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. Me too. There is always some wonder fix around that should fix all problems what ever they are. The fact is that a centre punch mark will pull a drill for the simple reason that it has slightly less work to do where the mark is. If the drill is centred on it all is ok. One school of thought suggests light centre punch marks another sort of average what ever that is. With rather small drills that bend easily the punch even needs to be held vertically when it's used. There are drills about that can be used on a lathe without using a centre drill first. It's interesting to note that if the tail stock is accurately aligned with the headstock and the end of the work is squared off and smooth ordinary drills are ok used like this too. There is no reason why they should wander if they are sharpened correctly. Bad practice, centre drills are used first. These work because they are designed not to bend and are perfectly happy within reason just cutting on one side. When they do that the work is rotating so they form a hole which is on centre - some people break the tip of the smaller sizes because the tailstock is too far out. Switch to a don't need the centre drill types and given some reason to bend such as the tail stock being out or the surface being rough or out of square and they will. Exactly the same factors apply how ever a hole is drilled - so much for giving them funny ends. That's why square to the work was mentioned along with them. They are an un needed fix as far as small drills go and in larger sizes the ordinary types are far more tolerant so who needs them really. While it is a good idea to have drills that can bend some what without breaking there isn't much point in expecting this to totally fix problems with deeper holes or aluminium melting and sticking to the cutting edge. It just helps when they have low diameters. One aspect of why bent drills have problems is that they don't cut very well on the sides. One important aspect is that they are sharp. A jig can be bought to do this on an off hand grinder also other things just aimed at just sharpening drills. Not sure just how good the later are if they are affordable especially as small diameters need precise re regrinding. The accuracy needed relates to the diameter. Must admit I have an ebay box that contains maybe 10 of each size in 0.1mm steps up to something like 3mm dia that I use rather than resharpening. I haven't bought a box of drills for a very long time. I have one box that I only use for holes that will be tapped or reamed, 2 actually to cover the size range. Another that I just use for run of the mill stuff and sometimes over a period of what is must be well over 10 years some small ones break and I top them up from the ebay box. That's why I bought it but these break more readily than the ones that were in the box. These are not terribly good drills, sizes can be a little off but I suspect the main problem is sharpness. What I should have done probably is buy specific sizes of some where like this http://www.drill-service.co.uk/ Jobber drills unless something specific is wanted. Or some where else similar. There may be cheaper sources about and some that sell them in packets with several in. John -
  2. I think you need a good spell with the larger lap now Damian to minimise the possibility of a hole, centre over centre and zero overhang, probably even stopping a bit short of that. This will also reduce any hill in the centre and a 10in lap on a mirror of this size is a small lap really. Pass on what might happen if the pitch is too soft. I feel I would be thinking and maybe struggling with a 15 or 16in lap on a mirror of this size. Even bigger if I could manage it. One thing I am curious about is if you are using cold tar or pine pitch. John -
  3. I bought 5ltr of cutting oil from Morris Lubricants mail order some time ago. The price is very reasonable and 5ltrs will probably out last me. I mostly use it on my lathe - a smear on final cuts on nasty stuff such as stainless or silver steel applied with kiddies paste brushes from a £ shop. I prefer the wooden handled ones. It's ok on taps and dies too but I generally just use ordinary solid hand soap. It or any oil really shouldn't be needed drilling ordinary materials. The thing to look for is the end of the drill pulling in some direction when it engages with the work. If that happens small drills are likely to break. If the drill is brought down on a centre punch mark slightly to one side even that will tend to pull a drill. Once it's pulled it's bent and things get worse as it goes further into the work. The drill will also pull if it isn't square to the surface of the work. It's worth checking drilling machines for that problem also loose spindles. Clogging drills up with swarf will break them too - deeper holes in aluminium can be a problem as can drilling several holes in one go - it's very good at melting and sticking to drills and lathe tool especially parting off tools. It can be flicked off with a scriber. Little else seems to shift it. John -
