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.

Ajohn

Members
  • Content Count

    783
  • Joined

  • Last visited

Everything posted by Ajohn

  1. This links sums up what has to be done to make the OP's idea work http://www.dgmoptics.com/oa_what_is.htm and it still isn't that straight forwards. Notice how long the F ratio's are. That's due to the way images deteriorate off axis more and more as the F ratio gets shorter/faster. I think that the above link is taking liberties really - probably because they sell them. One simpler way of looking at the problem is using a purely spherical mirror. That can be tilted around it's centre of curvature and be in the same position as the surface of a bigger mirror would be if it had the same radius. A paraboloid wont do that. A spherical 8in F9 mirror will meet Rayleigh's criterion without any figuring at all. The problem is that if this is tilted so that the centre of the view is at the edge of an 8in tube the rays look like they are coming from part of an F4.5 mirror which in practice they are. An normal 8in scope made with this mirror would be pretty good - it wont be at F4.5. In practice more tilt is needed to get the field of view outside of the tube. If unobstructed reflecting telescopes is searched or tilted component telescopes lots of ideas can be found. They tend to be aimed at small field of view work and often have relatively slow F ratio's. One has spectacular resolution over a very wide field until it's realised that the mirror positions have to seriously limit the field of view. They have lots of catches including the need for accurate mirror radii and very careful setting up. The design looks like it has a mirror support anyway - that will result in 2 spikes. Might be worth mentioning that 3 supports would generate 6 that's why 4 are usually used. 1 will generate 2. I did own a scope that used an oval tube to support the mirror and the focuser. The far side of the oval was on the far side of the 2ndry mirror and formed it's support. That minimises the effect as well as the diffraction spikes are spread out over an area just as they are the wavy vane way. An oval is a lot easier to do. John -
  2. Optics of the Cassegrain Telescope, Robert L Waland ISBN 0-8062-3393-1. Carlton Press I happened on it by accident on holiday in the Scotish boarders. The seller reckoned it came from his estate sale so was his own copy. Pass on that aspect. It's bulked out a bit with calculations and the readings taken on a couple of mirrors. There is a chapter on figuring a 60in F2 for NASA and another on a 60in F4 precision parabaloid. There are some plates in the book but very poor black and white printing. Info on the design of all cassegrain types. The way he did it is pretty simple. Grind and polish to the curve needed at the margins and then grind and polish the central curve using a spherometer for the measurements. It's obviously going to be easier on larger mirrors. This could be done with tiled tools by removing the outer edges once the 1st rad is done. The central area is polished with full sized lap and was ground with a tool of about 40in diameter for a 60 in mirror. After all of this the usual small laps are used. It makes sense to me but I doubt if the central curve could be kept more or less where it aught to be by hand. As some one asked I've added some detail as I would guess that the book is going to be hard to find. There is a bit about him here http://www.futuremuseum.co.uk/collections/people/key-people/science-invention/robert-waland.aspx I don't want to hijack the thread but thought it might be of interest to some one trying to make an F3.3 mirror which is going to be tough as he says. I should add that though the forum might show me on line I may not be. I tend to retain a lot of tabs on my browser. John -
  3. Take that back. I've nosed around a couple of times recently because of something I want to make and a different search term bought up these 2 http://www.atmpage.com/epfocal.html http://www.jim-easterbrook.me.uk/astro/ep/ John -
  4. The only info I have ever seen on the subject is here http://www.cloudynights.com/topic/19323-focal-plane-in-eyepieces/ There may be more info about on the same site. John -
  5. I've been aiming at sub F3 for a cassegrain. All sorts of problems including doubts if can actually make a corrector I have designed for it. I came across a book by a Scottish optician who did some work for NASA and others on fast mirrors. He does his sums and works out how much glass needs to be removed for figuring and if more than 0.001 inch he grinds it off. This might involve more than two tools. I feel that this could only be done on a grinding machine with some sort of stroke arm. He basically reckons he's describing the old way of doing this sort of thing in case some one needs the techniques in the future. Big mirrors by amateur standards. I understand he upset NASA by making an IR mirror way more accurate than it needed to be in case some one wanted to use it for something else later. Not sure on that point. He did his tests with Foucault tester mounted on a lathe cross slide and pins on the mirror but he also checked for smoothness using a method he puts down to Dall. An achromat, moved around to null the mirror to spot lack of smoothness and ripples etc. I have used the wire form of the Hartmann test and reckon that is more fool proof than judging shadows in masks or via pins but I don't do much of this sort of thing. I have been pretty active on cloudy nights at times and there are all sorts of ideas about measuring mirrors but when they are ray traced they don't look so good as the set up needs to be rather accurate. Depends what some one wants to finish up with really. He also talks about mirror grinders I get a vision of one that follows a template at times surrounded by polythene curtains, water jetted with some form of lubricant. He points out that glass dust is good at blocking drains so needs a settling tank etc. Never tried his method but it does make sense but pass really. If I ever do one again I'm pretty sure it will be a Dall Kirkam as per some on CN. Put all of the problems together and it makes a lot of sense. Whoops - meant to add that is a pretty amazing piece of work. John -
