Testing a mirror with a hole in it is reckoned to be tricky so generally people use plaster to stick the core back in after it has been cut out. There is an update on the program but from what I can see it gives the same results http://www.cloudynights.com/index.php?app=core&module=attach&section=attach&attach_id=440930 This will run under wine on Linux if winetricks is used to install the usual plus msfonts and the vbrun stuff. WineHQ show how to do that from the console. Sounds like cassegrain making is more profitable than newts. I'm interested in one around F8 but like I said off axis it wont match a newtonian but then these don't have very good off axis performance anyway. The design I looked at had around 14 arc seconds of coma at the edge of a 1/2 degree field. John -
  11. Wondering what to do with my flat field diffraction limited cassegrain project mirror I've downloaded a small application some one called Mike on Cloudynights wrote for people to rough out the design. Might be of interest to some adventurous people. It's here http://www.cloudynights.com/index.php?app=core&module=attach&section=attach&attach_id=335273 It will do the sums for several types of cassegrains and is really intended to obtain numbers to put into an optical design package such as the free version of Oslo but I feel it's actually pretty accurate in it's own right. It also calculates baffle tube dimensions. I need to ask about those results as they seem a bit big to me - could be that they assume the tubes are flared. The ATM people on cloudynights are very helpful and generally always prepared to answer questions and even check designs. It's also a good place to ask about the various ways of testing mirrors but remember that any supplementary lenses needed have to be precision types really and all have some limitations one way or the other. I think that after spending rather a lot of time on my design that people who make dall kirkams with relatively low focal length magnifications aren't so mad after all but they still need a rather fast primary and have rather a lot of coma compared with the classical cassegrains. On the other hand they can be made with a lot less difficulty. John -
  12. Ajohn

    All Sky Camera

    The sort of thing I would use would be the LM358N, £1.93 for 5 off ebay. There is a data sheet here www.jameco.com/jameco/products/prodds/839367.pdf As OP amps go they aren't spectacular but are suitable for all sorts of things. The LM258N would be a -45 to 85C part but to be honest they make so many of these types I doubt if there is really any difference. There are other variants as well, most manufacturers produce several types. There is usually a matching comparator as well so if some one is unsure of the circuit arrangements it's likely to be shown in the data sheet for those. National Semi often give the biggest range of circuit ideas and OP amps can be used in the same arrangements but will probably switch a mite slower. One thing to be aware of when looking at op amp circuits is differential arrangements. They look great but often require really closely matched resistors to function as they should. For single ended use inverting or none inverted they are fine. If some one wants a true differential amp the best bet is analogue devices, gain will usually be programmed with a single resistor. I have used that sort of thing in data loggers without any problems providing the signal levels are suitable. John -
  13. Ajohn

    All Sky Camera

    I'm pretty sure I would use a single rail op amp or comparator for that sort of thing. Probably an op amp as I'm a cheap skate. The type to look for has pnp transistor inputs. Some should be very cheap as the automotive industry uses buckets of them mostly quad versions but duals will be available as well. They can be so cheap the fact that one is spare hardly matters. Afraid I am not up to date with the current flavour of the month so any number I quote is likely to be more expensive than it needs to be. They are pretty vice free but can miss behave if used in an analogue fashion such as a low drop voltage regulator where there will be other semi conductors in the feed back loop to it's inputs. Max input currents are usually specified so that a series resistor can be used to limit it if the input goes over the rails. That is one factor that makes them popular plus the input range includes ground. If some one is into solving simultaneous equations comparator type 3 resistor feed back networks can be solved by setting one of the resistors to 1. The others will come out as fractions of 1 so can be used to select suitable actual values. For some reason not many people seem to know this and just add the feedback resistor by trial and error and see what happens. There may be resistor ratio tables about on the web. These can help a lot with many resistor network solving problems. John -