  4. I think Agnes was asking you to test it as a sphere which means that the led and paper will be twice the focal length of the mirror away from it. Rather than just using a led it would be better to put say a 1mm dia hole in something opaque and place the led behind that. Smaller the hole the better really. Your F12 mirror should be so close to a sphere that the difference will be negligible and it will focus pretty distinctly when it's at the right distance - more usual F ratio paraboloids wont. The Faucault test makes use of that fact. They might with a suitable aperture mask. I'm going to ask something really silly so please don't take offence. I assume both mirrors are front surfaced coated? If not multiple images are more or less to be expected. When the coating is behind glass the glass reflects surprisingly well so usually 2 images are formed from one surface. It's possible to make a decent pin hole with a LED. Cover the end with aluminium foil and the rest with say blue tack. Make a hole in the foil with a pin, rotating it rather carefully keeping it vertical usually makes it round. The problem with testing a mirror as Agnes outlined is that the size of the image of a planet on the focal plane of your mirror is really tiny. A led is way way bigger and size could very easily hide any problems. People have been known to strop good quality needles on news paper to sharpen them further and get holes down to 0.002in by making it over lightly ground glass. If it's a led just press very lightly. If the return image doesn't look round rotate the led to see if that is causing odd shapes. Or rotate the mirror to see if it moves with that. John -
  5. The contrast plot doesn't really get the point across Agnes. What is actually happening is that increasing the size of the 2ndry diverts more and more light from the diffraction spot into the rings which limits resolution, the spot is dimmer and surrounding areas are lit more than they should be by the rings, net effect reduced contrast especially important when viewing low contrast detail in objects such as planets but it also has implications on other extended objects. It might be more interesting to show a plot of Strehl ratio, not sure as it might have limited effect. The usual 99% etc ratio represents 99% of the theoretical encircled energy which even on a perfect optic isn't 100% due to some going into the rings. MTF curves are really the best way to compare various effects as they also show how contrast is what actually limits resolution. That is around 7 1/2 % at the resolution level usually specified on astro scopes - useless except for spitting double stars yet some people claim to have seen to this level on low contrast objects. A worse problem in many ways is the accuracy of the mirror figure that is needed on a Newtonian, Even just a 1/4 wave error causes the brightness of the spot to drop to 80% of it's peak value. I have no idea what this means in terms of energy encircled in the diffraction spot but I'd guess it makes that figure worse. It generally reckonned better than at least 1/10 wave is needed. This is why NASA shot for 1/30 wave on Hubble. Opinions vary. One other factor that might interest some one aiming primarily at planets is the off axis performance of newtonians - others too really. I'm fairly sure a simple formulae for Shrehl down to 80% can be found in the Newtonian section on this site, the numbers get worse with faster F ratio's. http://www.telescope-optics.net/index.htm There isn't much talk in amateur circles about encircled energy. Strange really as this is how the pro big scopes tend to be specified. X% of light into a spot of such and such a size over some field angle. It's probably irrelevant on most amateur sizes but at some point as they get bigger and seeing conditions limit resolution/magnification anyway it can be relevant as the conditions can swamp optical errors. One solution at some point that might be used is to settle for 1 arc second resolution rather than the theoretical maximum of some bigger scope. John -
  6. People often mention % area of the 2ndy mirror Agnes and don't appreciate the % diameter effects. 25% by diameter is generally considered ok for general use but planet people often try for less than 20%. I posted some screen shots of what happens as it goes up in the F12 4in thread. Past 25% it goes up rapidly. I showed the coma on a double sized scope too, F6 8in, up like a rocket but forgot to change the size of the 2ndry mirror otherwise it would be worse. I could post some more shots of the effect against no obstruction if any one is interested - when I have 1/2hr to spare. Size of scope doesn't matter. It just changes the numbers on the resolution axis and faster Newtonians have rapidly worsening off axis performance. SCT / compound scopes etc can help with that if they are designed correctly but usually suffer from having larger central obstructions. John -
  7. Try rotating the mirror if you can some how and see if the pattern moves with the rotation. I doubt if mounting is the problem but you could improve the support - 3 screw heads for the back of the mirror to rest on placed at around 70% of it's radius. For location it's hard to beat cork just touching it other than the difficulty of getting good cork. It needs to be pretty hard. John -