  6. If you are using a truss scope somewhere that isn't a really dark site the mirror box needs to be long enough to stop ambient light from reaching the mirror, even light from the sky, same for the 2ndry box. If you do make it be sure that any light coming through the trusses doesn't interfere with the view when it's reflected off the mirrors. Also at the open end as light can get into the eyepiece if it isn't long enough. With something like a 10in F4 or so you might find that the trusses finish up rather short so more of a two piece scope. Mine used a solid tube but a wrap round type cover could make long trusses feasible - maybe black out cloth or something similar, heavy gauge black polythene with a cloth wrap 1st as it's reflective. Sonotube makes a decent tube - carboard. It can be painted and wrapped with plastic film or what ever. A quick look in the UK came up with this site http://www.crayford-tubes.co.uk/#industrial-tubes Another route might be to ask some of the larger concrete suppliers where they can be obtained from. They are used as formers when casting concrete. I have owned a Meade that used this. It works out rather well. Pity about the mess they made of the 2ndry. It should have been a bit larger. John -
  7. I would suspect buying a used 10in and then building yourself using the bits would be the cheapest option. You get the major bits that way and wont have to worry about making mirror supports at both ends which can be made but ....... It might even be the cheapest option if you want to refigure the optics. There used to be people around that would make and sell mirror sets at any size. I guess the only option other than some one who does it part time now is to ask Orion Optics. If I ever did a dobsonian again I would forget some conventions. Maybe bolt ball bearings to the mirror box rather than messing about with teflon - wooden disks etc. I have also always wondered about the basic ideas shown in Texereau's book for his standard scope - some have used the same ideas for an equatorial. Fork mounts have a lot going for them when it comes down to weight. I'd need to think about where the eyepiece might finish up but I'll leave that to somebody else but if based round a used scope the mirror box is likely to be round. John -
  8. I may as well elaborate on the x-y knife testing for if some one wants to measure rather accurately. It's mentioned in one of the ATM books. It uses a mask but treats them as separate small mirrors and a double edged knife its used - a wire in other words. Each of the small mirror produces it's own cone of light, It's a little like the pure Hartman test. With that the path of each beam used to be recorded on film. Some might be doing it on dslr's now. Each beam produces a line in the final image. Take 2 shots from different distances and the profile of the mirror can be calculated. With the wire test the wire is traversed across the path of the beams and viewed with an eyepiece so it just has to be moved back along the axis by a set distance., I used a simple 10x loop. When the wire is exactly centred in one of the beams diffraction patterns form on either side of it so the position is adjusted until these are symmetrical. The original article reckoned that a 1in diameter hole in a mask is accurate to 1/100 wave in relationship to a sphere so they test rather well this way even on fast mirrors. I can't remember what the accuracy claims were for the test as a whole but it really does need a delicate touch on a real micrometer spindle to get the diffraction pattern to be dead symmetrical. The maths to convert to normal Foucault readings were in the article. As I couldn't get the type of lens Dall suggested I used one out of a 5x eye loop and knocked one up an open one on a piece of square bar illuminated with a pin hole over a LED. The diffraction colour effects were spectacular especially in the mirror itself but I could see that the mirror was smooth and that the distances were about right. I used perspex to hold the lens which is probably explains the colour effects. I've since read about an professional optician using an achromatic in a similar way but more as a smoothness test to spot ridges etc on very fast mirrors that might even need some of the figuring grinding in. With this one the lens is moved to null the mirror or at least get a gradually changing shadow on it. Both tests need decent slits. Texereau describes a good way of making them but various blades can also be polished. John -
