Exactly what constitutes the optimum lunar/planetary (L/P) telescope has been a constant source of debate in astronomical circles probably since Kepler introduced the positive inverting eyepiece and Galileo's friends disagreed. Over my lifetime I've seen many things come and go, I've even indulged in a few oddities myself (I remember that Gregorian); but the recent refractor love affair is perhaps the most interesting yet. Bitten by the bug and curious, as usual, I have designed and made now perhaps half a dozen f/15 refractor doublet objectives, including several oiled ones and an experimental glass liquid objective of 6" aperture, and even one 4.75" doublet out exotic KZFS01 that is a joy to use. I must admit, they produce lovely images, but at what cost? The tubes are quite long (the exotic KZSF01 job had to be f/18 to keep the internal curves reasonable) and with the exception of the exotic glass objectives, the secondary color can be obtrusive on bright objects and certainly ruins serious attempts a color estimation. When used with a color filter, planetary and lunar viewing is nothing sort of superb, and even without a filter, binary stars are wonderful - perfect, clean disks.
But refractors have a serious limitation, they require transparent glass of high quality and glass that even in the most common varieties, is expensive. And if they are made with the common, less expensive glass, the tubes must be very long compared to a reflector. As a result, people have come to accept the fact that refractors must be of small aperture. Some commercial refractors are made with exotic glass combinations and can be made rather short for their aperture, but they are still small (6" appears to be about the max for reasonable prices - and still multiples beyond the costs of an equivalent reflector) and the glass is wildly expensive in larger sizes, when you can get it.
Because refractors can be designed having all spherical surfaces, manufacturing can be mechanized a good bit, and given a nominal refractive index of 1.5, refractor surfaces need only be made 1/4 as accurate as reflector surfaces to produce an equivalent wave-front accuracy.
One of the things that has made the refractor so popular, and overridden the obvious limitations of small aperture, is that they happen to produce much better images than the typical reflector. Views through some of these instruments are really startling. The reasons for this are interesting and sheds some light on why the reflector has been given a bad rap over the past few years.
First of all, there have always been a separate set of standards for refractors as opposed to reflectors, the most obvious being the allowed small aperture and the other, the long tube in relation to that aperture. Reflectors, by contrast, are not permitted to have long tubes and are usually seen in larger sizes. This places the reflector in a disadvantageous position on at least five counts: 1, larger optics are more difficult to make and have perform to the diffraction limit; 2, larger optics are subject to the deleterious effects of atmospheric turbulence, which become pronounced after six inches in aperture and quite severe for high resolution work over ten inches; 3, short focal ratio Newtonian mirrors are subject to strong off axis aberrations; 4, short focal ratio mirrors are difficult to align and keep aligned; and 5, short ratio mirrors are extremely difficult to figure accurately. Put all these together (which you do in the typical Dobsonian) and you have a telescope that is likely to produce soft images. Put side by side with the smaller but optically superior refractor, the refractor image is likely to be much clearer appearing and the surrounding sky blacker.
Greed is a terrible thing and reflector owners seem to place more value on aperture than anything else and have a penchant for wanting (and getting) a lot of it. The Dob craze in many ways was a wonderful learning experience for the telescope making community. We learned to deal with and accept the alt-azimuth mount (but with a new name) and learned to build telescopes that could be transported and set up easily; but with this came some downers. We learned to accept even poorer optics than we were used to in the 50s and 60s. At least the venerable 6", f/8 was very forgiving of less than perfect correction of the surface. A 50% parabolized mirror would yield a very nice image in a 6", f/8 - actually diffraction limited; but when the Dobs came along the demand was up for f/5s and then f/4s, and in sizes in the 12.5" and then into the 16" and bigger range until we had 30" and 40" mirrors at f/4, and 2" thick. They certainly do collect a lot of light, and for their intended deep sky use, can be quite effective.
But, for serious planetary observing we need something quite different, we need a telescope that produces a high accuracy wave-front at the eyepiece but with an aperture of at least 8" and up to perhaps 10" and maybe even a bit larger. I specify "wave-front at the eyepiece" because I wish to emphasize that it is necessary to produce a telescopic system rather than just a telescope mirror of high accuracy. This is really the crux of the matter. The wave-front must be preserved intact until it reaches the eyepiece, and this demands an effective telescope design that solves a variety of problems inherent to the reflector. The refractor appears to be relatively free from those design flaws that plague reflector, such as spider and secondary mirror diffraction, tube currents, and, in larger telescopes, thermal equilibrium of the primary mirror.