  8. I understands some people have used paint that will pop up if ebay is searched for military paint. Machinery paint will bring up more as well but it tends to be more expensive. Ideally under that there needs to be a good coat of a suitable high build primer. The layer of that on some lathes which I assume were just cast in ordinary sand is often pretty thick. Personally I would just clean up what's there. A few chips add to the charm. Repaint if lots are missing - many home repaint jobs viewed from close up don't look very good. It takes a lot of work to get things smooth enough for a glossy finish. I just found the thread on the end my Pultra's spindle. Hidden under a screw on cover that in turn is covered by a pull off one. I can now look at fitting it with Taig chucks. That will allow me to get 1/2" bar through the spindle. A fair bit more than the Taig will accept. It's about the same as the Taig when the collet draw bar is used. John -
  9. Here is the Pultra, complete with it's built in work lamp - just to show what mad moments are all about. The smallest collet seem to be for work that is 0.4mm dia. The centre height is 50mm but it also came with rising blocks that will lift that to 90mm. As I was allowed to use a lathe at work if I wanted my first lathe at home was a Peatol better known as a Taig in the US. That has a centre height of around 56mm and it is surprising what can be made even with a lathe that small. The biggest problem is the need for a lot of stub drills as the centre distances are rather low. It's also easy to cover one with swarf to a point where it's hard to see what's going on underneath it all. The Peatol was extremely accurate all round when I bought it but over time the heads bend and they start turning a taper rather than trully round work. I'm hoping this one wont have that problem but now have the joys of carefully adjusting everything and hoping that I don't come across some bad fixes applied by previous owners. It dates from around 1955 and was owned by an instrument maker. No problem there but I worry a little about what his son may have done while "restoring" it. This lathe uses phos bronze bearings, probably on a hardened spindle in this case. Might be worth mentioning that most people reckon that SAE 32 hydraulic fluid is the correct lubricant to use with both bronze and cast iron bearings. The Pultra has small oil reservoirs. On many lathes a small drop should be added to the bearings each time the lathe is used. The bench drill behind it came with it. 3 speed going from 4 to 8 thousand rpm. I bit fast for anything over 1mm dia and even that might be too fast. It seems it can be used as a sort of surface grinder when fitted with a suitable grinding point. Have to see. That could be useful. PS The sweet tin has all of my Boxford gears in it. The tidy up is still on going. My PC speakers are a bit bigger than usual too. John -
  10. Not really. It's not that easy a Not really. It isn't an easy thing to do that way. A better option is to drive a low side switch via a resistor. The gates have significant capacitance so this will slow the switching down but also increase the dissipation in the switch. The usual aim is to switch them as fast as possible to minimise that. It can actually be tricky to make these things switch as fast as the data sheets suggest. It could be done by driving the dew heater through a resistor with a capacitor on the other side of that. This could give what ever time constant some one desires but it wont be very efficient when amps of drive are needed. The resistance value would have to be chosen to keep the turn on current within suitable limits - just sticking a 10uF capacitor across the wires to the dew heater might well destroy the FET as in principle the current that can be taken out of the capacitor would be very high. Actually capacitor have ripple current limits so that could cause problems any way. If some has had interference problems due to switching I would hazard a guess that it has nothing to do with the switching speed - more likely to be a problem with supply decoupling to the various bits and pieces being powered from a single supply. The box of tricks with the switches in it should have a substantial capacitor in it to help hold up the supply when sudden pulses of current are taken. John -
  11. There is or maybe was now a man in B'ham who rubbed something on rust on lathes and over a period of time it just flaked off leaving nice shiny metal. He wouldn't tell anyone what he used. He seemed to rub a bit of what ever it was on now and again. Rust remover works well but the metal discolours rapidly afterwards. Some one on the telly derusted a square by cutting a potato in half and rubbing it on the rust. Slow but it does work, Lathes with shimmed bearing caps often have no shims left to remove. Myford ML7's use this sort of set up originally with white metal bearings but they later supplied phosphor bronze - the 1st batch had way to much excess metal that needed scraping out and don't I know it. They refused to supply the more sensible ones they later supplied unless I just paid for another set. No way at £80 odd a set. The way these types are initially made is pretty simple. The bearing housings are only truly round when a certain size of shim is fitted under the caps. They make the bearing shells so accurately that they will come out correctly when fitted or maybe a few may need a tiny bit of hand scraping. When the bearing has worn the general idea is that a little bit of the shim thickness is removed to close the bearings up and they are then scraped to fit the spindle correctly. A razor sharp 3 square (triangular) scraper is used to do the scraping. The