  14. I haven't used it for some time but dow corning did 2 grades of clear silicone, sealant and adhesive. The latter is for sticking glass together. It's also probably possible to buy small tubes from aquatic shops - aquarium repair and also for windscreen seal repair from motor factors. The correct stuff has a much stronger smell as it cures than the sealant types. I've also used double sided tape for this - the type that is sold to hold down carpet tiles, it's thicker than the usual type. Even that might not stick all that well to bare wood. It worked for at least 10 years. the scope was sold so pass after that. John -
  15. I wouldn't laugh to hard about Texereau's comments on mount weight and scope size and some of his other comments too such as mirror thickness and accuracy too really. Perhaps Damian should consider using his mirror in a scope at max magnification for viewing say mars and being able to touch and focus the scope or even keep it pointing at it. Most people would quickly find that conditions can mean that the view was hopeless anyway at this sort of size so that mag would have to be dropped way back. Accuracy comes into that as well. Not just the mirror but the entire structure of the scope. He still makes a number of very fair points really but they have to be taken in context. Change to a relatively low mag light bucket and things are different but not as much as some might think especially in relationship to mirror thickness that will hold the sort of accuracy he is after. The other point about accuracy is that light buckets don't really need it unless they are going to be used at max magnification but on the other hand too serious a departure will spoil the view of a number of things. The term diffraction limited has changed a bit too - he really means it. With a 1/10 wave PV and not RMS PV and the sort of obstruction sizes he suggests things are very close to it. His book is also more or less totally aimed at visual work but he does touch on the photographic problems of his day. That area is now totally different but the same principles apply. Making mirrors - nothing has changed really as far as the actually making is concerned. There are numerous methods of testing mirrors but really his comments about ronchi still hold. From memory MB proposes it but follows it with star testing - that ain't as simply as people might think. Some one posted on here ronchi so far and then Foucault. There is also a variation on Ronchi where a supplementary lens is used to make the lines straight when the mirror is parabolic. Then there is the Dall null test and the other with a different name that is based on it but the lens goes in front of both the knife and the source. The lens for the latter one is pretty expensive. Sounds great until people realise just how accurately it has to be positioned in either case. Both can be used with a Ronchi screen as well. Some one has also come up with a "Ronchi screen" that consists of circles that stay round when the mirror is parabolic or something like that. Most of this type of testing has the same problem - things are set to give the correct result with a very specific mirror that has to have an exact focal ratio not what it just happens to turn out to be. Foucault software works on the basis of the best fit parabola it finds when the measurements are taken. Maybe the is a match the ronchi pattern set up where the screen is just moved a precise amount either side of the ROC and software analyses the image and it comes up with the best fit parabola. If so that would be a bit different but there would still be a need to know exactly where the ROC was. The pro's often use supplementary optics in what is essentially a Foucault test but the optics cause the mirror to null out when it has the correct shape, no shadow reading - eg Hubble . They usually use several lenses in the tester though not just the usual one used by amateurs. They have also use the Hartmann test sometimes just as a final check. It's probably possible to do that one with a DSLR. Foucault - easy to make the gear that is needed and no real drawbacks that can't be circumvented. The usual complaint about Texereau is not taking his suggestions on scope size etc in context and also the sums used to work out the mirror profile. I posted links to several packages that will do the sums. Another comment is sometimes pre Dobson but in real terms that has nothing to do with mirror making His comments about size and usability in typical urban situations are not far off the truth as far as resolution goes. 8in F6 was a bold step in his day. The usual was plate glass 6in F8 and Rubbish instructions as well. No one in the USA had even heard of Plossl's until his book was translated. It's a pity good ones aren't easy to come by and that they had to go super etc. I understand good ones are expensive to make and have little eye relief at shorter focal lengths. If anyone does it don't make the source too bright - after several hours of minor figuring and testing I looked in the mirror and found one pupil was the size of pin [removed word] and the other as big as it could be. It took a while to get back to normal. Hope Damian doesn't finish up like this
