English
American building materials supermarket.Measure the sideband conversion effect of the power supply change. The power noise of AM sideband conversion is a transfer function through which the noise on the power supply is converted into the distortion of the output signal.
However, measuring this transfer function (transfer function) is not easy because there are very large impedance differences at the circuit points of interest: the power impedance is very low, the transistor output resistance (hopefully) is very high, and the load is close to 50 ohms. The normal calculation of the gain and transfer function in the RF cascade (assuming a consistent 50-ohm impedance environment) does not work here. A different technology is needed.
The best technique to calculate how the power supply voltage changes affect the output signal is to treat this distortion as an amplitude modulation (AM) process.
This avoids the need to measure any current or impedance. Using this method, we only need to measure the change in the voltage of the power supply and the sideband around the output signal. It's hard to get any easier.
This process is shown in figure 5-9. Any pure AM process has symmetrical sideband. If necessary, the measured sideband is indeed AM by passing the output signal through the limiter, which eliminates any AM sideband but does not change the sideband modulation through any phase. If any spectrum asymmetry is measured, it indicates that there are two effective modulation processes, including amplitude and phase modulation.
Firstly, the power supply with small sine wave is applied, in which the rms amplitude VA is added to the rated DC voltage VDC. After the RF signal is input to the amplifier, if there is any sideband conversion of the power supply voltage, two sideband will appear on both sides of the output signal and will be separated from the main signal with the sine wave frequency of the power supply.
The power (PC) of the RF output signal of the central component (here referred to as carrier power) and the power of each sideband (PSB) are measured. Knowing that the RF signal is in an environment of R = 50 ohms, then we use the following equation to determine the voltage of these carrier and sideband signals:
Use this to calculate the modulation index of the AM process:
The desired transfer function, that is, the DPS to AM transfer gain, is:
Only two simple measurements are needed:
1. Use an oscilloscope to measure the AC RMS voltage (VA) of the power supply voltage (independent of the specific power supply voltage), and.
two。. The AM sideband power (the arithmetic average power of the sideband or two sideband, if they are asymmetric) is converted to the equivalent root mean square voltage (AM).
Note: AM has nothing to do with signal power. When evaluating gDPS_AM, it is not necessary to carry out the calculation (5.2).
No additional information is required, including any information about PA and its design. This measurement process is completely independent of the characteristics of the PA, including the impedance of its power input. It is also not necessary to know the output RF carrier power, which means that the transfer function is independent of the output signal. The change of power supply voltage is directly converted to amplitude modulated sideband power through equation (5.5). The modulation index (5.4) does vary with the output power, but this is secondary, if any.
Because the PA transistor 0 ≤ AM ≤ VA. Converting the low impedance RL on the transistor to an output impedance of 50 ohms R0 (or possibly any other impedance) output matching network (OMN,output matching network) will result in AM measurements greater than VA at the output port of the PA. This is not a problem because we simply determine the measurements made at the observable point and are consistent with the results. Then, we need to change the expected range of DPS to AM transfer gain:
When AM «VA, PA is suppressing the noise signal on the power supply. We expect and require this to be used in envelope tracking. In fact, if this is not the case, the circuit will not perform envelope tracking. It is doing something else: multiplying the RF signal by the power signal VA. This is a modulator process, not an amplifier process.
However, measuring this transfer function (transfer function) is not easy because there are very large impedance differences at the circuit points of interest: the power impedance is very low, the transistor output resistance (hopefully) is very high, and the load is close to 50 ohms. The normal calculation of the gain and transfer function in the RF cascade (assuming a consistent 50-ohm impedance environment) does not work here. A different technology is needed.
The best technique to calculate how the power supply voltage changes affect the output signal is to treat this distortion as an amplitude modulation (AM) process.
This avoids the need to measure any current or impedance. Using this method, we only need to measure the change in the voltage of the power supply and the sideband around the output signal. It's hard to get any easier.
This process is shown in figure 5-9. Any pure AM process has symmetrical sideband. If necessary, the measured sideband is indeed AM by passing the output signal through the limiter, which eliminates any AM sideband but does not change the sideband modulation through any phase. If any spectrum asymmetry is measured, it indicates that there are two effective modulation processes, including amplitude and phase modulation.
Firstly, the power supply with small sine wave is applied, in which the rms amplitude VA is added to the rated DC voltage VDC. After the RF signal is input to the amplifier, if there is any sideband conversion of the power supply voltage, two sideband will appear on both sides of the output signal and will be separated from the main signal with the sine wave frequency of the power supply.
The power (PC) of the RF output signal of the central component (here referred to as carrier power) and the power of each sideband (PSB) are measured. Knowing that the RF signal is in an environment of R = 50 ohms, then we use the following equation to determine the voltage of these carrier and sideband signals:
Use this to calculate the modulation index of the AM process:
The desired transfer function, that is, the DPS to AM transfer gain, is:
Only two simple measurements are needed:
1. Use an oscilloscope to measure the AC RMS voltage (VA) of the power supply voltage (independent of the specific power supply voltage), and.
two。. The AM sideband power (the arithmetic average power of the sideband or two sideband, if they are asymmetric) is converted to the equivalent root mean square voltage (AM).
Note: AM has nothing to do with signal power. When evaluating gDPS_AM, it is not necessary to carry out the calculation (5.2).
No additional information is required, including any information about PA and its design. This measurement process is completely independent of the characteristics of the PA, including the impedance of its power input. It is also not necessary to know the output RF carrier power, which means that the transfer function is independent of the output signal. The change of power supply voltage is directly converted to amplitude modulated sideband power through equation (5.5). The modulation index (5.4) does vary with the output power, but this is secondary, if any.
Because the PA transistor 0 ≤ AM ≤ VA. Converting the low impedance RL on the transistor to an output impedance of 50 ohms R0 (or possibly any other impedance) output matching network (OMN,output matching network) will result in AM measurements greater than VA at the output port of the PA. This is not a problem because we simply determine the measurements made at the observable point and are consistent with the results. Then, we need to change the expected range of DPS to AM transfer gain:
When AM «VA, PA is suppressing the noise signal on the power supply. We expect and require this to be used in envelope tracking. In fact, if this is not the case, the circuit will not perform envelope tracking. It is doing something else: multiplying the RF signal by the power signal VA. This is a modulator process, not an amplifier process.