fit is judged with engineers blue - an incredibly thin layer, so thin it's not that easy to judge the high spots that need scraping down. It can be done in 2 halves. Lower shells first checking for fit and that the spindle is level and aligned with the bed using a DTI. Then the top shells just for fit. They have been using laminated shims for this sort of thing for some time. The idea is that changes can be made by peeling small thicknesses off. To honest I have found it easier to buy shim material and select thickness to suite. After I did my ML7 my father told me that some people used to make things that allowed them to bore new bearings from the lathe bed. the tool being turned by hand. Ok as very little material has to be removed. He did this sort of thing early in his career - I found his scraper after he died. It's size was around 1/3 of what it originally was due to sharpening. He must have done a lot of it. The sides of the scraper should ideally be hollow ground. Some people have ground them from a 3 square file being careful not to over heat it and wreck the temper. They should be hardened and tempered to a very pale straw. John -
  12. I'd be interested in exactly what the washing soda rust removal process is and how well it works. Others may be too. In the past I have had chucks that have a sort of brown patina rather than rust. It's almost shiny. I think it's caused by soluble oil coolants. I leave that alone also just plain discolouration. I've used paraffin and a kitchen type scotch scouring pad to just get the crud off the later. Paraffin helps because most lathes have been well oiled at some point in their life. Theory has it that some soaks into cast iron so some rust can be simply lifted off with wet and dry. I suspect it's more a case of it acting as a lubricant and preventing the wet and dry from clogging up. Some of the gearing that forms the back gear is often part of the pulley. I believe that lathes.co.uk can supply flat belt if some one wants to keep that. There may be other sources. An inverter can be used to slow lathes down but there is risk of burning the motor out if this is done with higher loads for long periods. Most lathes come with the slides set loose. In order to produce the best work that they can this isn't ideal. The best way to set them on the cross and compound slide is to remove the lead screw and then adjust until there is slight resistance when pushing the slides by hand. It's also possible to do this by detecting a slight increase in resistance at the handles with the lead screws in. Most older lathes will show signs of wear usually going too tight as the cross slide is wound further out. Sometimes it's possible to fix this by adjusting the gib strips in stages where they pass over the dovetail they are running over - hard to explain but the dovetail might be 3in long and the slide a lot longer than that so the gib strips can be set with the tool wound right in, half out and fully out. If a lathe looked promising tested and adjusted via the handles like that I would strip it down, clean it up and use some slide way oil when putting it back together. Headstock bearings are a pain. Some spindles just run in cast iron, split on one side with a pinch bolt. The castings sometimes crack when these are adjusted. Other have phos bronze bearings. Some of those are tapered with a nut on one side which pulls them into the taper to close them up and another one on the other end to lock the setting up - it's important to slacken that one off before adjusting the other. It's possible to make new bearings as well. The outside diameters need to be correct, taper too if there is one. The inside bore is made a few thou undersized and then fitted to the spindle by hand. A scraper or maybe an adjustable reamer. Some lathes use taper roller bearings etc. Unless they are really bad and even then it's best to just adjust them as lathe precision bearings are very expensive. I've found heat the best way of adjusting these - they should warm up a fair amount after the lathe has been running for say 20min at medium speed. The bearings themselves might be running at around 100C so the castings around them will warm up noticeably. If thing have to be loose on a lathe, wear etc the best answer when a good finish is wanted is heavier cuts and a very even feed. How heavy is TBD but people shouldn't be afraid of trying say 1 to 2 mm or even more. Much depends on the feed rate but basically the net effect needs to load the loose parts of the lathe sufficiently to force them home and keep them there. Another approach is to just keep running the same cut until no more metal is removed. This works well on some. If all else fails it's files and emery cloth etc but the lathe needs running at a low speed when a file is used. In my experience Chinese lathes are different largely down to the cast iron they use. It can also be very difficult to correct any built in errors. John -