Tube currents are by and large the most serious problem effecting the reflector and various proposals have been made to correct this problem. I have experimented with some of them and have found a few to be successful. The most singularly effective thing I have done to eliminate tube currents is the installation of a fan at the bottom end of the tube. This requires that the lower end of the tube be left open (which I recommend in any case) and that the fan be so arranged so as to blow the air out the end of the tube. This results in a smooth and continuous flow of air from the upper end through the telescope and out the lower end so that pockets of warm air do not develop within the tube and slowly work their way out of the top - destroying the shape of the wave-front as they go. An important secondary byproduct of this continual flow of air from the upper down through the lower end of the tube is that the air drawn in from the top is drier than the wet air likely to rise up from the ground during the normal process of convection. As a result, dew is less likely to form on the interior of the tube and surface of the optics. Of equal importance is the fact that due to the fairly rapid movement of air through the tube, the mirror, which frequently begins the observing session much warmer than the ambient air, more rapidly cools to a state of equilibrium. This is extremely important with mirrors 8" aperture and larger and becomes critical with mirrors 12" and up. Mirrors that sit in a closed tube (and frequently, and unfortunately, with the bottom end capped) have really very little opportunity to adjust to ambient air temperature and spend most of the observing session giving off their heat and contributing greatly to the slowly moving pockets of warm air that rise up the tube. Use of the fan has caused my 10" mirror to cool completely in approximately 1/2 hour. Prior to this, and with an open truss tube, the mirror took approximately two hours to reach full thermal equilibrium.
It is appropriate at this time to digress for a moment and discuss the desire on the part of many people to want to have mirrors made of exotic materials such as Zerodur and other low expansion substances. These materials have largely been designed for use in large aperture instruments where unequal expansion and contraction can cause severe distortion of the optical surface and a serious degradation of the final wave-front. In small mirrors, say 16" and under, this simply is not as much of a problem as other factors relating to less than full thermal equilibrium. What I mean by this is that a thermally unstable mirror in an amateur's planetary telescope offers problems that exceed those generated by thermal instability affecting the actual optical figure of the mirror. The more severe problems are those resulting from the discharge of heat in the form of warm air currents rising upward off the face of the mirror. Anyone who has conducted optical tests understands that any thermal disruption of air has a far more severe impact when it is close to the surface of the mirror than when it is close to the final image plane. Translated to the problem of telescope tube currents, this means that any thermal disruption close to the mirror is going to have a significantly greater impact than any thermal disruption further away from mirror along the tube. I have confirmed this by practical experimentation in such a way that can be observed by any telescope owner. To test the thermal stability of your telescope mirror at any time during an observing session, point the telescope at the brightest star you can find that is close to the zenith (one does not wish to be confused by poor atmospheric seeing). Remove the eyepiece and look through the eyepiece holder. You'll see the objective brilliantly illuminated by the light of the star. You'll also notice that if the objective is not in a state of thermal stability the surface of the mirror will appear to be crawling with a mass of squiggling, moving shapes. What you're seeing is the effect of non-homogeneous refraction due to small cells of air of unequal temperature. When the mirror reaches thermal stabilization, and the interior of the tube is free of tube currents, and the atmosphere is reasonably stable, the mirror will be equally illuminated. This usually happens after the telescope has been sitting outdoors for at least two hours. Using an exhaust fan I can achieve this effect with my 10" reflector an approximately 1/2 hour.
As to the design of the tube, I am a proponent of a type of tube frequently seen in Dobsonian telescopes. That is, a solid walled box at the lower end extending high enough beyond mirror so as to retard the formation of dew, surmounted by an open truss arrangement and topped by a solid walled end piece holding the eyepiece and diagonal. The top end piece should be long enough so as to keep any stray light out of the focuser barrel and to prevent the formation of dew on the surface of the diagonal. As a general rule, I would say that the solid walled end piece of the tube should have a length approximately 1.5 times the diameter of the mirror. And the mirror box should extend above the surface of the mirror by an amount approximately equivalent to the diameter of the mirror. The rest of the middle part of the tube is open. I have had no problem with body heat affecting the image quality. In fact, I have seen many more problems resulting from closed tubes than open ones. In my case, the use of the fan is then effectively confined to removing air from the mirror enclosure box. Just to emphasize the fact that I believe heat leaving the mirror to be the basic cause of internal tube currents, the use of the fan even in an open tube telescope such as mine has had an enormous impact on rapid stabilization of the optical system. Even with the open tube design I was initially plagued by tube currents that were entirely confined to the small mirror enclosure box. One can only imagine the impact of these currents as they work their way up a long solid tube. I vividly remember being a star party a couple of years ago with a fellow who had an Astrola 10", f/7 reflector which turned out to have excellent optical properties. At the beginning of the observing session just after sunset he was having trouble with distorted star images. When we performed a star test we noticed that the expanded Fresnel rings were flattened on one side. I looked at the bottom end of his tube and noticed that it was covered with a plastic end cap. I suggested he remove the cap. While he did so I continued to look through the eyepiece. The cool air rushed up through the tube and the stellar rings assumed a circular shape immediately. While this telescope had not reached total equilibrium, one cannot deny the beneficial impact of allowing air to move freely through a telescope tube.