  16. There are toooooooooo many video's on youtube about mirror testing and making the bits for that and even the mirror. I think there are a few purely on the Foucault test too. They should aid understanding. There are some good ones. There should also be examples of fixed source and the other type of tester. I can't write a book on here. Texereau wrote 2 books. How To Make A Telescope and the 2nd edition of the same book which as far as this thread goes doesn't add anything useful. The book is fairly old now but explains Foucault testing and mirror making better than any other source I am aware of. There was a copy of the 1st edition on the internet archive. Like many famous books it's been bought up by some one "William Bell" I think and so now any ebook copies are removed. It mentions Ronchi and some other tests but doesn't really rate it as an accurate test - as some one on here has already mentioned perhaps the best option for making a very accurate mirror is Ronchi followed by Foucault. There are other tests as well that use additional optics but like most things there are catches. Perhaps the easiest way of getting to grips with Foucault testing is to do it once the basic idea is firmly in the head. It sounds a lot worse than it is in practice. This page and any links off it might help but I don't know of any web source detailing fringes ahead of the knife edge http://www.telescope-optics.net/foucault_test.htm I've come across another twist on Foucault but it's past my bed time so will add it later. John -
  17. Texereau shows a demo of a ronchi using 5 lp/mm but only 4 lines across the mirror using a 40um slit. From the photo diffraction doesn't look to be much of a problem, it's a 125mm diameter with a 2m rad and a turned down edge. The faults are pretty evident including surface finish. This area is a bit of a pass for me. I couldn't get my hands on one easily on my 1st mirror so knocked one up using fine wire but wasn't impressed as his method actually measures the problem. Damian could use the same method. The knife would be down below the led but that doesn't matter. Just fasten the knife blade 1/2 over the LED, find something shorter to form the other half of the slit and adjust it by eye with the led on - just look for a narrow parallel slit. Put that some way in from the ROC of the mirror and fringes will be seen. These need to be straight too but the advantage is that the first fringe will be 3/4 of a wave from the knife edge. He also uses this to set the slit width. A very fine slit will cover half of the mirror with fringes and this will also give strong diffraction effects. If the slit is widened so that only 2 or 3 are seen the test can still be accurate and the diffraction effects will be minimised. You hope. Adjusting it isn't as bad as people might think. The radius of curvature can be measured roughly with a torch or led etc. Wet or oiled when only fine ground. Wave the light around at the side of your eye and look for the reflection back from the mirror. Probably best to start too close. As this moved back the direction the reflection moves will reverse as it goes past the centre of the ROC. Hone in on the position so that the mirror looks to be full of light and slight movements of the light makes no difference. Lining a tester up is fairly easy but if it's truly slitless it would be best to put something thin centrally over the led with a small hole or rough slit etc in it. The mirror will reflect it back and the position can be found with piece of card. Move things around / tilt the mirror until it's where it should be. A slit and knife edge are handy again here - focus a loupe onto the knife edge and adjust to bring the return image sharply in focus with the knife edge about 1/2 way across it. An eyepiece used backwards should make a decent loupe. 25mm will give about 10x. With source stationary set up there is no need have the knife cut into the beam as the knife can be moved independently. Lining up towards the mirror is a bit more tricky but the return image should remain half masked by the knife through a significant amount of movement. There is a need to worry about this when shadow style testing is used during figuring. Some people add an x traverse to keep the shadow generation even when source and knife move together and only worry about getting it exact when things are nearly finished. I've wondered about using a camera macro slide/rail for this and making the knob bigger to get fine adjustment. It's not so much of a problem with stationary source as the knife edge is usually swung into the beam via mounting it on an L shaped part. Probably all difficult to understand but his rig is shown here http://astro-foren.de/index.php/Thread/4323-Foucault-Test-Einrichtung-nach-Texereau/ a better shot of it here http://www.iucaa.ernet.in/~scipop/Literature/hbt1/fig17000.jpg The basic idea is that the sliding part runs on 2 V's formed by metal plates running on a piece of bar or tube supported by a screw that runs on something slippery. (I just used a bit of metal rather than glass) That screw is used to move the knife edge but the same set up can be used for source and knife moving together. The table