Hi Support
I have inherited a power detector circuit that uses the LT5538 power detector.
Across multiple DuTs I measure the response of the detector in 2dB steps from -5dBm down to -15dBm, or so, in order to derive the 'linear' gradient (m) and intercept (c) response: as y=mx+c [with y in mV, m in mV/dB, x in dBm and c in mV].
There are two plots that seem relevant:
1) Slope Distribution vs Temperature at 2.14GHz == distribution of mV/dB (ie. m)
2) Logarithmic Intercept Distribution vs Temperature at 2.14GHz == distribution of extraplolated intercept 0mV point
As the gradient and intercept points are correlated it is hard to gauge how the intercept point relates to a linear response whose c term is determined at 0dBm. There >12dB spread in the extrapolate data for 0mV. Is this also seen in the 0dBm c term?
The Vout v IP Power Plots suggest the power detector does not suffer from a wide spread in the c term. The reason I ask is that I am trying to establish production limits for the c term at 2.5GHz and also want tor determine if I have ESD components (such as RF switches or even the power detector itself).
Are you able to enlighten me as the 6 sigma spread I might expect for the c term please? And also what the ESD rating is for the LT5538?
Many thanks