  9. Might be of use to some people. While nosing around for a diy eyepiece calculator I came across this collection of SCOTCH mount links no idea why it's called a barn door. I have seen some barn doors but never a type 3 etc. http://www.astronomyboy.com/barndoor/links.shtml The few I tried are still live. Not sure how it will work out photographically but the pole star round here is a problem at times. For visual use I set the wedge according to the scale on it. Orientated the mount with a compass taking care to level the platform and then used star drift to get the compass alignment correct. It may be possible to do the same thing with a camera using magnified live view. It doesn't take long with a bit of practice. In fact after a while I forgot the compass and lifted the lot up and turned it to get close. I still use a compass and level to align go to mounts. Maybe every one is aware of the star drift method but with all of the aids around now maybe not. I'm pretty sure Meade made a finder with a built in polar alignment reticule at one point. I don't think anyone does now, just polar scopes. I'm a little dubious about comments like just look along the hinge and line it up especially for exposures that need a more complicated scotch mount. Those do reduce the need for rewinding though. John -
  10. I did buy a very large bucket of resistors off Ebay. 1% and if I really need that I will measure them - just in case. It was time to replenish stocks again or more fill in the gaps that the last lot left really. This outfit has been reliable for me as well, all sorts of odd things show up every now and again. They still look pretty good on packs to me. I haven't bought any of those for some time but would do so again. http://www.greenweld.co.uk John -
  11. RS Components are probably your best bet but things sometimes come in packs. Ebay is another but companies like RS do declare what they are actually selling. For instance I bought a lot of variable voltage regulators off RS made by a Chinese company. The spec is nothing like the ones from the main manufacturers poorer in all respects but will do. On Ebay there is always a chance that items like this are being sold unless they actually state who made them I still wonder then at times. Farnell is another alternative and Rapid. It's no problem opening an account with any of them. Some years ago I used to wonder why my analogue designs worked at work but not at home then I found if I bought the bits from the right supplier they worked either way. That was when certain shops sold stuff really cheap. Now I think it's more a case of who actually made them rather than them being rejects. John -
  12. While looking around for some info on something else these popped up Using a crudely modified servo as a geared motor for focusing - there are plenty of source of information on the crude method http://emediadesigns.com/focuser/ And a page which I was looking for doing the same thing to a servo but retaining the servo action. I think he is being a bit OTT matching the resistors that accurately. https://learn.adafruit.com/modifying-servos-for-continuous-rotation/overview John -
  13. Reading the other thread and noticing mention of timing belt drives I fixed a variable magnification microscope head that used these to move bits about. They used a plastic sector on the end that was being turned and it had broke in places. I removed it and rolled a thin cylinder of epoxy putty, rolled it around the part that was being turned, placed some cling film over it taking care to keep it smooth and then pressed the belt into it and smoothed out the edges with my fingers. When it had gone very firm I removed the belt and cling film. It worked well. I took a bit of care that the belt teeth weren't pushed down hard onto the part that was being turned but found it was difficult to displace all of it which would have weakened it too much. There's not much chance of making a full diameter timing belt pulley with all of it's teeth like this, just a sector. I used the epoxy putty that £ shops sell, a packet with several small rolls of it. Milliput is reckoned to be the best and as they say it can also be drilled and tapped. It's useful stuff to have around. It also sets under water - so no need to worry about it getting very wet. No guarantees etc and perhaps it would be best to try it on something that doesn't matter first. I didn't but me £300 2nd hand head was useless anyway. It now works perfectly. Due to the cling film it more or less turned out shiny. I had the belt on and off several times while pressing it in to see what was going on. Where there is a lot of load it might be best to wrap a slab or something similar of the stuff part way round what is being turned to increase the area that grips. There are other options, maybe even make it and then fix it in place some how. John -
  14. I'm a heathen. The link can help to see if this could do the job. I believe there is some sort of tape available as well http://shop.pimoroni.com/products/electric-heating-pad-10cm-x-5cm?utm_medium=cpc&utm_source=googlepla&variant=411771581&gclid=CMyEtJyF6cMCFasBwwod6FkAoQ Next I would wonder about a metal canned power resistor fixed to some soft aluminium that can be hand bent round some of the the dew shield - it's metal so heat will spread. Last as it's probably hard to find rather thin stainless steel wire but it might be possible to do it with mig wire. I did make a 1m foam cutter once out of that powered via a transformer with a dimmer switch on the mains side but can't remember the voltage, it wasn't that low. It was thicker than 0.6 mm as well. Plus I might solder some resistors together. A metal resistor would probably be ideal for smaller bits if needed. John -