  13. The poll is a difficult one really. I mostly use Linux for everything so I voted for Linux. I do have windows about - I'm on a vista laptop now away from home but that plus updating software in cameras etc is about all I use it for. I have recently bought another windows laptop but haven't got round to sorting it to my liking yet. Waste of money really but I feel I should ditch this one but I don't like the look of win 8 at all. Of late I have developed an interest in telescope remote control. There is a LInux package based around Kstars for this. If interested it would be wise to check that it can work with the gear that people already have. There is a problem with it though. It is always using what is referred to as bleeding edge software. Some one who uses their machine for all sorts of things is very likely to find that it's buggy. Stable versions of Kstars are available as with the vast majority of Linux software but for some reason the people who maintain the telescope control aspect don't maintain one. There are a couple of packages about that offer similar facilities that can be run via a tablet. It runs on a TV dongle or Ras Pi etc and functions as a web server. All of the hard work is done in the Pi. Sensibly one person that produces this is very keen on stability. These are also Linux based. There is a well known Windows telescope control application and as far as I am aware most equipment manufacturers produce drivers that can be used with it. Linux has the problem that some one has to decide to write one. This may or may not be a problem depending on the equipment that is being controlled. It looks to need more power at the telescope end than say the ras Pi linux approaches. In terms of image processing I don't think there is much difference but Photoshop won't run on Linux. The GIMP just like it has a steep learning curve and often required entirely different techniques. I do a lot of normal photo processing and find under Linux that I use more than one package. At the professional end of astronomy I am told that Linux has 100% of the super computing market in all areas. I have also seen at least one package that is used by professional observatories. It was too obscure for me. It might pay to remember that windows desktops are an add on to Linux. It's a console application. Some ms windows applications can be run under Wine on Linux and on a Linux desktop. What will run can be a bit hit and miss. One method that can be 100% successful is to run ms windows under a virtual machine. Probably the easiest one to use is VirtualBox. There are others. My main use of Wine is for odd bits and pieces that were written for ms windows or dos via the dos emulator years ago. Such things as telescope mirrror Faucault analysis software and odd bits and bobs like that. It will also run an earlier free version of a comprehensive optical design program now. Previously only some parts of it would run. More recent free versions are severely crippled so of no interest to me. It's very expensive software bought out right. Wine is subject to continuous development so what it will run evolves over time. John -
  14. If I had 2 Portass lathes I would rig both up some how and adjust and try them to see which one was best. Grade 400 wet and dry plus paraffin will make short work of the rust on the bed. Maybe a grade coarser for some things. Cast iron is so hard the chances of removing a significant amount are slight. Chucks can be hard to clean up fully. John -
  15. I'v had a mad moment. Making small things such a m2 thrumb screws for microscopes on a Boxford isn't much fun. I've meant to do something about this for some time. Nosing around I came across a Pultra 1750 complete with all of it's collets that seems to be in fine condition. I'm off shortly for a few days so no time to take a photo. It's neatly mounted on a small cabinet that just fits by the side of the miller but needs wheeling out to use. It's on castors! There are some details about them here http://www.lathes.co.uk/pultra/ Everything is scaled down from a full sized lathe on this type including feeds so they are ideal for making smaller parts. Max rpm is around 5,000 which again helps. However I suspect I have made all of the microscope bits I need to . However I'm wondering how I could make some thread chasing gear for certain telescope bits. John -
  16. That will make a small difference. Not bored any more but. This is 4in F12 optimum 2ndry, 28mm 2ndry and 1/3 main mirror 2ndry. Which is which is pretty obvious. The 1/3 is outside Rayliegh's limit so wrong about the software. Silly me it's the axial one. I didn't alter the obstruction of the 2ndry on the 8in F6 otherwise it would show some sag. The problems with the larger F6 mirror are nothing to do with the size of the 2ndry. The difference is purely down to aberrations. Like most things it's not that simple but if a modulation of 0.5 is picked the F12 mirror is better across the field than the F6. 0.5 loosely means that the actual contrast going into the scope is reduced to 1/2 of what it should be when it comes out. As stars are black on white that doesn't matter much, contrast is way way down at Rayliegh's diffraction limit anyway but on extended objects which generally have low contrast levels it has a noticeable effect. One other thing that the table on the 4in F12 shows is the radius of the moons image so when at it's closest it will be about 13mm dia. Well within the scope of 1 1/4 eyepieces. The side effect of all of this is that as scopes get larger and shorter the angular field they can cover well with simple optics gets smaller and smaller. If you read up on some of the big observatory scopes where they use extensive techniques to improve this they can still finish up with field sizes that are measured in arc seconds rather than degrees. This is why they also have smaller scopes about as well. It's a curious area really. It doesn't mean that people shouldn't strive for bigger and bigger mirrors it just means that on things like the moon at some point it might not be as good as expected - also that the whole moon might not fit in a 2in eyepiece at some point. John -
