File talk:Capt FV Thompsons Stereoplotter.jpg

Notes attached to photo by Professor Kenneth Mason:

"CAPT. F.V. THOMPSON's STEREOPLOTTER"

1. Preliminary work in the field After a very careful reconnaissance of all ground to be surveyed, and all main Trigonometrical points observed, it must be decided exactly where the photos are to be taken from, and how many can be dispensed with. This depends on:

(1)Contour of the ground (2)Time available (3)Saving of labour in the field and in plotting (4)Expense

In Cumberland, where the hills are not excessively high, but where no very careful reconnaissance was carried out about 46 pairs of photographs were taken in an area of about 24 square miles. About 2 pairs to a square mile would be a very rough estimate - probably an underestimate.

2. Photography It is only intended very briefly to run through the process. A rough base line of 200' was measured by means of a string and levelled, and two tripods set up say at A&B. On tripod A is placed a vertical stick on a baseplate; on B the solid wood box camera is placed and a small telescope of which the axis is at right angles to the camera's axis is fixed to the top of the camera. By aligning this exactly on the "finger" of A the axis of the camera is brought at right angles to AB and the photo taken through a 'x32' orthochromatic screen - by actinometer rate. The camera is then placed on A tripod and the "finger" on B, taking care not to shift the tripod. The telescope is revolved on its transverse axis through 180 degrees and again alinged on the finger. The other photo is now taken. In selecting stations, A&B should as near as possible be on the same level - a slight difference does not seriously matter.

A hand sketch of chief points should be drawn. With a little practice 20 mins will be found sufficient to unpack, set up, level, take a pair of photos, and pack up again.

In Cumberland an average of 10-12 pairs of photos was maintained, omitting one day which was lost owing to climbing Helvellyn and getting lost in cloud. No photos were taken.

3. Development and production of transparencies Development should be done on the evening of the day of exposure so as to enable fresh ones to be exposed in case of failure. L&R transparencies shold also be made at this period, to prevent any unnecessaryfield work in case of breakages. Development should always be done by time to get exact similarity between L&R

4. The Stereoplotter The Stereoplotter shown is an instrument designed for mechanically plotting by aid of exaggerated stereopscopy maps from the transparencies produced as shown above. It consists of a wooden base A, on which is mounted a brass frame B for holding the transparencies. Rigidly attached to B is lug C, on which the stereoscope D is moved by means of a vertical screw E. F is a hand wheel actuating a horizontal shaft G moving the transparencies in a plane at right angles to the line of sight of the stereoscope. H is a graduated drum on a spindle J, screwthreaded in a socket K, for moving R to of from L. M is a vertical height adjustment. N the drawing board. OO' is the mean ray in plan from observer O-O' being actuated by the spindle G.

P is the distance scale corresponding to the drawn H. Q1 Q2 being the indicating rule. Q2 Q3 gives the adjustable height scale and S the mean vertical ray pivotted at S1 and actuated at S2 by a weighted string T attached over a pulley V to the slide D. W, W1, W2 are adjusting screw for the transparencies. The main fault in this, the first instrument, is the unrigidity of the support X and variation in tension in T; this has been improved upon. The instrument was made by Watson & Sons, High Holborn, London.

5. Action and Use The transparencies are placed in their frames, 'L' in the left hand one, and 'R' in the right. THe first adjustment to make is for true horizontality, which is done by turning the wheel F and getting the side marks on the two plates in a straight line. True verticality is then tested; if any adjustments are necessary these are done with W. Having satisfied onesself that the plates are situated correctly - close the right eye and by means of F, bring one of the known points, (either trig station or already observed point, the former for preparing) directly under the hanging chain. Now open the right eye, fix both eyes on the chain and move the right plate until the chain appears to hang immediately over the given point. The scale P is now to be selected by trial. Several scales near the scale required are photographed and presented. The point in plan is marked say A. Similarly a point B may be found. The distance between A and B is known so if in plan AB is not correct a new scale must be attached. These scales will not differ much and the nearer and more accurately to a fixed base (say 200') is measured, the nearer and more often is one scale fixed. These scales are photographed on paper and clipped at zero. The same height scale is always used and adjusted by screws Q2 Q3 to read the height of A or B when the stereoscope is set on the point A or B, whose heights are known. The instrument is now set to start work. I found after a short time the best way to go to work was to work along roads, rivers etc. first, fixing one or two well defined objects, such as cottages. Constant check readings are advisable on A and B both for height and distance. Heights should always be taken even with very distant objects as they always serve as checks on other maps. After main roads, isolated trees, etc have been filled in it is a good plan to work on a main countour following round and getting the general shape of the ground. By this splitting up the map a number of readings may be saved in filling in less important readings. If owing to awkward ground it was impossible to take both photos on the same level L photo must be adjusted before such reading by M, but this is not necessary for an inch or two. All screws must be finished in the same way before reading. M,E & H right handed. F left handed. One pair is plotted in 3.5-4 hours, or about 36 sq. miles per month.