Theory of the Negative Impedance Converter 



By J. L. MERRILL, Jr. 



{Manuscript Received July j, iQ^o) 



This paper presents a relatively new approach to the solution of negative im- 

 pedance problems related to vacuum tube circuits. The approach consists of 

 reducing the vacuum tube circuit of a device for producing negative impedance 

 to an electrically equivalent four-terminal network from which the stability and 

 the operation of the device as an element in a system can be predicted accurately. 

 The theory is of interest at this time because a negative impedance repeater, the 

 El, has recently been developed for use in the exchange plant. It has been found 

 that such a repeater can be placed in series with a voice frequency telephone line 

 to provide transmission gains which are ample for many purposes. 



Introduction 



A NEW type of telephone repeater known as the El has been developed 

 -^ ^ recently to meet the large demand in the exchange area for an eco- 

 nomical means of providing transmission gains of about 10 db in tw^o-wire 

 telephone lines. This repeater costs less than the 22 type v^hich has been the 

 standard tvi^o-wire, tv^o-vs^ay means of amplifying voice currents in the Bell 

 System. The difierence in cost is made possible by a difference in operating 

 principle. The El repeater employs a type of feedback amplifier the action 

 of v^hich can be said to have the properties of a negative impedance converter. 

 It is the purpose of this paper to describe the operation of the negative im- 

 pedance converter, which is a device for transforming positive impedance 

 into negative impedance. 



Negative impedance like positive impedance can have two components: 

 reactance and resistance. The reactance component can be either positive or 

 negative. However, for an impedance to be negative the resistance compo- 

 nent should be negative at some frequency in the range from zero to in- 

 finity. 



The idea of negative resistance originated over 30 years ago, and in the 

 beginning was associated with the concept of resistance neutralization. This 

 concept grew from the observation that a two-terminal device could be 

 found which had an unusual property when inserted in series with a single 

 mesh circuit : it could produce the same flow of current as would flow other- 

 wise at some frequency provided a resistance R were removed from this mesh. 

 Apparently, the addition to a circuit of a two-terminal element could neu- 

 talize an amount of resistance equal to R. Thus within certain frequency 

 limits this two-terminal device could be treated as a negative resistance equal 

 in magnitude to — R. 



88 