  15. That's great Nebula - One of the reason I started this thread was that it might produce more information in one place - in this thread. I hoped for more links actually. Where I said looks like least painful I should have added for me as I want a completely portable set up as round here I will take some excellent light pollution shots. The set up I linked to uses a web interface which means that many things can be used to drive it that also have long battery life. It also looks like the 'pi B's may be able to handle a sensible amount of USB current load. Times on a model B don't look too bad but I am not sure which one has been used. It's also not clear what the astrometry.net software is using to do it's sums - multiple cores or would the FPU and one be better etc. Makes me wish I had done work on what is essentially PC software but I have only worked on automotive ECU's generally in native assembler and some very boring after a while application software a long time ago pre windows. I'm also starting to try out an INDI set up using a netbook for Kstars / Ekos but I'll bet I will need to ask a question about the feasibility of just downloading jpg's from the camera. If you have read all of the posts you might have noticed that I am concerned about getting a decent bandwidth on the connection. That shouldn't be needed if raw remains in the camera. Wifi here is Rubbish. There is too much on it and it doesn't go very far either. I haven't stuck Kstars on the netbook yet. I aught to know what should go on an Atom processor but installed 64bit OpenSuse and then 32bit. The 32bit (legacy) seems to be noticeably quicker. Also that OpenSuse may be a touch quicker than Kubuntu. It's a lot quicker than windoze 7 to the point where an application is actually usable and the system type applications and browsing run with some snap once they are loaded. It just has 1GB of ram as so far I have failed to get the cover off the thing. KDE on 13.2 seems fine though, not even bitching about too much work to do on the graphics - it probably will at some point. Unfortunately it uses a 19.5v supply. That's rated at 40w. Some one on the INDI forum commented that they do use this sort of set up but that it is slow. What ever I do will take some time as I am starting from scratch. Even a new mount. I just like to have a good idea what I will do before I get round to doing it. At the moment I suspect INDI will be a fall back for if the other way doesn't work out. John -
  16. I'm interested in trying to do it with model aircraft type servo motors but not sure about the torque needed. Not sure on that point with steppers either. I need to measure it. From memory steppers are driven by a phased pattern on the 4 wires but I have no idea what the arduino driving boards need so can't comment really. Patterns can be generated by having an array of them and stepping through it at the rate you need to switch at. The processor does have more than one timer that could do that via interrupts or I suppose a simple loop could be used. Has to be from memory as I haven't had an interest in them for a long time. The servo motors are driven by a square wave and altering the mark space ratio sets the position the arm on them goes to. There is also a type that rotates. Some people convert the arm type to these, there are details about on the web. The mark space ratio then sets the speed and direction. The control can be directly connect to a pin on the processor. I haven't fully looked at the data on the processor yet and these things need precise timing if they are to stay in position. The digital ones usually have something like a 2 uSec dead band - any more than that and they will move. The other problem is hooking them up to the focus. I have seen one rated at circa 10 kg for about £9. If one is firmly hooked up to a focus knob turning it by hand is likely to be hard work. They use small electric motors and a gear cluster. It's also possible to buy servo testers - set them with a knob but they usually have one or two other facilities as well. Don't quote me but it's probably possible to take a servo motor apart remove various bits and pieces and just drive the motor - nice compact motor and gearbox set up. I assume the 10kg is at the end of one of the arms they usually fit. Like most things though it's probably an OTT rating and might let the smoke out. All electrical things really work via smoke because once it comes out they don't work any more. On the other hand if used as servo motors they will apply power to maintain position or rotational speed on that type. John -
  17. There is some info on another way of making the heaters here http://www.dewbuster.com/heaters-330ohm-resistors.html I found it useful because I had no idea what sort of power is needed. 330 ohm resistors at 12 v will dissipate 0.436w of heat and take 36ma each. The only thing I would warn people about is the power rating. I used to work at a place that had rather a lot of resistors that they couldn't use as they had to move to metal film. Free to take. I gained an impression of what a 1/2 w resistor could take from these. Later I bought some of ebay and had a shock about just how hot they got. Seems there is an industrial and a commercial rating. I would make sure they are 1/2 w and not go lower. These for instance will run at about 70C above ambient at 12v continuous http://uk.rs-online.com/web/p/through-hole-fixed-resistors/0148382/ John -
  18. Any particular reason for using a nano rather than a uno ? Curious as I intend to do one for 2 channels but using model aeroplane etc servo's rather than steppers so bought a uno but may have missed something doing that. Bit held up at the moment as I need to upgrade my Linux to load the software I need. I've just had a hell of a game installing it on a netbook in case I get tempted to try the INDI approach to telescope control. John -
