Passive Intermodulation(PIM)

Size: px

Start display at page:

Download "Passive Intermodulation(PIM)"

Tabitha York
5 years ago
Views:

1 Passive Intermodulation(PIM) Causes of PIM in the Passive Equipment`s such as FM, VHF, UHF Antenna and What does a field technician need to fix a PIM problem?- By Asghar Bahrani Desember 12,13,

2 Contents What is passive intermodulation, PIM? Causes of passive intermodulation, PIM products Passive intermodulation Effects Passive Intermodulation Measurements How to prevent PIM Conclusion 2

Intermodulation Active versus Passive intermodulation Intermodulation is caused when 2 or more RF carriers are mixed in an active system and from unwanted

3 Intermodulation Active versus Passive intermodulation Intermodulation is caused when 2 or more RF carriers are mixed in an active system and from unwanted signals When passive components containing non-linear elements those are the source of this interference We refer it in this case as Passive InterMdulation (PIM) 3

4 Linear system In linear devices, the output is linearly proportional to the input. When two signals at different frequencies F1 and F2 are mixed, the result at the output is two signals at F1 and F2. Ideally, no other frequency components are generated. (see figure 1 below) 4

5 Non-Linear system In non-linear devices (see figure 2 below), when two signals at different frequencies F1 and F2 are mixed at the input, the result is a series of harmonics and high-order frequency components at the output:these new frequency components become a source of interference and therefore need to be carefully controlled. \ The 3rd order intermodulation (2F1-F2) is the strongest product. The 5 th order PIM product is about 15 db lower than the 3 rd order and the 7 th order is lower by an additional 15 db. The PIM figure increases as well with increased input power levels by an approximate 2:1 ratio. (when the input power is increased by 3dB, the PIM figure will be increased by approximately 6dB) \ When the 3 rd, 5 th or higher order mix fall within the RX band, the level must be below the squelch point. If not, it will cause significant desense (receiver desensitization) issues. 5

6 IM Mathematics 6

7 IM Mathematics Frequencies generated by intermodulation distortion 7

8 PIM Model In this model Is and Vt are just fitting parameters and they can be tweaked to any value until the wanted PIM is obtained. For example, Is can be used to control the power level of PIM and Vt can be used to adjust the power distribution between different order products Diode-like Model 8

9 Nonlinear Diode Effect at ferromagnetic metals A low signal operating in a linear region and a large signal operating in the non-linear region of a ferromagnetic metal is creating additional spectral components in the output signal. 9

10 PIM is a result of signal mixing at nonlinearities IM3 PIM nonlinearity increases, in theory, at a ratio of 3:1 (PIM to signal) A 1 db increase in carrier power correlates to a theoretical incre ase of 3 db in PIM signal power. In practice, the actual effect is closer to 2,3-2,5 db as the thermal noise constant -174 dbm/ Hz becomes an error contributor. 10

11 PIM are clogging up complete RF bands PIM multiplies bandwidth If bandwidth of f1 and f2 is 1 MHz then BWIM3 = 3 MHz BWIM5 = 5 MHz BWIM7 = 7 MHz 11

12 A real example - TELSTRA Next GTM UMTS 850 Example f1 = 887 MHz, 5 MHz UMTS T f2 = 935 MHz, 200 khz GSM T fim3 = 839 MHz, CDMA RX 12

13 Summary of the phenomenon PIM is of particular concern when PIM products fall in the RX band Two or more transmitter channels share a common antenna TX signal levels are high RX sensitivity is high TX and RX are diplexed 13

14 ACTIVE VS PASSIVE: Active devices require sources of energy where as passive devices do not. Active devices are non-linear and thus, the main source of intermodulation distortions and spurious emissions in RF systems. The linear passive devices that are supposed to be linear show nonlinear behavior that results in minor distortions, commonly referred to as Passive Inter Modulation (PIM) distortions. 14

15 Importance of PIM Modern systems indeed require: much tougher frequency plans, use of higher transmitter power levels more sensitive receivers, PIM manifests as an interfering signal which may degrade the performance of the receiver. 15

