It took him 18 months to obtain a 20-inch blank: he finally purchased one for £36 (bout $90) from Chance Brothers, Limited, of Birmingham, England. While awaiting its arrival and, as he writes, “working backwards in the good old fashion set up by ATM”, he first prepared on the roof of his house an 18-foot, metal-roofed, wood-lined dome turning easily by hand on 13 ball bearings.
Four years later Commander Silva wrote that he had finished his telescope and had it in working order, “ready for stargazing, variable-star-observing and photography”. He sent a collection of photographs made at its Newtonian focus. Three of these are reproduced on pages 181 and 186. That the telescope gives a high-grade optical performance is shown by the small round star images in the corners of the photographs (some of the images lose their roundness in the halftone engraving process).
“I choose the double-yoke English type of mounting” Commander Silva writes in his perfect English, ”because of its sturdiness and ease of construction and despite its one limitation, the inaccessibility of the Pole. Yet it can be pointed as far as 75 degrees north declination”. The yolk of the telescope is built of half inch oak plywood. The octagonal tube is built of the same material reinforced by iron ribs. The focal length of the mirror being 10 feet, the focal ration is f/6. A coudé Cassegrain secondary mirror, when used, converts this to f/18 for visual observing with powers up to 800. There are identical screw capped openings on the four sides of the tube for insertion of the eyepiece-diagonal unit, and four in addition for the coudé unit.
Commander Silva continues: “The mirror blank, originally 3 by 21 ½ inches, was worked throughout face up with a 12-inch subdiameter glass tool on which I had previously made two 12 inch mirrors, and thus it was worn down finally to only one-half-inch thick. The strokes used were those described in Amateur Telescope Making-Advanced. I found the instructions adequate when supplemented by thought and the usual willingness to experiment, fail, try again and maybe fail again but keep on learning until the results are attained. The subdiameter tool method permits a larger element of personal idiosyncrasy than the conventional method. The job was easier than I expected. The most tedious part was the roughing out, but the grinding went very well with the tool on top. Astigmatism was avoided by walking as regularly or, rather, as irregularly around the mirror as possible. It was always supported on the nine-point equalising system later installed in the telescope. Polishing and zonal correcting were done with honey-comb foundation, but final figuring was done on a pitch lap since HCF gives bad contact. By the subdiameter lap method the work advances very rapidly, and frequent testing is necessary. Raised zones were treated by pressure on the edge of the lap a little outside of their crest. I used cerium oxide because I was unable to get scratch-free rouge.
“Parabolization with the subdiameter lap proved easier than with full-size laps. By the Foucault test, the mirror was found to be a little under corrected. I let it stay and called it a job. Having caught chronic “mirroritis”, I should enjoy making a 30-inch mirror by the sub-diameter tool method.
“I am satisfied with the telescope’s performance. Visually it easily reaches its theoretical magnitude, 15.5. The drive is powered by a fan motor and drives the sector attached to the polar axis through a train of reducing gears and a nut and a screw which may be run back with a hand crank. Here at Alfeite the electric power is not synchronized and the voltage, nominally 220, often changes more than 10 percent; hence I built an automatic centrifugal regulator or governor to control the drive. The motor is belted directly to the regulator. The belt is not tight and may slip if the motor turns too fast. I had to calculate the position of the points of suspension of the two rotating masses so that for three revolutions per second they would be in equilibrium in any position, but if they turned faster they would suddenly rise and bring the rotating disk against a fixed at the top. To obtain the result accurately they should theoretically rise in a parabola instead of a circle. This being difficult to arrange, the points of suspension of each mass must be made to coincide with the centre of a circle tangent to the middle of the parabola.
“It was a pleasure, and perhaps a surprise, after the regulator was built, to test it and, counting the revolutions made in one minute, find the designed 180; for I must confess that I was a little doubtful about the results of my geometrical lucubrations. It is, however, possible to change the critical velocity a little if necessary by screwing the two masses up or down on their rods. This is the Young isochronous governor, not the Silva, and it works very well.
“The flow of current may be controlled by two resistances, which may be either taken out or put in the circuit by two buttons held in the hand during the photographic exposures, In this way it is possible to drive the motor faster or slower than its normal speed and, notwithstanding the governor, the driving mechanism obeys”.
Commander Silva, whose chapter on a double-star micrometer is included in Amateur Telescope Making-Advanced, is a professor of ballistics in the Portuguese Naval Academy and director of the Naval Laboratory of Explosives. He mentions that his hobby resembles his vocation - both consist of aiming hollow cylinders at the sky. “A few months ago”, he says, “I wrote some articles on telescope-making in a Portuguese popular science magazine. As a result there are now seven fellows grinding glass disks and hoping to see the moon at arm’s length”.
Instructions for calculating the design of the isochronous governor, with a worked-out example, which Commander Silva kindly supplied by request, are held over for the later publication for lack of space, as are description of his solar eyepiece, his method of making Ramsden eyepieces, and his solar prominence spectroscope with a Thollon (carbon dissulfide) prism.