  17. As I'm bored and curious I just laid out a 4in F12 newtonian. As I expected it's not far of perfect for a 4in scope over a field angle sufficient to cover the moon which at the size given needs a 22mm diagonal but 25mm is ok. This is the result It includes the effects of the 2ndry mirror. The mtf plot would be more or less a straight line without any sag in it if that wasn't there. This is a plot of contrast against resolution. If I increase the diameter to 8in so it's then F6 this happens The black circles in the spot diagrams are the size of the diffraction spot so set just what can be resolved but the contrast will be very low at that level. So if looking at Uranus, careful choice of words, the 8in can F6 can clearly do better than the 4in as the field angle needed is tiny and newtonians can be perfect on axis. Over wider fields things aren't so simple as the plots show. The blue line in the 2nd mtf is the perfect one for comparison. The software throws that in if when the plots exceed Rayleighs limit. The plots represent perfect optics - worth bearing in mind too. John -
  18. It's better to think in terms of the angular field size rather than just say I want a field size of X mm. Then look at the implications of the size of secondary needed which is pretty simple really. 25% by diameter is generally considered ok but 20% is better. Some people who want a planetary scope might try to get smaller than that. 30% etc is more likely to be considered as satisfactory for photography but when it gets to this sort of size there are most definitely contrast penalties. Contrast in real terms relates to resolution. Focal length sets the image scale. Diameter the degree of light level amplification but on newtonians the off axis aberrations climb rapidly as the diameter goes up for the same focal length. They go down if the focal length of the scope is increased so Texereau starts on that basis by setting just how big the aberrations can be. Pretty sensible really. Borrowing an image the effect is shown here This shows 2 scopes same diameter one has twice the focal length of the other. The aberration level of the longer focal length is 1/4 the size of the shorter one. It's a square law which is why the aberration climbs so rapidly. There are all sorts of approximations about for determining the field angle that a secondary covers. The borrowed image shows how they work. The dashed horizontal line shows an axial light ray. The vertical one the focal plane. Forget the aberration and assume some angle of incidence as shown in the diagram, the point where the reflected angle hits the focal plane gives the actual dimension this represents on the focal plain. It's easy to sketch this out add the 2ndry mirror and then calculate sizes via say a spread sheet so that the numbers are easy to change to see what happens. A popular angular field size might be sufficient to cover the moon, slightly over 1/2 degree total. If some one is limited to a certain ccd size there is no point really in exceeding what that represents in angular field size. The same idea and basic trigonometry can be used to work it all out including the diameter of the cone of light where the 2ndry has to be placed. For 100% illumination that mirror needs to catch all of it. Past that point it will slowly vignette and eventually not capture any light at all as the off axis angle increases. Forgetting the aberration might seem strange but another borrowed image can show why A lens but that doesn't matter coma is coma. This shows the light distribution in it. The blob in the right place is similar to the diffractions spot size of the telescope = small. The light in the wrong place fades slowly so any vignetting will make it fade more quickly. It's pretty small in regions of interest anyway - eg Texereau sets a limit of 0.1mm and then considers the other factor - % secondary diameter of the mirror. People tend to rabbit on about % area which is misleading really as it only accounts for the light loss it causes and not the effect on resolution and contrast. Compound telescopes, cassegrain types, macs, sct etc sound bad in terms of the size of the secondary mirror often up and over 30% but telescopes are a compromise and as these have more active optical components in them they can reduce things like off axis coma. A mac for instance can be designed to have none at all. All it means in practice is that a smaller scope with a smaller or no obstruction can match them for resolution. This essentially also means higher contrast for the same aperture. In my case I have actually seen this visually comparing a good 8in F10 SCT with a good 5in F9 apo. I'm pretty sure the 5in apo wouldn't compete with a 10in sct resolution wise. It's relatively easy to come up with a good quality F9 5in apo using the more exotic glasses. As with all scopes shorter focal lengths makes things get harder and even more so as the diameter goes up. John -