16 What actually causes PIM? Junctions of dissimilar materials. rust, corrosion, loose connections, dirt, oxidation, and any contamination of these factors. The result is a diode-like nonlinearity that makes an excellent mixer. As nonlinearity increases, so does the amplitude of the PIM signals. 16

17 PIM Causes Non-Linearities take two different forms Contact Non-Linearity Material Non-Linearity 17

18 PIM Causes - Contact Non-Linearities Causes of contact Non-Linearities Junction capacitance due to thin oxide layer between conductors Impurities on metal surface Semiconductor tunnel / schottky effect at point of contact Contact restistance caused by two dissimilar metals 18

19 PIM Causes - Material Non-Linearities Causes of material Non- Linearities Hysteresis effect in ferromagnetic materials (nickel, iron, steel) Thermal Heating due to p oor conduction rate (torque, corrosion, cracks) 19

20 Root Causes of PIM in a real RF environment 1- Loose and / or inconsistent 2- metal to metal contacts 3- Not enough contact pressure. 4- Cracked solder joints 5- Cold solder joints 6- Scratches or dents at mating interfaces 7- Burrs 8- Metal flakes, chips, dust 9- Improperly formed or Misaligned parts 10- Rough mating surfaces (saw cut) 11- Loose metal to metal contacts 12- Loose or rusty bolts 13- Ferromagnetic materials (steel, nickel, etc.) 14- Wind induced vibrations 1- Trapped between mating surfaces 2- Trapped between plating layers 3- Solder splatters 4- Dirt or debris 5- Surface Oxides 6- Insufficient thickness of plated metal causing RF heating 7- Too much or too little torque at 8- Contaminated conductors & Interfaces (Dirt, Dust, Moisture) 20

21 Field Examples ferromagnetic materials cracked solder joints Antenna showing oxidation within the power divider 21

22 Field Examples 22

23 Field Examples LDF4-50A RF- Repeater feeder cable 10 db Return Loss after in stallation no Repeater operation po ssible due to high noise level in Don or-site RX band 35 db Return Loss after c onnector swap 23

24 Copper Foil Copper foil is of huge influence on PIM 24

25 Very Low Profile ED Copper Foil 25

26 SEM Photo of Microstrip (C1 copper) copper debris left in at the bottom of the etched track 26

when designing a system, development teams will choose passive elements with minimal or acceptable levels of PIM as specified by the component manufacturer.

27 when designing a system, development teams will choose passive elements with minimal or acceptable levels of PIM as specified by the component manufacturer. Circulators, duplexers, and switches are particularly prone to the effect. Designers may choose to accept higher levels of passive intermodulation by selecting lower cost, smaller size, or lower performance options. Component design trade-offs, size, power, rejection, and PIM performance 27

equipment for intermodulation measurements.

28 The diplexer A30089 is specially designed for very low passive intermodulation and enables an improvement of your existing equipment for intermodulation measurements. The Antenna-Port includes an directional coupler to monitoring the forward and reverse power. 28

29 What other conditions affect PIM? PIM tends to increase as components age. Environments wide temperature variations, salt air polluted air, excessive vibrations 29

30 What conditions are necessary to cause PIM? two relatively strong RF signals relatively close in frequency are required to trigger PIM effects. The outputs from two or more high-power (20 W or so) transmitters are enough to create the PIM effects. The higher the power used, the greater the PIM signals generated. 30

31 How are the various signals are formed? PIM nonlinearities produce a specific frequency spectrum. 31

32 How PIM could be controlled? Avoid the use of ferrous metals. Minimize the number of contact junctions. Design all contact junctions such that they are precise and under sufficient pressure to maintain good contact. 32

33 How PIM could be controlled? Solder or cold weld all junctions where possible. Avoid dissimilar metals in direct contact. Plate all surfaces to prevent oxidation. Make certain plating is uniformly applied and of sufficient thickness. 33

34 Real world deficiencies that occur: Poor alignment of parts. Inadequately torqued screws and fasteners. Bad solder joints. Insufficient or incomplete cleaning of parts prior to plating. 34

