Cheap Solar Projector Telescope

Note:  This page is NOT finished, but you can get the general idea of what I’ve done from it now.  Last updated 6/22/04.

 

If you want to teach your kids a few things about astronomy or optics during the day without spending a lot of money, then this web page is for you.

I've enjoyed teaching my kids the names of a few constellations, and setting up a telescope for them occasionally, but if you've tried this you know it's hard to do without keeping them up late in the summer, or freezing them to death in the winter.  I was looking for something to do with them during the day, and I thought of looking at sunspots, eclipses, and the like.  There are only a few ways to view the solar disk safely, and most are either kind of expensive or very low quality.  What I've come up with seems inexpensive and of at least moderate quality.  Here are the methods of solar observation I could think of:

What I built was a very simple projection telescope that works well enough to see sunspots, only cost about $20, and was easy enough to assemble in one evening.  I was so pleased with the result, I thought I should share it with others, hence this web page.  Time for the disclaimer:

SAFETY DISCLAIMER:  Looking at the sun can cause blindness.  Looking at the sun through a telescope can cause blindness almost instantly.  Anything you do with this design that burns a hole in anything at all, including your eye, is not my fault.  The whole point of this design is to prevent anyone from looking through the telescope at the sun, or even toward the sun.

Several different optics professors told me "you can build a telescope out of just about any two lenses."  So, I thought I'd give that theory a test and ordered a set of lenses from Edmund Scientifics (item number 30404-14). Since they were out of stock last time I checked, these guys seem to have the same thing.  I also ordered a couple of the Edmund books on telescopes and optics which were inexpensive, but only moderately helpful.  I did get some good ideas from them though, as you'll see.  These lenses are 50mm diameter (about 2 in.) and there are three positive (convex) and three negative (concave) lenses.  When you're building a telescope there are two basic approaches:  two positive lenses, or a positive and a negative.  The objective lens (the one that goes toward what you're looking at) is the one with the longer focal length (which is a "weaker" lens) and the eyepiece lens (the one you put your eye up to) is the one with the shorter focal length (which is the "stronger" lens).

If you want to build a telescope to look through with your eye (obviously not for solar viewing unless you have a filter), the equations are really simple.  You need to space the lenses as far apart as the sum of their focal lengths (even if one is negative), and the magnification is just the ratio of the focal lengths

 

 ().  These formulas seem to display well in Internet Explorer, but not in Netscape.  Sorry about that.

 

You can ignore the sign on the magnification. If you use two positive lenses, it's called an "astronomical telescope" design and it has an inverted image.  If you use a positive objective lens and a negative eyepiece lens, it's called a "Galilean telescope" design, and it has an upright image.  With a negative focal length, the lens spacing is shorter, and the field of view is narrower.  I chose a Galilean design because I wanted the whole thing to be shorter, and the sun isn't very wide.  Inverted image or not isn't very important, since nothing in space seems upside down when you look at it.  I'll get into the math more later in case you want to modify my design, but this is about all I knew when I started.

I needed two tubes that would slide inside each other (you know, "telescoping!").  It turns out telescoping cardboard mailing tubes work really well for this.  I got mine at a university bookstore, but I think most office supply stores carry them, too.  Amazingly, I couldn’t find any on the web except in quantities of 25, like this, but I got mine for about $3, it was a 3” x 25” tube with metal ends, and you can telescope it to about 40” or so.  I pried the metal ends off with an old bottle opener.  It’s hard to saw a straight cut on these, so avoid that if you can.  I think the only real disadvantage of cardboard is that sliding them seems to make dust on the optics, but I can live with that.  It wasn’t really necessary to cut the mailing tube at all (although I did initially, of course), since it came in 3 pieces, roughly the lengths I needed.

 

To get the lengths for my design, I just experimented with the lenses and a yardstick to see what would work.  This was rather tedious, and I spent a lot of time later working on understanding the lens equations to be able to predict the behavior I wanted.  I’ll give more detail about that below.  Note one thing I learned was that the lens spacings are different if you focus on a light bulb on the ceiling than if you focus on the sun (the ceiling isn’t nearly far enough away to act like “infinity”).

 

The oatmeal canister idea came from the Edmund telescope book.  I put a piece of white paper on the bottom (don’t glue it down, it gets dirty easily, and that makes “spots” on the sun that aren’t sunspots.  I used a piece of black foamcore board to make a stiff mount between the oatmeal canister and the mailing tube end. Whenever I needed a circle cut out on the foamcore, I just drew a circle around the tube I wanted to match and then cut a little around the circle depending on whether I wanted a bigger or smaller hole. With no more precision than that, it’s amazing how well the thing worked out.  If I were doing the oatmeal canister again, I might use the lid end for the screen (so it could be removed easily) and cut the bottom out and do the foam core on that end.  I say this because the plastic rim didn’t hold to the epoxy well (that’s the part that broke) and although I have now epoxied over the edge of the plastic to get a better hold, it might work better this way.  The downside of making the screen removable, is that now you could more easily look down the tube into the sun (AND LOSE YOUR EYESIGHT), which defeats the whole purpose.  The upside is that you could replace the screen more easily if it got dirty.