usually needs a bit of weight adding to keep it stable and another one can be mounted on top of it to give x-y movement if needed. I'd guess source and knife moving together is the most precise method but the fixed source is probably easier to set up and allows the amount the knife cuts into the beam to be easily adjusted. Trouble is it tends to finish up with a wide separation between the source and knife and that needs keeping down to a minimum really. I have seen figures given based on % rad of the mirror but I'm sometimes dubious about info on the web, i've seen a photo of a pro set up used to test mirrors up to and above 60in dia. Slit and source move together and I would estimate were about 6mm apart. It was mounted on a lathe top slide giving X-Y adjustment and modified to read to 0.0001in. John -
  18. Phew - just read all 13 pages. I reckon you are doing pretty well really. The problem you have now is removing relatively large areas of glass to cure a small area error. Slow work it but should happen eventually. Some might just go back to fine grinding again but I can understand why you haven't done that and you do seem to be making progress. One reason I read the lot is wondering how this happened. I'm at a bit of a loss but some maybe's. I noticed a comment that the centre polished up more slowly. Things usually work from the centre to the edge. Say a small lap was being used on the whole mirror. Some series of narrow w's would be used across the entire mirror. This means that the centre of the mirror has the lap passing over it more often than the edge - the w's might run left right and then right left. This is what usually happens how ever it's done even during grinding - the centre of the mirror spends more time in contact with the tool. As the laps are light you might be pressing down to speed things up. I feel that is a no no. It's better to add weight to the lap. Work height can be used to avoid this as well - if it's a bit higher than elbow height it's harder to push down and easier to just push, pull without varying the pressure on the lap. Less tendency to rock the lap as well. To much effort needed to move the lap - only a maybe but perhaps the mix of polishing compound is too thick and needs more water added. Not a subject that is well covered from what I have seen. As I need to make a new tester I found the testing interesting. The test for a sphere was mentioned where fringes are seen ahead of the knife edge. The knife edge needs to be say 25mm ahead of the light source which should be at the roc (slightly off centre in practice). With a slit source the fringes are strong. The source and the knife edge should be equidistant from the axis of the mirror during any testing. There is some mention of that in places. It's done by catching the return image from the mirror and adjusting so that the source what ever it is in focus and ideally as close to the source as it can be while remaining usable as a tester. It needs to be level with the source as well so that everything is then also square to the mirror. The easiest way to do that is probably to rest the mirror back on 3 adjustable screws or have the base the holder sits on tilted by the same sort of idea. Then things have to be aligned for when the tester is moved back and forth to take measurements. I spent some time looking around at testers as slitless became more popular. The impression I had was that people were prepared to spend a lot of time making a mirror but not much on a tester. For me Texereau's basic method of making a sliding table makes a lot of sense and is easy to make even if it doesn't use the same materials. Given that all sort can be placed on it. I did see one form of slitless that interested me. The usual led in a hole but with a craft knife blade edge of some sort running vertically over it and on centre. Sort of one sided slit where the rest of the blade forms the knife edge while always lining up with it. It would be easy to convert this to a true slit source too. I have my doubts about using large sources. I feel that people in the past that used pin holes or short slits were not idiots and must have had some reasons for not just using larger holes. The reason given how the tests function is probably accuracy. I don't think I would use a wire brush on a lap - lots of fine detail to catch and retain Rubbish. Any mesh that is finer than the squares of pitch on the lap will provide smaller facets with space round them so that they can work correctly. Some one expressed doubts about financial aspects of making a mirror like Damian's. Well Orion would charge from 2 to 3k plus VAT for a mirror of this size. Time is likely to vary according to experience and to some extent luck. John -
  19. I haven't found anything to nudge the telescope pointing around Steve but am surprised if it stops tracking when something does that. I'd expect keyboard shortcuts to be available to correct pointing. Even on a high res 27in monitor it's difficult to select say Jupiter and slew to it rather than a nearby star - with a full sky view. I'll play more when I get back home again. John -