The fan I use is a 12 volt muffin type purchased from an electronics supply store and operated by a 115 volt AC/DC power supply now so commonly available. As far as mounting the fan is concerned, the most advantageous method appears to be suspending the fan radially by three or four 1/4 inch, steel tensions springs running from the fan to the sides of the mirror mount or tube. This effectively isolates the fan mechanically from the rest of the telescope and I have had no trouble with vibration. I operate the fan continuously, though I have also sometimes stop the fan and then unhook the springs and remove it from the telescope so that no heat from the motor will find its way up through the optical path.
The business of how to mount the secondary mirror and how big it should be is another excursion into endless debate. And the efforts of these debaters can result in extreme mechanical concoctions of the most bizarre kind. Discussions with individuals who know much more about theoretical optics than I ever will, combined with my own, as well as their own, practical experimentation, has resulted in the opinion that discussions regarding size and support of the diagonal mirror largely over-rate the problem. The short end of it appears to be that any diagonal 20 percent of the diameter of the primary or less will not contribute in any significant way to the apparent degradation of the image, and that supporting the diagonal on thin, straight, metal strips is as efficient and unobtrusive a way as possible of doing it. Granted, curved diagonal spider veins are available for purchase, but what they essentially do is smear the diffraction more or less evenly across the field rather than concentrate it in a thin radial line. But once again, the business of spider diffraction, I believe, is one of those grossly over-rated issues. For one thing, spider diffraction is really not visible except when observing bright stars. When observing second or third magnitude stars I do not find that such diffraction spikes are visible. Certainly, they are of no consequence looking at the moon and really do not appear to be a problem when looking at the planets. The sum total of the impact of spider diffraction appears to be extremely tiny so that the only real problem is an aesthetic one when looking at bright stars.
Along the same lines, the pursuit of an extremely small secondary mirror can actually lead to problems far exceeding the unobservable deleterious effects of a moderately small mirror. What I mean by this is that a diagonal should not be made so small as to not allow for some area beyond that required for containing the complete bundle of axial rays. Some allowance should be made for fully illuminating a reasonable off axis field as well as accounting for the fact that the very extreme edge of the diagonal is likely to be turned down, as are almost all optical surfaces no matter how well fabricated.
Mirror clips that extend over the front surface of the primary mirror should also be avoided if at all possible. Practical observation has shown that these also produce diffraction, but this diffraction can be easily eliminated by simply using side pads as described in A Simple Mirror Cell. It is really quite essential that the edge of the mirror, the actual pupilary opening of the system, be kept clean and round with no jagged or discontinuous edge.
As regards investing in one of those high-priced, ultra-smooth focusers, I find these relatively useless except for photography or CCD imagery, were they become critically important. But for ordinary visual work the standard rack and pinion focuser is just fine. I have experimented with helical focusers and found them to be quite useful for very fast Newtonians but offer no special advantages to f/6 instruments and above. In fact, they are particularly clumsy to use when various people are observing from the same instrument, as each individual wishes to adjust the focus to his own needs. When observing with the general public they don't work at all. Once again, spend your money where it counts.
I have no special preferences regarding mounts except that they should be solid and track the motion of the objects as they cross the sky. Push and shove mounts may work well enough for low-power, deep sky observing, but high-power, planetary observing requires that the observer be able to concentrate on the object under analysis and not have to suffer the distraction of constantly adjusting the telescope tube or waiting for the object to pass through the field. Shaky telescope tubes are a horror and frustration. Russell Porter's requirement that telescope be so solidly constructed that one should be able to punch the side of the tube with his fist and not see the image shake does seem to be a bit excessive, but something between that extreme and the wobbly, hamstrung rigs I've seen weaving about in the breeze, or falling to one side with a dismal clunk, should not be too much to aspire toward.
In the end, the essentials of creating an optimum telescope for lunar, planetary and binary star observing is a result of combining a first-class mirror and diagonal, a well ventilated tube system and placing the entire business on a stable mount that tracks reasonably well. Add to this a decent eyepiece (a nice Plossl will do fine) and you have the basis of an excellent lunar/planetary telescope.