The location and the number of calibration points can have a significant effect on a system’s performance. The four plots here show measured data from the AD8318 log detector when swept with an input signal at 2.2GHz. In Plot A, the AD8318 has been calibrated using a 2-point calibration at –12dBm and –52dBm.  Looking at the resulting error plots, it is shown that the error performance is fairly evenly distributed over the -12dBm to -52 dBm calibration range. Note that the post-calibration error is not zero at all RF power levels. This is because the log amplifier does not perfectly follow the ideal Vout versus Pin equation, even within its operating region. The error at the two calibration points will however be equal to zero by definition. Plot A also includes error plots at −40°C and +85°C. These error plots are calculated using the 25°C SLOPE and INTERCEPT calibration coefficients. Unless one is willing to implement some kind of temperature-based calibration routine, the 25°C calibration coefficients must be relied upon and the slight residual temperature drift must be accommodated. In many applications, it is desirable to measure the RF power most precisely when it is at, or close to maximum power. By changing the points at which calibration is performed, the achievable accuracy can, in some cases, be greatly influenced. Plot B shows the same measured data as Plot A. However, in this case, the calibration points have been moved to -10dBm and -30dBm. This results in improved accuracy at the top end of the range coupled with poorer performance at lower input power levels. Plot C shows how calibration points can be moved to increase dynamic range, albeit at the expense of linearity. In this case, the calibration points are −4dBm and −60dBm. These points are at the end of the device’s linear range. This extends the range over which the AD8318 maintains an error of less than ±1dB. The disadvantage of this approach is that the overall measurement error increases, especially in this case at the top end of the detector’s range. Plot D shows the post-calibration error using a more elaborate multi-point algorithm. During factory calibration, multiple power levels (separated by 6dB in this example) are applied to the transmitter and the detector’s output is measured at each power level. These measurements are used to break the transfer function down into segments with each segment having its own SLOPE and INTERCEPT. This algorithm tends to greatly reduce errors due to transfer function nonlinearity and leaves just the temperature drift as the main source of error. The disadvantage of this approach is that the calibration procedure takes longer and more memory is required to store the multiple SLOPE and INTERCEPT calibration coefficients.