  19. I doubt if you will have the problem but I came across 2 scopes where the entire mirror wasn't being used. One by design from Meade I suspect plus some one trying to collimate an offset 2ndry which can be tricky. The other one was down to the focus tube not being square to the tube. I fixed and checked it by removing the mirror and replacing it with a paper target. The 2ndry was fixed on this one so that aspect wasn't a problem and it did line up. I then literally bent the focuser to centre it on the target which just deformed the metal tube that the scope had or bent it back to where it should be. I found this was more effective than trying to judge that the entire main mirror was centred in the focus tube. The same sort of thing can be done by sticking a dot exactly on the centre of the main mirror but I didn't have any adjustment interaction problems by separating it out. John -
  20. I think I have concluded that this option is the most painless or will be in the near future http://sourceforge.net/projects/iastrohub/ There is a link to a cloudynights thread with lots of discussion on it. Why will be ? Currently it's built round a TV dongle. The problem with that is that they are likely to change periodically for all sorts of reasons - different hardware in them for instance. This could mean that people may buy one expecting it to work and finding out that it wont. Anat the author is very helpful with this aspect but is now looking at it running on a 'pi which as a platform should be a lot more stable. The Odroid is another possibility but they produce new products more often but I suspect if they wish to succeed they too will have to offer backward compatability and a stable platform. One of the advantages I feel this set up has is that it just downloads jpg's from the camera, That should speed things up a lot even with limited bandwidth. These days cameras usually put 9 stops into jog's often with settings that will add even more. I feel that is sufficient to judge a shot and helps a lot in the area of needing watty pc's around at all. Ideal for a portable set up. John -
  21. Maybe they can't afford HP servers. The install on the netbook went ok. It seemed to get a little confused because Linux was already on it. Had to hit a create partitions button. Previously it's always proposed one. Boot may be a bit slower than Ubuntu but the install dvd was still in so it looked at that twice. The install DVD offers several options including boot from hard disc, a safe mode and memory tests. It might also be a touch slower in all respects but it's 64bit and I loaded 32bit Kubuntu. I feel OpenSuse still has an advantage in that area. It will load all of the 32bit libs as well as 64. Probably not so important now but some useful old app might need the 32bit stuff. It's currently doing a mass of security updates. That will take some time as 13.2 has been out for a while now. John -
  22. Yeh - I have used a number of shells over the years. Old hat really so I don't unless I need to. I even have been known to use the root desktop. They covered that with images of bombs once and severly crippled it on KDE. I moaned on the mailing list and the following release was as it should be. After all root should know what not to do. I've not had the need of late so no idea what it looks like now. I was pretty fed up with windows and while walking around PC World noticed a nice thick big box with Suse on it. It even had 2 thick books in it. Bought it and over the next say 6 months stopped using windows at all. I later bought at least one upgrade from the same source and then found I couldn't so downloaded OpenSuse which is free. I have to upgrade now - can't load the arduino package thanks to dependencies. It's not much of a problem really except that I still run Kmail from KDE 3. I think that option is still available for OpenSuse 13.2 by searching for kdepim3 on their build service. Just checked and it is. Nothing compares with the old Kmail. Being able to run KDE on that netbook surprised me. I did find some figures showing % ram and cpu usage of all of the desktops. There ain't much difference according to that so I thought I would try it. I feel it's more usable than win 7 and might be fine if I fitted more ram. Next thing will be to install OpenSuse 13.2 on it and have a look around. They have being having problems getting people to maintain YAST but I think it now uses the new console mechanism. They have added a nice feature there - type a command and instead of just throwing an error if it isn't there the suggest running cnf what ever it is, command not found. That searches the repo's and generates a copy paste line to install it. To be honest I suspect OpenSuse still wins out over others because it's widely used on servers. Even HP qualify their stuff on it. Debian is an alternative but installing that isn't easy especially for beginners. OpenSuse is graphical in a similar way to Ubuntu which makes that aspect easy. Given the choice I would run ARCH but I can't spend the time needed to learn to install it - more of a will to really. I have a PC so that I can use it and that's that really. When I install it on my main machine I hope it still offers import partitioning but will have to make notes just in case. Odd system writes are directed to a small partition on one of the discs. I run an SSD but nothing ever writes to it unless I add or update software. My home is on raid, swap never used but on another disc and one other I use as a sort of back up at times. It seems the install will just update now but I'm not sure if I should trust it. I'll see what the forum thinks about that. There are several very helpful people on it. Out of interest this is the easy to use part of the build service. Many distro's use it in other areas. It's where updates come from when packages move on from the original release long before the normal update service offers them - if they ever do, I use several that aren't included in the distro anyway. http://software.opensuse.org/package/kdepim3 John -