  20. You've lost me their Steve. I have one indi control panel. I assume more ccd tabs come up if there is more than one. It's tracking Jupiter on the other side of the planet at the moment. I wondered what happened if it went off the Kstars screen. I'd say carry on as it has is the prefered option. There was another window last night but it seems to have gone. I minimised it. John -
  21. Maybe I wasn't clear Steve - OpenSuse has had Kstars for yonks and INDI too but for some reason Kstars didn't like INDI being installed. There is no need to install from source. They work together now. The OpenSuse package search offers one click installs and also unofficial installs as an option. The Kstars and INDI I have installed is official. OpenSuse is an RPM based package just like Fedora but the owners are Novel who are mostly server orientated. The syntax for installing is a little different. Most can be done with a graphics utility called YAST. It lists everything in their repo's and also in any repo's that the user adds so installation just involves clicking a check box and hitting continue. It then sorts out all of the dependencies and installs them. It will also indicate when updates are available. Some things that aren't in the repo's such as a photographic package I use a lot called Fotoxx is available in the package search. Other packages are available in the build search which is where lupinix is doing his work. Some debian packages are built there too. I've posted that OpenSuse have "done it" in lupinix's thread on the INDI forum as what they have done may help him sort bleeding out for OpenSuse. That covers graphical installs from the desktop except that areas are categorised. I do build from source at times so click development and lo all of the available dev packages are shown. It's also possible to search packages by description. Kstars is under educational. All of these can be selected during install as well if the full dvd is used. Not sure about the live one. Console installs are different. A utility called zypper is used but to be honest there isn't much point in using it except if YAST can not do a multiply package exchange but I think that is sorted out now. The other feature in the console is say a command that has been typed doesn't exist on the system. It suggest typing cnf <command> which searches the repo's for it and comes back with the zypper command to install it. I use copy paste for that. There is also a very unofficial repo that goes by the name of rpm.pbone.net. Unlike the other unofficial stuff this may depart seriously from the other sources of update. Adding that to the repo list can be a disaster. It's possible to load files from several repo's that are not compatible and finish up having to fix it in safe mode at the prompt. Zypper is useful for that and some one on the OpenSuse forum can usually guide people through sorting out what went wrong and fixing it. There is also some sort of snap shot idea about now but I've not tried it yet - that should allow roll backs. Installing from source isn't difficult. Usually the instructions for installing are in the enclosed readme file or an install file. INDI were naughty on that score but now do provide enough info with the package download. Earlier ones just had a web link - out of date. Compiling needs header files - these are simply name-dev. Also if serious debugging is needed there will usually be a name-debug available. Kernel headers can be needed as well at times. The biggest problem is things like C++ headers but a search for C and C++ in YAST fixes that - there are a lot to click on. To be honest I don't have much interest in bleeding especially if it includes KDE5 as I wont be installing INDI on a separate machine. It will have to fit in with the other stuff I use. I've been on the KDE mailing list for donkeys years and can't wait to see what happens. There has already be one question which suggest a complete difference to 4 and comments about certain oddities about 4 still keep rolling in. The list is a shadow of what it used to be though. On 3 there might have been a couple of hundred posts in some weeks and always some one about who could help and also at times change things. The KDE forum is more active now but it can be frustrating - suggest a change - great idea - I'll look at that - and it doesn't happen. I have more luck with Novel's bug reporting. It usually happens. John -
