Making a Refractor

All books on telescope making start with a comparison of the various advantages and disadvantages of different types of telescopes and then come to the conclusion that a Newtonian is the obvious way to start your telescope making activities. This is undeniably true but in the comparison the difficulties attributed to making a refractor are, I believe, greatly exaggerated. The purpose of this article is to show that making a refractor is very much easier than most people think and to encourage would-be telescope makers to "Please Consider".


To summarise the advantages in making a lens as opposed to a mirror let me say that:

  • All surfaces are spherical. A refractive surface is four times less sensitive to figure errors than a reflective one.
  • You can test the lens as you go and do not need to send it away for coating. What you make is what you get and if any slight imperfection starts to bug you in a few months time you can always fine tune it a little more.
  • By oiling the space between the two lens elements you can effectively eliminate small surface errors and scratches from those surfaces and any figuring need only be done on the front surface. I suppose what I am really saying is that you don't have to be a crazed obsessive to come up with a really good telescope and that by applying the same standards of accuracy to my refractor as I have to my reflectors I have come up with a result that really surprised me.


I first started my project by contacting fellow ASSA member Ian Robinson who was able to supply me with glass blanks for a 105mm achromat lens for $120. That's the first myth debunked! (The major expense in my telescope turned out to be for the focuser.) These blanks consisted of a disk of dense flint glass about 15mm thick and a disk of crown glass of similar thickness. This crown glass is officially known as BAK 1. (It means nothing to me either!) Had I wanted to make the telescope an f/15 instead of an f/12 I could have purchased blanks for $95. Apparently BAK 1 has better refractive qualities and is more suited to shorter f-ratios and is a tad more expensive. Ian also came up with a set of radii for my f/12, generated in a few minutes on his computer. Second myth - you don't have to be a mathematician. The curves went something like this. No 1 radii = 650mm, No 2 = 430mm, No 3 = -430mm (concave) and No 4 = 0 (flat). All this added together gives, for some reason, a focal length of around 1200mm.

Grinding: I started grinding the flint on top of the crown in exactly the same way you would grind a mirror with the tool on the bottom and as the curves we were looking for were the same, (albeit one concave, one convex) in no time at all two of the four surfaces were ready to polish. In the past I have tried to make my own flats for diagonal mirrors and have been dreadfully disappointed with the results so I was a little apprehensive about this stage of the process until I found out that for a flat to be successful as part of a lens it doesn't have to be very flat at all. In fact, if you hold a straight edge against the lens, hold it to a light and you see no gaps, it's perfectly flat enough to work very well. Three down!

It's best to leave grinding the front of the lens until last because if your radii for No 2 and 3 are not spot on you can adjust No 1 to suit. This happened to me and a quick call to Ian gave me the required length to work to. This curve required that I cut a tool from plate glass 10mm thick (available from Bab's Bargains, $5 per sheet). Luckily I have access to a lapidary saw so this was quite easy but I have cut this glass in the traditional way in the past.

Again, grinding the last suffice of the lens was just like making a mirror but with tool on top, lens on bottom and paying more attention to getting the radius correct. In general, grinding lenses of this size is a very fast process and in all didn't seem to take as long as grinding my first ever 6-inch mirror. One thing to watch for in grinding is stressing the glass by fixing it too tightly or pressing too hard. Make sure it is well supported underneath as it is relatively thin and prone to distortion.


One thing I haven't mentioned so far and which seems to loom large in any discussion on lens making is wedge. Wedge is where one side of a lens is thicker than the opposite side causing the two lens elements to be at an angle with each other and this affects the chromatic (colour) correction of the telescope. The thickness around the edge must not vary more than about 0.005mm. Having said that let me point out that eliminating wedge effect was easy and predictable, almost to the point of being a non event and should not concern anyone contemplating a similar project. To measure wedge you need the only thing remotely resembling a sophisticated testing device you will need for the entire project. I made mine with a second hand micrometer and a piece of plywood. It is very simple and I will willingly show it to anyone interested because it is beyond my threadbare ability to describe it here. Once you have decided which side of the lens is high you simply apply more pressure to that area and the wedge steadily disappears.


I did all the initial polishing using optical polishing pads (Coburn Optical Industries, Lonsdale). These pads leave weird zones but are very fast at getting pits out which is just as well because flint glass seems to take forever to polish. The pitch lap very quickly eliminates any zoning or roughness and the combination of the two speeds the process up no end. I used the same lap for both convex surfaces, just repressing for the deeper curve and the same lap for the flat and the concave. By taping the lenses together with a thick cloth between them you can support well and avoid any stress.

Figuring and Testing

This was the most surprising part of the whole project as it turned out to be so easy and straightforward, especially considering the difficulty I have had in the past when trying to parabolise mirrors accurately. I have actually come to believe that the Foucault zone test is some sort of voodoo ritual but I had no such difficulty testing a lens because the star test images are so clear that it is obvious what needs correcting. Ian's experienced eye was a great help in this area but believe me, at least in my (limited) experience, testing and getting an accurate figure is much easier than on a short focus mirror The first moments of real satisfaction I gained from this telescope were not seeing the clearest images of Jupiter and Saturn that I had experienced or even seeing an Airy disk for the first time but when I took it up to a vantage point in the hills one day and looked out at the airport and ships at sea just like with a "proper" telescope. It's portable, gives the best images I have achieved and, with an erecting prism, can be used for anything you like.