35 Real world deficiencies that occur: Contaminated plating tanks. Plating material build-up. Using wrong materials. Poor plating adhesion. 35

36 Summary of the phenomenon PIM is measured acc. to IEC Ed RF connectors, connector cable assemblies, a nd cables intermodulation level measure ment Standard specifies the use of t wo 20 watt carriers ( 2 x +43 dbm) Typical IM3 value is -165 dbc 36

37 IEC : 2012 Passive Intermodulation Standard IEC :2012 Passive RF and microwave devices, intermodulation level measurement - Part 1: General requirements and measuring methods IEC :2012 Passive RF and microwave devices, intermodulation level measurement - Part 2: Measurement of passive intermodulation in coaxial cable assemblies IEC :2012 Passive RF and microwave devices, intermodulation level measurement - Part 3: Measurement of passive intermodulation in coaxial connectors IEC :2012 Passive RF and microwave devices, intermodulation level measurement - Part 4: Measurement of passive intermodulation in coaxial cables IEC :2013 Passive RF and microwave devices, intermodulation level measurement - Part 5: Measurement of passive intermodulation in filters IEC :2013 Passive RF and microwave devices, intermodulation level measurement - Part 6: Measurement of passive intermodulation in antennas 37

38 How is PIM Specified? Absolute power (dbm) The absolute IM power (in dbm) of intermodulation signal Relative power (dbc) IM signal power relative to the standard signal carrier Example: -110dBm IM signals generated by to +43dBM tones is a -153 dbc IM level 38

39 Test Methods when two +43 dbm carriers are injected into the device under test, i.e., -153 dbc.a typical specification requires a passive IM level no greater than 110 dbm To put this into perspective, this is a ratio of 1:2,000,000,000,000,000, or the equivalent of trying to measure the distance to the sun to an accuracy of one-tenth of a millimeter. 39

40 Realize the Necessary test Setup Rack and stack two synthesizers feeding two high power amplifiers that connect to an array of RF components. Route the signal of interest through a low noise amplifier to a spectrum analyzer. Power meter is used to set the proper transmit powers and readjust on a periodic basis to compensate for drift. Measurement results are sometimes difficult to repeat, and setups are often unstable and susceptible to damage 40

41 Example of Real Test Considering a device used in a PCS1900 Band network. Setting Carrier 1 to 1930 MHz and Carrier 2 to1990 MHz. (Note: There are some PIM test setups that, due to their design, must use one carrier set to a frequency in the receive band while the other carrier is set to a frequency in the transmit band. The validity of this approach open to debate.) The test is initiated. After the transmit frequencies and powers are established and the spectrum analyzer configured to measure the IM response, During the measurement, care is often taken not to disturb the device under test or the test setup In case instabilities might produce a data spike at the precise moment the spectrum analyzer happens to be sweeping past the IM frequency. 41

42 Dynamic Measurements IM performance is monitored during the application of an appropriate stimulus. Dynamic measurements has been given a great deal of attention with regard to cable assemblies. The concern has to do with the vulnerability of the connector /cable interface as well as IM created in the cable (by micro-cracks in solid conductor cables and discontinuous contact in braided cables) Testing involves measuring the IM as the cable is flexed and/or a bending moment is applied at the connector/cable interface. Two measurements routinely performed on all PIM critical components used,the tap test and the bending moment test. 42

43 Tap Test The tap test is simply a matter of tapping the device and watching the IM response. For example, tapping the tuning screws on filters frequently generates high levels of passive IM. After the tapping is stopped, the IM sometimes returns to its low IM condition and sometimes it remains high. The tap test has been demonstrated to be highly useful in screening devices and cables that will fail at some future time. 43

Dynamic IM Measurement the Strip Chart Mode of the analyzer is used to record the IM response over time. The device under test is a PCS1900 bandpass filter.