  22. Something interesting as happened. I decided to install Kstars on my desktop machine as the one in the astro distro seemed a bit odd to me. Much to my amazement it picked up that indi was about but when I fired it up and connected the whole thing crashed. I then tried the package search for INDI and it came up so selected one click install. The stuff I had tried to install manually caused dependency problems so I selected down grade. All seems ok now and proper KDE windows which can be minimised. I've not used root or sudo other than root password to install. So far I have parked the scope simulator and then slewed to Jupiter. It's tracking it too. I need to bitch about what else I can do with it on the INDI forum before connecting anything real to it. This is OpenSuse 12.3. As that's not supported any more I was surprised that I could update without any problems. I'd guess that the stuff is also available for 13.2. OpenSuse seems to be a bit of a culture shock for Ubuntu users. I don't see why but it is best to understand partitioning when installing it if anything other than defaults are wanted. I always set mine up manually but the defaults were fine. If Linux is already about it probably wont wipe these areas by default and that can cause problems if it's a different distro. I had that happen playing around on a netbook. From memory custom partitioning has to be selected to format any of these but it will retain the default layout or even import the existing one. If some one wants to try it they may need the package search to get the correct files. That's here and will default to INDI. Then search Kstars. https://software.opensuse.org/package/libindi Don't expect INDI 1.x it's a 0.9x John -
  23. I've been determined to get something up and running and have spent days trying to get the INDI "stable" up and running away from Ububtu. I had a feeling that I might be able to do it if I could find older source but not much luck so far. More recently I noticed a show all downloads option on the INDI site. There is a virtual image of the lot I assume to allow windows and mac users to use it. I've given it a go on Linux under virtual box. First problem was it took 5 downloads to get one with the correct MD5 checksum. Then I found that it wont run in VB, ubuntu options come up followed by a black screen of death when it boot fully. At least it's not blue. To check that my VB was OK I then installed AstroDistro (AD) in it. No problems running it as a live DVD other than VB has changed since I last used it - have to click the cd icon and navigate to the ISO file which I have on my hard drive. I can't install without changing the virtual hard drive size. It all works though complete with a resizeable virtual window which functions correctly. It's double click as far as the desktop is concerned - not the way Linux should do that. Oddly AD doesn't put Kstars on the desktop but it is available from the start button so I have now at last seen INDI / EKOS in it's full glory but only with the telescope and ccd simulators plus guiding via the same ccd. I have this thing about software - should be possible to do something with it in 1/2 hr at the worst without any help files etc. No luck so far, none at all. It might help if I could drag the scope view around in Kstars but so far haven't found out how. I thought I might be able to take some perfect images via kstars maps - again no luck so far - just ccd noise. I noticed that the EKOS window remains on top and can't be minimised and that I can't find what version I am running from that either. I've already found that playing around isn't a good idea and couldn't find the abort button - maybe that should always be on top. I can't see why the Ekos window has to remain on top etc. It hasn't got the usual KDE window options either except remain on top which isn't ticked. I've just found out how to slew the scope but when I tried to calibrate the focus it crashed. I'm getting the impression that it's too complicated. Also note that it's shifting to KDE5 shortly. KDE4 still has irritations and some feel that they wont get fixed. There have been various comments about what KDE5 will be - including a VM. John -
  24. Maybe a bigger reprap Gina. On the other hand it is possible to cut aluminium on a band saw and finish the edges off with a router. Routers can follow straight edges or even wooden patterns. My brother works for an aerospace company and they needed some large aluminium discs so they mounted a router on a trammel. I've had no need yet but tried routing the edge of some 6mm plate. Looks feasible to me given smaller cuts than people might take in wood, slower feeds seemed to be the main thing. John -
  25. I think I would try a large star lap before deliberately changing part of the mirror Damian. That way you can work with some overhang without working the edges much and still keep an even figure. The other way is to press the lap over a ring round the circumference. Didn't work for me, I'd guess it depends on how soft the pitch is. I'm guessing but with an 18in lap I would try 5 or 6 points going say 4 in into the lap and using an over hang of circa 1/2 that. More points = more work biased to the centre. You'll probably find that this deepens the centre but I'd be surprised if it removed all of the glass that has to be removed from there to form a parabola. What I found with laps was they had to be tried to find out if they were going to work out. A bit suck it and see despite all of the info that is about. The classic small tool on top for getting to a parabola would be 1/2 diameter. Used to the edge to deepen the centre and it's centre run round the edge to correct that part. John -
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.