  19. When I mentioned an image of a star not hitting the mirror as quoted below I should have said the secondary mirror. Any form of secondary mirror has the same characteristic no matter how small or large it is. At some field angle it will only catch part of the cone of light heading for the image plain. An image will still form but will be dimmer because of the vignetting effect the 2ndry mirror has had. Please excuse my use of don't rather than doesn't. Done on purpose actually. John -
  20. No Agnes, vignetting caused by the 2ndry mirror how ever big or small it is but true the tube can vignette the main mirror. That is just going to relate to the required view angle and the tube length from the edge of the mirror. TeleVue go through the eyepiece aspects but few other manufacturers if any give field stop diameters http://www.televue.com/engine/TV3b_page.asp?id=79#.VVWsa-c0YoE The scope focal length sets what view angle can fit in the field stop as indicated there. Books give other formulae for field size, usually approximations. Wider eyepiece viewing angles allow shorter focal lengths to be used. That allows wider fields to be covered with smaller exit pupils that can still pass through our eyes pupil. Personally I only use them when they are needed to do exactly that. As I see it if an ordinary eyepiece gives an acceptable exit pupil I see no point in using a wide field type eg twice the angle and 1/2 the focal length. This attitude might not be good for expensive eyepiece sales. John -
  21. True about dimness after a fashion but imagine the image of an off axis star - if it don't hit the mirror you wont see it. At the mirror the light from a star is a cone so if only half of it hits the mirror approx 50% of the light will be lost. Some see that as acceptable and it will happen anyway however big the mirror is. A star further off axis might loose 66% of it's light etc. There are 2 aspects to the 2ndry mirror. Field coverage and light loss over the field. What ever size of 2ndry is used there will be light loss at the edge of the field it "covers". It's simple trigonometry to work out how big a mirror needs to be to give 100% transmission over a certain field size. One other aspect is when 1 1/4 eyepieces are being used. Not much point having a field over that size. As the eyepiece focal length gets shorter or more correctly it's field stop gets smaller the same applies. Texereau uses another method of determining it's size - a limit on radial fanning of stars due to coma. 100um if I remember correctly. He didn't anticipate amateurs using coma correctors but even these leave residues. He also points out that on longer focal length mirrors the rule can result in a very large 2ndry and that this has it's own problems though he doesn't mention what they are. His suggestion in cases where a large 2ndry could be used on his basis for say photography is to have 2 different sizes. John -
  22. No 4in F12. That will be more or less totally free of coma. I have a 4in F8 kicking around some where with a spherical mirror. They can give surprisingly good views. The only problem really is that the size of the 2ndry limits the field of view. Make that small enough and this sort of scope could compare well with an apo refractor- except for field of view. Make it bigger and the contrast drops - biggest problem with SCT's where it gets up close to 33% or more. Mac's too. One other mistake I made on my wooden tube was making it hexagonal. Octagonal is a better option when bits and pieces get fitted. John -
  23. It's a common problem. More expensive drills are hardened in a different way so that they can bend. By a surprising amount in the past. These days they tend to be hardened along their full length and I suspect different alloys are used. Technique needn't have much to do with breakages. The main one is speed. Taking an extreme watch maker types are likely to want a lathe that will run at say 4,000 rpm or even faster. While holes can be drilled in principle at any speed they will only accept a certain amount of cut per revolution so if they tried to drill a hole at 100 rpm they would have to feed it in incredibly slowly, same at 4,000 rpm actually with very fine drills. Also using the same analogy but little to do with technique the tail stock and headstocks on the lathes they use are aligned to very fine limits. It will also remain aligned as the drill goes into the work. If they were not aligned like this the drill would have to bend at some point as it went into the work. The fact that it has bent will also mean that the hole wont be straight. Poor sharpening can cause drills to wonder as well. The main thing technique wise is drilling into centre punch dots or holes started with a centre drill and using a feed rate to suit the speed. If the drill pulls off centre as it starts to cut one fix within limits might be to lift it out and cut again and so on until it doesn't. This will help but not completely fix the problem. In terms of what is being used to drill the hole making sure the drill is square to the surface of the work is the most important thing. If the bearings in a pillar drill are loose this probably wont help, same with lathes. I break drills in wood sometimes well sub 1/8in dia using a battery powered drill, I never do with an old elu mains powered one. Expensive battery drill too. On a lathe it's not too difficult to make a test mandrel out of ground silver steel if a dial test