44 Dynamic IM Measurement the Strip Chart Mode of the analyzer is used to record the IM response over time. The device under test is a PCS1900 bandpass filter. What we are observing is the changing IM due to tapping. Later testing also showed this device to have much-degraded IM performance with temperature changes. 44

45 Bending Moment Test The bending moment test is performed by applying a modest side force to the connector on the device under test. If the connector is not adequately attached to the body of the component, or if the launching mechanism within the device is not solid, IM will be generated. 45

46 Swept Frequency Measurements he accepted method for measuring passive IM was to set the carrier to two fixed frequencies in the transmit band, and measure the IM generated. on many devices, this is not adequate, because these devices exhibit IM characteristics that vary as a function of frequency. The Passive IM Analyzers make this measurement at multiple frequencies practical and easy. 46

47 Swept Frequency Measurements Example A PCS1900 band duplexer with a bad connector is measured. It should be noted that the connector damage was caused by over-to rquing the connector and is undetected by visual inspection and swept return loss measurements. The comparison between The good and the bad duplexer return loss is shown in the following figure. The good duplexer meets the 115 dbm specification at all frequencies The bad duplexer has a severely degraded IM as the transmit carriers are moved from the band edges. Both duplexers would pass through QA screening if only band-edge carriers were used for the testing. To fully characterize the device under test, the swept frequency capability of the Passive IM Analyzer records the IM performance as each of the two carriers is swept in frequency across the transmit (down-link) band. Only one carrier at a time is swept, while the remaining carrier is held fixed at the band edge. 47

48 Swept Frequency Measurements Example 48

49 3 rd Order IM Response Carrier 1 is fixed at 1930 MHz Carrier 2 is swept from 1950 to 1990 MHz producing third order intermodulation products (IM3) that occur at frequencies from 1870 to 1910 MHz. 49

50 Forward and Reverse PIM Measurements Many passive RF components, like attenuators, can be utilized in both directions. Forward and reverse PIM measurements of components may display slightly different results. PIM ratings on manufacturers'data sheets should include both forward and reverse PIM characteristics. International standard for PIM measurements is IEC

51 PIM Analyzers Analyzers used by field personal are typically single port systems Dual port analyzers are used at production floors. Dual port analyzers are not to be confused with dual band PIM analyzers. What is the difference? The latter have the capability to measure two different frequency bands; the former provide two ports, one to perform reverse PIM measurements, the other for forward PM measurements 51

52 Block Diagram of Single Port & Dual Port PIM Analyzer 52

53 Block Diagram of Single Port & Dual Port PIM Analyzer RF generators which drive two high-power amplifiers (HPAs). The generated signals are variable in frequency and level. The generators can be set to any frequency in the specific band for which the PIM analyzer has been designed. An ultra low PIM combiner brings the amplified signals together and delivers the combined carriers to the Tx port of a duplexer. 53

54 Block Diagram of Single Port & Dual Port PIM Analyzer The duplexer has to provide very high separation between its wireless pages high power Tx and very low power Rx ports. As better this duplexer is designed, as better the dynamic range of the PIM test system. The output of the duplexer is basically the reverse PIM port of the PIM analyzer. 54

55 PIM Analyzers With the exception of antennas and loads, all Devices Under Test (DUTs) require external terminations to perform PIM measurements. Antennas radiate the transmitted RF power directly into the air loads convert it into heat. Any test cable and load that is used for PIM testing, must have significantly better PIM ratings than the actual DUT to allow for accurate measurements. 55

56 PIM Analyzers PIM signals that arrive at the analyzer port are channeled to the duplexer's Rx leg. The duplexer frequency correlates with the receiving band of the wireless system. Dual-port analyzers consists of all these elements, but have a second duplexer an additional internal high power, low PIM termination and a low PIM switch. 56

57 PIM Analyzers The second port feeds into the second duplexer, which has a terminated Tx port. Since an internal termination is provided, dual-port analyzers do not require external loads to terminate 2-port DUTs. The Rx port of the second duplexer measures forward PIM power. A low PIM switch toggles between reverse and forward PIM measurements. During PIM measurements, system PAs and receivers of the RF systems are disconnected and instead PIM analyzer are connected. The analyzers will now generate two high-power CW tones within the transmit band of the measured system. It is important to apply enough energy during PIM measurements to ensure the RF system is tested under conditions similar to the actual utilization. 57

Forward and Reverse PIM Measurement of Cable For many measurement applications single-port analyzers are suitable for both, reverse and forward PIM analysis.

