1154 the bell system technical journal, october 1951 



Thickness Loss 



The effect of the thickness of the recording medium shows up in the 

 term (1 — ^"^tS/x)^ ^^ Jq^ frequencies for which the wavelength is much 

 greater than the thickness of the medium this reduces to lird/X. In this case 

 the reproduced voltage is proportional to the thickness of the medium and 

 to frequency. This is the familiar six db per octave characteristic. 



At high frequencies, however, when X ^ 5 the term reduces to unity 



10 

 15 

 20 

 25 

 30 

 35 

 40 

 45 

 50 

 55 

 60 



0.8 1.0 



0.01 a02 0.03 0.05 0.1 0.2 03 0.4 0.5 



SPACING IN WAVELENGTHS, d/A, 



Fig. 6 — Computed spacing loss as a function of d/\. 



and the computed "ideal" response is flat with frequency and independent 

 of the thickness of the medium. 

 If the term (1 — g-2»«/^) is rewritten as 



then the part in parenthesis accounts for a 6 db per octave characteristic 

 and the part in brackets accounts for a loss wilh respect to this 6 db per octave 

 characteristic. This loss, which will be called Thickness Loss*, is given by 



* It seems somewhat awkward to speak of "Thickness Loss" when nothing is actually 

 lost by making the medium thick. The only excuse for this way of splitting the terms 

 is that it makes for ease in comparing measured and computed curves. 