  23. I didn't really look but did wonder about that way of doing it. I also suspect USB can cope with power at both ends. Might be wrong about that but I did have a machine where one USB hub turned off at times - no power on it - and could only get it back up by repowering the machine. On Linux. I just loaded Kubunto into an HP Mini 200-400 netbook that did have Win 7 on it. 1.6GHZ processor and 1gb of ram. I ran it without installing initially from a DVD. Not really enough memory for that but it tried and was semi functional without any crashes. Pity they don't offer the choice of nicking a bit of disk space. I then hit the install button. Had a surprise here. It gave me 2 choices really. Use up the entire disc or set partitions manually - no help which I suspect some would need and no apparent way of leaving the machine in a state where it would dual boot. Sort of all or nothing approach - sad. I'd guess there is info on the web on how to get round that. To just try it on a machine like this one it might be worth loading it into a USB stick but really more memory is needed. Having installed it boot up is slow but the desktop is usable as soon as it comes up. Application loading is also slow but functions well once it's up. I'd say overall quicker than the Win 7 that was on it. Much snappier once an application is showing. Can't expect to run huge apps really but from experience swapping isn't usually that bad as the core parts of Linux can remain in memory. I then for my own interest looked at software installation. Kubuntu/Ubuntu do have a graphical interface for this. I tried a search for Kstars - not there, then a package called Arduino, provides another way of programming them - not there. I then tried some topic searches such as dev and compiler and gcc - no good. I'm comparing with YAST ( Yet Another System Tool ) in OpenSuse. That is way way better. One of the problems with open source is knowing what is about and what does what. A simple search for AVR would bring up all related packages in that. Kstars - well it's under the educational section. More importantly when compiling source the header files for any library file that the applications uses are needed. Easily found in YAST. Ok Ubuntu/Kubuntu can use apt-get and there are loads of packages that can be obtained that way but I firmly believe that these days there really shouldn't be any need to work at that level in a console at all / maybe occasionally. Some people do love it though. So for me while I might install a pre built Ubuntu for a server etc no way would I use it as a main work machine. The other difference is install. OpenSuse comes with over 4gb on the dvd. Ubuntu 1.1. Many of the things I need are on the OpenSuse distribution. Not sure if they will offer a dual boot install, This may have disappeared on all of them due to the availability of VM's. I'll post a useful link on Arduino programming. I fancy an electronic focuser similar to the one shown on the INDI pages but also with the opportunity to use it manually locally. https://www.cypherpunk.at/2014/09/native-assembler-programming-on-arduino/ The code posted shows a cardinal sin where I come from - only one of the interrupt vectors specified (reset) which means it will crash if one happens to pop up for some reason. Tedious filling all in but can be used again on any subsequent applications. I was dismayed when I saw the method of using GCC on Atmel's pages. There is no way I would write code like that. My way is what is shown in the link - pure native assembler. Much depends on what is wanted and there are plenty of links off that page at various depths that go through several ways of doing it even mixed with C or C++. John -
  24. Probably the obvious way taking power from the plugs at each end. anyway http://www.corning.com/opcomm/OpticalCablesbyCorning/products/USB-3.Optical.aspx Horses mouth - they make them. Semiconductor manufacturers are also producing truly tiny signal conditioning chips, much like the normal USB cable length booster components only a lot smaller. In the case of USB 3 I would guess they are essential at much shorter cable lengths. I keep looking for a USB 3 sata hub plus card reader. Seems to be thin on the ground. I'd guess cable length is why. Dongles = no leads, those are around. John -
  25. There is an American "shop" that sells cameras and such like BH ???????? can't remember the rest. There are some optical USB 3 cables on there. Getting on for $200 from memory. It's a natural way for things to go. Actually fibre optics have their transmission line aspects as well. Single mode and multi mode. I see single mode as being like microwaves travelling along a tube of the correct size. Losses are very low and there used to be a device around where a probe could be moved along to measure voltages/current available at specific points . They varied sinusoidally so the frequency could be measured - effectively near perfect damping. I had to rent a line off BT once running from B'ham to Bolton. Complicated boxes at each end and running at an amazing 1200 baud - read bits / sec. One of the faster ones at the time. Some one had a VT100 on the other end running a stock control application that was in a DEC mini in B'ham. John - PS there is a bit on waveguides on the wiki https://en.wikipedia.org/wiki/Waveguide -
×
×
  • Create New...

Important Information

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