NA TURE 



[August 12, 1922 



was no rapid diminution of strength after passing the 

 horizon line from Carnarvon. 



The range of the system was also tested wholly over 

 land. A site was chosen at Hendon, and a reflector and 

 transmitter for 15-metre waves erected with the reflector 

 pointing towards Birmingham. Tests were commenced 

 in February 1921 from Hendon to a portable receiver 

 on a motor car. Very good speech was received up 

 to 66 mik-s. and fair speech in the neighbourhood of 

 Birmingham. A reflector station was then erected at 

 Franklev near Birmingham, 97 miles from Hendon, 

 and tests were started there in August 1921. 



Measurements with and without the reflectors 

 indicate that the energy received when both reflectors 

 are up is about 200 times the energy received when 

 not using the reflectors. Local measurements of the 

 polar curves taken round the station show that the 

 electric field in front of the station is increased ap- 

 proximately 4 times by the use of the reflector, and 

 that the same order of increase is obtained during 

 reception ; the increase of energy received due to the 

 use of the two reflectors should therefore be 4 2 x 4 2 = 256 

 times. 



During the continuous-wave tests at Carnarvon it 

 was found that reception was quite possible on the 

 transmitting aerial while the transmitter was operat- 

 ing. The heterodyne may be either the transmitter, 

 or ait independent small heterodyne in the receiver. 

 Both the transmission and the reception utilise the 

 same aerial and reflector, and the transmitter is left 

 going and can be operated while receiving. 



There is no reduction in strength while the trans- 

 mitter is on, but a practical trouble has appeared. 

 Owing to the comparatively large power, strong currents 

 are induced in all conducting structures and circuits 

 close to the reflector and transmitter, such as the sup- 

 porting towers and buildings, and every variable con- 

 tact produces a noise. The elimination of all variable 

 contacts in the neighbourhood of the transmitter has 

 proved a work of some magnitude. 



Reflectors besides giving directional working, and 

 economising power, are showing another unexpected 

 advantage, which is probably common to all sharply 

 directional systems. It has been noted that practi- 

 cally no distortion of speech occurs, such as is some- 

 times found with non-directional transmitters and 

 receivers. 



Although the results between Hendon and Birming- 

 ham constitute a record for telephony for ratio of 

 range to wave-length — for such results were believed 

 to Lie impossible two years ago — they are only a first 

 attempt and do not represent the best that can now 

 be done after the experience gained. But it has been 

 demonstrated that wave-lengths of the order of 20 

 metres are capable of providing point-to-point direc- 

 tional commercial service over very considerable 

 ranges. Such services will be comparatively secret 

 as compared with the usual non-directional type of 

 transmission. 



The directional effect obtained with reflectors which 

 are large compared with the wave-length is so good 

 that it was suggested that it would prove very useful 

 for position finding for ships near dangerous points. 



The general idea is that a transmitter and reflector 

 revolving will act as a kind of wireless lighthouse. It 



NO. 2754, VOL. I 10] 



is not intended at present for long ranges, but rather 

 that revolving reflectors should be erected in position, 

 similar to those at present occupied by fog signals, 

 and be capable of similar ranges, so as to give the 

 position to ships during fog when within about 10 

 miles of the danger point. 



An experimental revolving reflector was erected on 

 I m likeith, and tests were made to s.s. Pharos, the 

 lighthouse tender of the Northern Lights Commis- 

 sioners during the autumn of 1920. With a 4-metre 

 wave, spark transmitter, a reflector of 8 metres' 

 aperture, and a single valve receiver on the ship, a 

 working range of 7 nautical miles was obtained. The 

 reflector made a complete revolution once every 2 

 minutes, and a distinctive signal was sent every half- 

 point of the compass. The bearing of the transmitter 

 could then be determined within \ point of the compass, 

 or within 2-8 degrees. 



The best method of giving the direction to a ship 

 by means of such a revolving beam requires considera- 

 tion. When listening in a receiver to a moderately 

 sharp revolving beam the signals are heard only for 

 a very short time. The exact time of maximum signals 

 is not easy to determine by ear, but the times of start- 

 ing and vanishing are easv to determine, as the rate 

 of rise and fall of the signals is extremely rapid. The 

 time half-way between these two times gives with 

 great exactness the moment when the beam is pointing 

 to the ship. 



It would be quite possible to arrange to send a 

 general broadcast signal when the beam passes through 

 true north ; then by arranging for the beam to revolve 

 at a perfectly uniform rate, the bearing on the ship 

 could easily be determined by means of a stop-watch; 

 This method is probably the most accurate, but has 

 some disadvantages. It entails accurate timing 

 mechanism at the transmitter, the use of two waves, 

 and three, or perhaps four receivers on the ship, as 

 well as the use of a stop-watch. 



For the short wave two receivers are required, one 

 at each end of the bridge, or one fore and one aft. 

 This is necessary to avoid screening by the ship itself. 

 If the broadcast wave for giving the time when the 

 beam passes true north is another short wave, then 

 two more receivers would be required. 



The method provisionally adopted avoids accurate 

 timing mei hanism at the transmitter and the use of a 

 broadcast wave. On the base of the revolving reflector 

 contact-segments are arranged so that a definite signal 

 is transmitted every half- or quarter-point of the com- 

 pass. 



The apparatus proposed is of a very sturdy nature. 

 The spark transmitters are robust, and last for years 

 without attention. The receivers are simple valve 

 rectifiers with fixed adjustments except for a " backing 

 off " potentiometer for dealing with powerful signals 

 at close range. The attenuation of these waves ovet 

 sea is so strong that a little experience enables distance 

 to be judged by strength of signals, and this can be 

 measured by means of the potentiometer. The only 

 qualification necessary for a person determining the 

 bearing is the ability to read a few Morse signs. 



The success of the present experiments indicates a 

 wide sphere of usefulness for the new short-wave 

 directional wireless system. 



