|
There are several amplifier performance parameters which are important to a system designer. One of the most important is linearity. When specifying linearity two tone third order intermodulation measurements are usually taken because the measurement is simple using readily available microwave equipment (noise power ratio and multitone tests are also a measure of linearity, but require specialized equipment). This parameter is usually specified as a minimum level in dBc at a given total output power for both tones. For comparison purposes two CPI amplifiers were chosen which have nearly identical performance characteristics. The VZC-6964A4 is a compact medium power TWT amplifier. The SSCI-200 is a CPI solid state power amplifier. Both operate at C-band (5.9 to 6.4 GHz) but are representative of the entire CPI product line in both C-band and Ku-bands.
Tube amplifiers have twice the power added efficiency than SSPAs. Tube amplifiers have about 50 to 60% efficiency and SSPAs have 25 to 30%. The effects of this can be seen in the prime power consumed versus output intercept point. Output intercept point (OIP3) is a figure of merit and is equal to the total output power of two tones when the third order intermodulation products are down 0 dBc. This effect can never be seen because the amplifier saturates before it is reached but it can be calculated from the intermodulation distortion levels when the amplifieris operating in its linear range. Figure 1 shows the results of plotting OIP3 versus total prime power consumption. Notice that the TWTAs have typically higher linearity (higher OIP3) than the SSPAs. This is an effect of their higher efficiency. When the OIP3 falls below 56 dBm there really is no great difference between the two types of amplifiers. The reader will notice there is a slight difference between the two amplifiers chosen for this comparison, however the differences are not significant. The conclusion here is that for a given intermodulation requirement a tube amplifier above 56 dBm OIP3 will consume less prime power than an equivalent SSPA. Below OIP3s of 56 dBm both amplifier types will consume about the same prime power.
Figure 1.
Figure 2 shows a comparison between the TWTA and the SSPA transfer curves for the amplifiers chosen which have the same linearity. What the transfer curves show is that both amplifiers have the same output power 1 dB gain compression point which is what was expected for two amplifiers with the same linearity. The point marked P1dB shows both transfer curves deviating from the constant gain line by one dB at the point where the output power is 53 dBm. However, the similarity ends there. The SSPA transfer curve bends over sharply after the 1dB compression point whereas the TWTA output power continues to rise saturating at about 56 dBm. This has been a source of some confusion to users of power amplifiers in the past. How is it that two amplifiers with the same intermodulation linearity can be rated as a 400 W TWTA and a 200 W SSPA?
Figure 2.
The answer is tradition as determined by the marketplace. Traditionally TWTAs have been used for FM applications where they have been operated in saturation thus the TWTA amplifier in this case which has a saturation output power of 56 dBm minimum is called a 400 W amplifier. In the SSPA case the naming convention is to use the typical output power at the 1 dB gain compression point.
In this case the SSPA is specified at 53 dBm or 200 W. SSPAs are typically operated below the P1dB. Because of this difference in the naming convention for rated power for the two types of amplifiers the Satcom systems engineer cannot specify linearity at a specified output back off alone to determine the amplifier to use. To do so would result in comparing distortion of the TWTA at an output power 3 dB higher than the SSPA. He must also specify the absolute output power requirement at a given intermodulation distortion level, or the OIP3, in order to come up with the correct answer; which is that these are equivalent amplifiers. Before we leave Figure 2 it is also important to note that the SSPA has 3 dB less "burn through" reserve for high path loss situations such as rain fade than the TWTA.
Figure 3 shows clearly that both amplifiers exhibit the same linearity. A theoretical line through the data with a slope of 2:1 intersects the x-axis (IM3 = 0 dBc) at 62 dBm output power.
Figure 3.
This is the OIP3 for both of the amplifiers. With this number and the known relationship between the fundamental signal slope and the intermodulation slope the designer can estimate the IM3 levels at any output power level up to 55 dBm for the TWTA and about 51 dBm for the SSPA.
Figure 4 is a plot of the final measure of linearity of the two types of amplifiers. AM to PM in dimensions of o/dB for the SSPA are slightly better than the TWTA up to about 50 dBm output power and then the TWTA is slightly better. As was the case for AM/AM earlier the TWTA characteristics hold for power levels above 53 dBm where the SSPA amplifier saturates. With predistortion linearization used in either amplifier the AM to PM characteristics can be improved.
Figure 4.
Reliability is also a major concern for users of power amplifiers. The largest study performed on the relative MTBF's for the two types of amplifiers has been done on amplifiers used in all the Intelsat satellites. The failure rate of the SSPA population was higher than the TWTA amplifier population by about 15%. Therefore the user should consider similar sparing philosophies for amplifiers when developing maintenance plans for Satcom systems.
Cost is the final consideration. A good rule of thumb is that for amplifiers with the same linearity or intermodulation distortion characteristics (the OIP3s are the same) SSPAs will be more cost effective below 57 dBm OIP3 for C-band rack mount and 56 dBm OIP3 for C-band hub mount.
The numbers are 50 dBm for Ku- band rack mount and 46 dBm for hub mount. These tradeoff points are also valid for the size and weight of the amplifiers with TWTAs having the advantage at higher powers because of their higher power added efficiencies.
Please contact us for any questions regarding this paper or the amplifiers it discusses.
|