indicator and a fixed steady is available. Hold one end in the most accurate way possible with a collet or chuck, it might be a good idea to check for run out. Then hold the other end with the fixed steady and get that running as true as possible, no deflection on the DTI. The steady should close to the end. Face the end and centre drill it. Bought test bars will be drilled in a depression on the end to protect them. Don't drill in too far. An and angled ledge of say 1mm or so is plenty, that way the drill will self centre the hole in the bar. Reverse the bar and do the same to the other end. It can then be used between centres to check and adjust the tailstock alignment with the DTI. They can also be bought with a morse taper on the end. Personally I feel parallel ones between centres are better. It's important to use the DTI in a way that allows it to be moved across the diameter of the test bar both horizontally and vertically to measure at the highest points. There are some magnetic based mounts about that have an articulated arm. By testing with the tailstock barrel at 2 extensions how well that aspect is set up can also be measured. Shims can sometime fix problems. Minor errors can be removed with morse taper reamers - sounds awful but this is usually done by holding the reamer carefully in a 3 jaw. Extending the barrel of the tail stock say 1/2 way and pushing it along the bed onto the reamer. This works with heavy tails stocks as the weight causes the hole to self centre, the reamers are stiff. On light weight set ups it might be better to do the same thing followed by clamping the tailstock and advancing the barrel a touch. It's not possible to remove much metal this way as at some point the morse socket will be too big. If some one is unsure of what they are doing they could wreck their lathe. I've done it a couple of times following advice from a fully fledged tool maker. Not clamping the tail stock is an important aspect. They all say that as silly as it sounds. It's really a method of correcting thous not gross errors. John -
  24. Cotton reel maybe? John -
  25. I thought it was worth straying a bit on this thread Richard because lots of people want lathes and it can be a bit of a mine field. The Raglans are very capable machines. In respect to ME10's the floor standing boxfords with the drive in the cabinet have several advantages but I needed rear drive :-) otherwise I would have lost a cupboard and would have to remove an unused sink. There is storage space under the sink too. My Boxford came with the 127T gear and many people want them but really for the majority of things it's best to google screw cutting conversion gears as I mentioned earlier. What I did though is use a spread sheet to see what I could do with the gears I had. It turned out that Boxford used one needed for some of the common metric threads as a spacer so I just needed one other. The pitch error is small enough not to matter. I have used these as it's quicker than changing the whole gear chain and looking at a massive table of set ups for precise pitches that will still really not be exact any way for a number of reasons. I always use imperial threads when I can as the screw cutting indicator can be used on all of the ones I am likely to cut. The machine has to be stopped and run backwards for metric threads to keep the tool in sync with the thread that is being cut. There are other ways but the time between cuts can get enormous. The same is true when cutting metric threads on a metric lathe other than some which have a certain relationship to the pitch of the lead screw. The boxford metric lathes have 2 gears on the indicator to help with this but it can still be a problem. Colchester did an extremely complicated one that could cope with lots of metric pitches but the time between cuts could still be enormous so when it's a problem people still use the screw cutting indicator after a fashion. ie Engage screw cutting and cut along the thread. At the end of the thread note the screw cutting indicator position and disengage screw cutting. Wind out the tool, stop the lathe and select reverse When the indicator shows the same position re engage the screw cutting Allow the tool to run back to the start position and stop the lathe Set the cut and then take it with the lathe running in the usual direction. Repeat etc. This is why some metric lathes don't have screw cutting indicators - other than working up to a shoulder which is pretty easy done as above they don't offer much help. Some pitches can just be engaged and disengaged providing they have a certain relationship to the pitch of the lead screw. More can be done with a screw cutting indicator, some don't benefit from one at all other than for use as a disengage re engage indicator as above. I did think that the ISO people might sort this out but while they have rationalised the range of metric pitches they don't seem to have. There were myriads of them. Imperial managed with 3 basic styles, fine and coarse with tpi suitable for use with screw cutting indicators. Then BA for instance which seems to be metric based (127 tooth gear) and perhaps the equivalent of metric fine. I suspect that the American fine screw pitches may be more sensible in this respect but maybe not. John -
  • Create New...

Important Information

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