RADIO ASTRONOMY — HAWKINS 281 



A single beam may be produced in an endless number of ways 

 which can become almost as complicated as the character of the de- 

 signer. If dipoles are connected together to cover a flat area they are 

 equivalent to a paraboloid telescope of the same area. The array 

 of dipoles, however, will operate only over a narrow band of wave- 

 lengths and it is difficult to point the sensitive beam to various parts 

 of the sky. A dipole may have five or more focusing rods placed 

 in front of it to form a Yagi-type antenna which is frequently seen 

 in use with short-wavelength TV receivers. Electrical energy may 

 also be picked up on a long metal helix. Both the Yagi and helix 

 are equivalent to paraboloids with apertures of from 1 to 2 

 wavelengths. 



It is possible to increase the quality of the image by means of the 

 interferometer. Two separate antennas are spaced at either end of a 

 long baseline and the signals are mixed together in the receiver. A 

 radio star perpendicular to the baseline produces signals that are in 

 phase at each antenna. As the earth rotates and the radio star makes 

 an angle with the baseline the signals will differ in phase and tend to 

 cancel out. In this way a radio star produces periodic variations as 

 it rises, passes due south, and sets. Now the effective aperture of the 

 telescope is equal to the length of the baseline, so that a narrow beam 

 can be produced with reasonable economy. Unfortunately, not one 

 but many narrow beams are produced, so that the results become diffi- 

 cult to interpret. Despite this limitation, however, the interferometer 

 has done much valuable work in determining the angular diameter 

 and exact positions of radio stars. 



The receiver is similar in many respects to those used in TV, except 

 that the voltage gain is high ('-'10 million) so that the radio noise due 

 to thermal motion of electrons at the input of the receiver is readily 

 detected. In radio astronomy great care has to be taken to maintain 

 a constant gain in the receiver because a fluctuation, say in the tem- 

 perature of the filaments in the tubes, would produce a variation of 

 noise at the output which would mask the faint signals being detected 

 from space. A standard source of energy is put in the place of the 

 telescope to calibrate the receiver as shown in figure 1. This is usu- 

 ally a diode vacuum tube since the noise power is accurately known in 

 terms of the current flowing through the tube. To minimize the effect 

 of variations in the thermal noise of the receiver the calibration is 

 sometimes carried out automatically at a rate of 25 times per second. 

 In this way a 25-cycle note is produced at the output and the ampli- 

 tude of the note is independent of receiver noise, being proportional 

 to the difference between the cosmic signal and the standard source. 

 There will always be slight ripples in the output, however, even with 

 an ideal system, because we are comparing two noise signals which 