 

A key element in the speed and success of this project:  5 Minute Epoxy.  This made it possible to just put the thing together, hold it for a minute and then glue the next piece, which made it practical to build the whole thing in one night.  It hardened faster than I could make the next chunk of foamcore.  I’ve gone to hotglue on the lenses so I can (theoretically, anyway) take them back out and remount them later or reuse them, but I don’t think it’s strong enough for the main joint, which I’ve already broken once.  The first version of lens mounts was done with epoxy, trying to keep the lenses untouched, so I made two outer discs to hold the lens inside, and an inner disk to attach to the end of the tube, and to glue the two outer disks to.  If I did this again, I would just make one disk with a hole smaller than the lens, and hotglue the lens to it.  The hotglue will come off the glass easily if you need to try again (I think), and there’s less chance of misaligning the lenses if they’re glued to a flat surface.  After I decided to use all three negative lenses for the eyepiece, I did a very different mount and hotglue, but I’ll deal with that later.

 

Parts List:

 

Design Math:

To do this design with different lenses, you just need a couple of things:  an equation to relate the screen distance and the lens spacing to the lens focal lengths, and an equation to tell how big the solar image will be.  All these equations will be approximate, since most by-hand designs are done that way.

 

The simple part is figuring the solar image size.  The sun is approximately one half degree in diameter.  If you magnify the image by 3x, you get an angle of 1.5 degrees.  If you convert this to radians, you can get the arclength which is approximately the image diameter using .  So, the image size should come out to about where solardiameter is the diameter of the solar image projected, mag is the magnification of your telescope (see equation above), and d is the distance from the eyepiece lens to the projection screen.  The 0.5 degrees is the angular size of the sun, and the pi/180 is the degrees to radians conversion.

 

The more involved equation isn’t that hard to derive, but it takes some diagrams to completely explain.  I’ll give the summary here and hope to put the diagrams in later.  See the hyperphysics links at the bottom of the page for some nice generic diagrams explaining the lens equations I used.  Assuming the objective lens is lens 1 and the eyepiece is lens 2 (negative focal length in this design), I get this relationship:

 


 

The lens spacing is just the distance between the two lenses, and the screen distance is the same as d above.

 

One more thing:  in the latest version, I stacked all three negative lenses together to form a “stronger” negative eyepiece lens.  When lenses are touching, you can add their reciprocal focal lengths (approximately).

 


 

For my design, I just did a spreadsheet with these relationships to figure out what range of tube lengths I could use.

 

Some assembly photos (soon).

 

Some Venus transit photos: 

Unfortunately, I still haven’t got a good system for producing well-focused images on the camera (there is a very narrow depth of field, and it’s hard to see the focus quality in the tiny viewfinder).  I am working on a method of getting highly focused images, but in this one you can see the shadow of Venus very near 3rd contact.  This was taken in Tuskeegee, AL, at about 6:03 am.  The image I saw on the telescope screen was much more “in focus” than what the camera produced.

 

Pictures:

If you open most of these pictures in a separate window (I’ll add the links later, I hope), they are higher resolution than what you’ll get on this page.


The telescope in operation.  If this looks precarious to you, well, let’s just say, I’ve fixed it since “the fall.”

 

The screen with opening, and a foam-core mount for the mailing tube end.

The objective lens mount.

 

The eyepiece lens mount, with the other two negative lenses taped on for experimentation.

 

What happens when there’s a tree between you and the sun.  You can tell the difference between the focus distance for the tree and the sun.

 

A focused image of the sun.  There is a blue fringe and you can’t see the sunspot in this image, although I could see it with my eyes.  I haven’t learned how to get the contrast and focus on the digital camera set just right yet.  It’s not quite round, because you can’t take a dead-on image from this device.

 

Here’s the SOHO satellite image from the same day.  The sunspot you see near the edge should appear in the image above at about “8 o’clock”.  http://sohowww.nascom.nasa.gov/sunspots/

 

Some links about telescopes and mailing tubes.  I’ll fix them up better later.


http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/lenscon.html#c1

http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/image.html#c1

http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/image3.html#c1

http://web.missouri.edu/~wwwepic/Telesc_Projector-MaryDiane_.pdf  (lesson plans for mailing tube telescope)
http://www.nasa.gov/audience/foreducators/informal/features/F_Build_a_Telescope.html

http://galileo.rice.edu/lib/student_work/astronomy96/mtelescope.html

http://www.solarscope.org

 

Copyright 2004 Wayne T. Padgett

Want to talk about it?  Remove the (nospam) from my address and email me at Wayne.Padgett@(nospam)rose-hulman.edu. I’d love to hear that you’ve gotten something similar working.

 

internet service providers

internet service providers