58 Forward and Reverse PIM Measurement of Cable For many measurement applications single-port analyzers are suitable for both, reverse and forward PIM analysis.figure 4 shows forward and reverse PIM measurements for a simple two-terminal DUT; a cable in this example. Forward and reverse PIM will not differ much, and can easily be measured in two steps. Reverse PIM is measured by connecting one end of the DUT (A) to the PIM analyzer, while the other end is terminated with a low PIM load (B). Forward PIM is measured with the direction of the DUT reversed (B-A). The resulting PIM measurements are quite similar. 58

59 Forward & Reverse PIM Measurement of a 3 port DUT Dual-port units help measure forward and reverse PIM of components in less time. This is very beneficial for complex components with more than two ports. The example in Figure 5 shows a power splitter under test. Single-port units require three test steps, feeding the analyzer's measurement signals sequentially into the splitter ports IN, OUT1 and OUT2. Ports that are not connected to the PIM analyzer need to be terminated for accurate measurements. Dual port analyzers offer more convenient testing. During the first test step, measurement signals are fed into IN port of the splitter, while the forward port is connected to OUT1, with OUT2 59

60 Residual PIM Residual PIM is unrelated to the DUT, but it is a characteristic of the PIM analyzer. Analyzers are built with components that are far superior to the ones used in network operations PIM analyzers generate internal PIM. Even if the amount of PIM generated by these special internal components is small, it sums up and can influence the measurement. Manufacturers provide residual PIM specs in their data sheets. Residual PIM should be at least 10 db below the measurement range of the analyzer. Modern PIM analyzers use not only ultra-low PIM components, but minimize possible residual PIM readings with sophisticated DSP technology. 60

61 Which companies make PIM test equipment? Agilent Anritsu Boonton Kaelus Rosenberger Summitek Tessco. 61

62 62

63 Conclusion 63

64 Rep. ITU-R SM.2021 The following values [Petrovic and Gosling, 1979] for intermodulation performances are desirable: for single frequency transmit and receiving applications: 100 dbc for multiple frequency transmission: 130 dbc for multiple frequency transmission and reception: 143 dbc Similar considerations hold for other non-linear components like circulator, combiner, isolators, etc., installed between output stage of the amplifier and the transmitting antenna. Examples for mobile radio are given in [RA, 1987]. 64

65 fintermodulation mf1 nf2 = + where m and n are integers and ( m + n ) is the order of the intermodulation product. Part of the output spectrum of a non linear device excited by two signals f1 and f2 is shown in the following figure. How much spacing to you really need between antennas at radio sites? Robert S. Mawrey, Ph.D. Vice President of Systems and Technology 65

66 The most effective way to combat shared site radio frequency interference is to isolate nonlinear devices from strong signal sources. As illustrated in the following figure this is typically achieved using a combination of antenna isolation, filtering, and the selective use of more linear devices and ferrite isolators. Passive intermodulation can be reduced by minimizing the number of loose metallic joints2 within system components such as antennas, cables and connectors, and in the external environment on towers and wire fences. M. N. Lustgarten, COSAM (Co-site Analysis Model), IEEE Electromagnetic Compatibility Symposium Record, Anaheim, California, pp , July

67 Actual antenna spacing requirements can be estimated using comprehensive interference analysis techniques. The interference analysis should include investigation of Intermodulation and harmonics Noise Desensitization Antenna coupling Equipment characteristics Interference Mechanisms Radio frequency interference at a shared site is typically caused by one of the following mechanisms that have their origin in some form of equipment or other non-linearity: Intermodulation Transmitter Receiver Passive Out-of-band emissions Transmitter noise Transmitter spurious emissions Transmitter harmonics Receiver spurious emissions Other effects Receiver desensitization 67

68 Thank you for your attention! 68

Intermodulation Distortion in RF Connectors

Intermodulation Distortion in RF Connectors Introduction Intermodulation distortion or IMD has always existed in RF transmission paths. Until about the early 1990 s, cellular communications had relatively