The problem is based on the double source method for the determination of the dead time of a G.M. counter. Two radioactive sources of comparable strength are chosen. In all four counts are taken. First, the background rate B per second is found out when neither source is present. One of the sources is placed in a suitable position so that a high counting rate N1 is registered. While this counter is in the same position, the second source is placed by its side to get the counting rate N12. Finally, the first source is removed so that the second counter above gives a count of N2. If n1, n2 and n12 are the true counting rates then we expect
(n1 + B) + (n2 + B) = n12 + B
or
n1 + n2 = n12 − B (1)
Now, for each particle that is counted, on an average there will be a dead time τ during which particles are not counted. Then if N is the counting rate (number of counts/second) then the time lost will be Nτ , so that the true counting rate
n = N/(1 − Nτ )
Thus n1 = N1/(1 − Nτ) etc.
We can then write (1) as N1/(1 − N1τ) + N2/(1 − N2τ) = N12/(1 − N12τ) − B/(1 − Bτ) (2)
Now, in practice N1 and N2 will be of the order of 100 per second, N12 is of 200 per second, B ≈ 1 per second and τ ≈ 10−4 second, so that N1 τ << 1, etc. We can then expand the denominators binomially and write to a good approximation
N1(1 + N1τ ) + N2(1 + N2τ ) = N12(1 + N12τ ) − B(1 + Bτ)
N1 + N2 − N12 + B = τ[N122 − N12 − N22 ] ≈ τ [(N1 + N2)2 − N12 − N22]
Or τ = (N1 + N2 − N12 + B)/2N1N2