Why do telescopes show things upside-down (and sometimes mirror-image as well)?
When you look through a telescope and see a gorgeous coloured double star, or maybe a sparking globular cluster, you can’t immediately notice that what you are seeing is upside down, and maybe (depending on the telescope and its setup) a mirror image as well. But, if you look at the Moon and compare it with what you can see with your unaided eye, you will quickly notice the difference.
So what’s going on?
I’ll explain it fully in a moment, but first, think about regular optical aids, like a simple pair of binoculars – there’s no image inversion there, so why do astronomical telescopes have it?
Well, most designs of binocular include special prisms (complex glass shapes) which compensate for the inherent image inversion, providing an upright image. This can be done for astronomical telescopes too: it’s possible to buy a such prisms, variously called an “erecting prism” or an “amici prism”. We have one here at GersAstronomie, normally fitted to a scope we often use to look across our wonderful landscape.
But these prisms involve quite a lot of glass, and thus absorb a little bit of light. And if you’re straining your eyes looking for really faint objects, that’s the last thing you want. So, usually, astronomers put up with upside-down images (you do get used to it, actually) in the interest of collecting as much light as possible.
So, let’s get down to a bit of simple optics. Here’s a picture showing what’s going on.
Ignore the big red arrows for now; they represent to object being viewed. Look instead at the parts representing the telescope and the observer: the primary lens, or objective, is labelled “p” (it’s true that many telescopes use mirrors instead of lenses, but the same principles apply otherwise they wouldn’t work as telescopes). It’s also true that the primary lens is never a single simple lens as we see it here: it’s a combination of 2, 3 or more (or many, many more in the case of camera lenses). The eyepiece, or secondary lens, is marked “s”, and the same caveat applies, but even more so: premium eyepieces these days may have 9 or more individual lens elements. But they still behave as the simple, single lens we show here.
Finally there is the representation of the eye of the observer, labelled “o”.
The horizontal dashed line is the optical axis of the telescope, and the vertical line marked “f” is the focal plane: it is here that the parallel light rays from a distant object converge to a focus. Think of it as the flat surface of the sensor chip in a camera, or the film in an old-style camera.
Now it’s time to think about the object we’re looking at, which in this case is the red vertical arrow, marked “A”. In our astronomical case it’s very far away, so the light rays from any part of it (a specific star in a star cluster perhaps, or in our case, the tip of the arrow, indicated by the yellow circle) are parallel to each other. These rays are coloured yellow.
Other rays, from a different part of the view, say another star in a cluster, or the foot of the arrow (shown by the purple circle), are parallel to each other, but not parallel to the first set. These rays are coloured purple in the sketch.
What we see in the picture is that the various sets of light rays converge in the focal plane to produce an image labelled ‘I”, which is upside-down compared to the original. If you would put a sheet of paper there, or indeed the imaging ship of a camera, this image will appear perfectly real, even though upside-down!
In the case of a normal telescope though there is an extra step. Instead of trying to capture this image at the focal plane with a camera chip, we use an eyepiece, which is in effect a powerful magnifying glass, to look at the image directly, and magnify it.
The focusing mechanism of the telescope can move the eyepiece horizontally (in this sketch) so that the image is in the focal plane of the eyepiece. In this position parallel rays from the image “I” converge to a focus in the eye “o”.
As the eye sees it the light rays from the tip of the arrow arrive at quite a large angle compared to the arrow’s base (which is on the optical axis) and so the virtual image, “V” appears to be very much magnified compared to the original, “A”.
But it is upside-down though. There’s nothing wrong with the optics: it’s just the way it works.
Over the years astronomers have just got used to it. Indeed there are many published maps of the Moon which are inverted so that they appear the same as the view in the eyepiece.
To give an idea of just how universal this is, here is a picture of the view through one of our camera lenses. Yes, the image is upside-down, so why isn’t the photo coming out of the camera? Well, that’s because the camera software turns it over.
A more interesting question is, what about our eyes? Well, it turns out that they also provide an upside-down image, but our brains switch the image the right way up! When I was in school I was told of an experiment where volunteers were given special glasses which inverted their view of the world. After a week or so, they were seeing things the right way up. The human brain is capable of a remarkable amount of signal processing!
But there is a further complication (as if what we have just described was not complicated enough!).
Everybody who has looked in a mirror (and that’s probably all of us) knows that the image you see in a mirror is reversed: if you wave your right hand, the reflection in the mirror is waving their left hand. It’s called a “mirror image”.
If you look at a mirror image in another mirror, guess what: it reverses again so the image is the right way around.
Now mirrors quite often find their way into telescopes (I’ll have to write another blog one day on different telescope designs). The earliest example of this was the reflecting telescope invented by Sir Isaac Newton to overcome the problems inherent in the lens-based telescopes in use at that time in the 1600’s. His telescope design, now called a Newtonian reflector is more commonly found nowadays in a very popular design known as a Dobsonian (or “Dob”). John Dobson’s innovation actually had nothing to do with the optics but was all to do with the mount; the telescope itself is still a Newtonian reflector.
This telescope design has 2 mirrors and so the mirror-images cancel each other out and the view is correct left-right, but still upside down. This argument is also true for all designs which have 2 mirrors, such as the very common Schmidt-Cassegrain scopes like our Meade 12-inch in the observatory.
But there’s yet another complication!
Telescope designs where you look in from the side, like the Newtonian reflector, are one thing, but if the view ends up being on-axis, that is, if you have to look through the tube of the telescope to see the object, then there is a risk of back or neck problems.
Think about it for a moment. A sailor looking through his telescope at the horizon is looking horizontally. And when we look at astronomical objects low in the sky we have a similar situation. But when you point the scope up towards an object high in the sky you will have to have the flexibility of a limbo dancer to crouch down low enough and tip your head up to see it properly.
It’s not comfortable, so astronomers developed, long ago, a device called a star diagonal, which is essentially a mirror set at 45° just before the eyepiece. This diverts the light rays by 90° and so, instead of risking a dislocated neck, the observer can comfortably look down into the eyepiece.
But, it is a mirror. So, if it’s used with a lens-based telescope or a mirror-based one, it will, once more, add the left-right reversal, mirror image effect. Note that these star diagonals are almost never used with Newtonians as they’re not needed to ensure a comfortable viewing position.
We do have one fitted to our big 12-inch Meade scope, so the upside-down, mirror image view is standard for us.
We have got used to it!