XC2000 and XE166 Family Derivatives / Base Line Electromagnetic Emission Summary AP16167 Test Report V1.0 2009-04 Microcontrollers Edition 2009-04 Published by Infineon Technologies AG 81726 Munich, Germany © 2009 Infineon Technologies AG All Rights Reserved. LEGAL DISCLAIMER THE INFORMATION GIVEN IN THIS APPLICATION NOTE IS GIVEN AS A HINT FOR THE IMPLEMENTATION OF THE INFINEON TECHNOLOGIES COMPONENT ONLY AND SHALL NOT BE REGARDED AS ANY DESCRIPTION OR WARRANTY OF A CERTAIN FUNCTIONALITY, CONDITION OR QUALITY OF THE INFINEON TECHNOLOGIES COMPONENT. THE RECIPIENT OF THIS APPLICATION NOTE MUST VERIFY ANY FUNCTION DESCRIBED HEREIN IN THE REAL APPLICATION. INFINEON TECHNOLOGIES HEREBY DISCLAIMS ANY AND ALL WARRANTIES AND LIABILITIES OF ANY KIND (INCLUDING WITHOUT LIMITATION WARRANTIES OF NON-INFRINGEMENT OF INTELLECTUAL PROPERTY RIGHTS OF ANY THIRD PARTY) WITH RESPECT TO ANY AND ALL INFORMATION GIVEN IN THIS APPLICATION NOTE. Information For further information on technology, delivery terms and conditions and prices, please contact the nearest Infineon Technologies Office (www.infineon.com). Warnings Due to technical requirements, components may contain dangerous substances. For information on the types in question, please contact the nearest Infineon Technologies Office. Infineon Technologies components may be used in life-support devices or systems only with the express written approval of Infineon Technologies, if a failure of such components can reasonably be expected to cause the failure of that life-support device or system or to affect the safety or effectiveness of that device or system. Life support devices or systems are intended to be implanted in the human body or to support and/or maintain and sustain and/or protect human life. If they fail, it is reasonable to assume that the health of the user or other persons may be endangered. AP16167 Electromagnetic Emission Summary Revision History: V1.0, 2009-04 Previous Version: none Page Subjects (major changes since last revision) We Listen to Your Comments Is there any information in this document that you feel is wrong, unclear or missing? Your feedback will help us to continuously improve the quality of this document. Please send your proposal (including a reference to this document) to: [email protected] Test Report 3 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Table of Contents 1 Introduction ........................................................................................................................................5 2 Emission Measurement Methods .....................................................................................................6 3 Emission Limit Curves ......................................................................................................................8 4 Microcontroller Operation during Test ............................................................................................9 5 Application-Typical Emission Tests ..............................................................................................10 6 Worst-Case Emission Tests............................................................................................................21 7 References ........................................................................................................................................36 8 Executive Summary .........................................................................................................................37 Test Report 4 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 1 Introduction This summary describes the electromagnetic emission behaviour of the following microcontroller products: PG-LQFP-64 package: XC223xM derivatives XC233xA derivatives XC2735X derivatives XE162xM derivatives PG-LQFP-100 package: XC226xM derivatives XC236xA derivatives XC2765X derivatives XE164xM derivatives PG-LQFP-144 package: XC228xM derivatives XC238xA derivatives XC2785X derivatives XE167xM derivatives Slight differences in the electromagnetic emission performance exist between the package versions (PG-LQFP64, 100, 144). The derivatives which come in the same package show similar emission due to the fact that they contain similar microcontroller designs. Detailled reports on electromagnetic emission – dedicated to every package version – are available on request as listed in chapter 7. The detailled reports contain much more information about the test board, the software settings wrt. module operation, additional probed pins, and comparison with other Infineon 16-bit microcontrollers. The following derivatives have been measured as representatives of the different packages to obtain the electromagnetic emission results found in this summary: PG-LQFP-64 package: XC2735X PG-LQFP-100 package: XC2361A PG-LQFP-144 package: XC2286M All EMI test results presented in this report are obtained from hardware and software test setups compliant to the “Generic IC EMC Test Specification” (“BISS paper”), open copyright 2004 by Robert Bosch GmbH, Infineon Technologies AG, Continental AG (former Siemens-VDO). All significant emission peaks around 950 MHz are mainly caused by transmitters in the GSM band, and not caused by microcontroller emission. Test Report 5 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 2 Emission Measurement Methods The setup used for electromagnetic emission measurement complies fully with the BISS test specification which can be provided on request. One test board was designed for every PG-LQFP package with similar layout. The test board is used for conducted and radiated emission measurements. Conducted emission is measured using the standardized 150 Ω network, see figure 1. This network is used for both port and power supply emission measurements. Only crosstalk noise is measured; i.e. the port pins under test are never actively switching. Frequency range is 150 kHz to 1 GHz. Fig. 1 shows the schematic 150 Ω network connection to the microcontroller (IC under test) and the general layout of each 150 Ω probing net. Figure 1 150 Ω probing networks for conducted emission measurement 150Ω networks are provided for conducted emission measurements according the international standard IEC 61967 part 4 and the BISS test specification for a set of signals and power supply nets. All digital power supply nets plus a subset of I/O pins (typically located in the center of all 4 package edges) are measured, see figure 2. Figure 2 Test Report Probed supply and signal nets for conducted emission measurement 6 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Radiated emission is measured using the standard mini TEM cell according IEC 61967 part 2 and BISS emission test specification. The frequency range is from 150 kHz to 1 GHz. Figure 3 Measurement setup for radiated emission Measurement instrumentation and conditions: Spectrum analyzer: Rohde & Schwarz FSP7 Detector type: Peak detector Measurement time: For all measurements, the emission measurement time (10ms) at one frequency is longer than the test software loop duration. Pre-Amplifier: internal Data generation software: Rohde & Schwarz EMIPAK 9950 Environment: temperature 23°C ± 5°C Supply: nominal voltage ± 5% For all measurements the noise floor is at least 6dB below the limit. Spectrum Analyzer Frequency range TEM 150 Ω RBW 150 kHz to 30 MHz 30 MHz to 200 MHz 200 MHz to 1000 MHz 10kHz Sweep time* ts = NP ⋅ LT ⋅ FR RBW *) NP = number of points; LT = loop time; FR = frequency range Test Report 7 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 3 Emission Limit Curves For reference purpose, all measured emission spectra are enhanced by the “external digital bus systems” limit curves taken from the BISS specification. Figure 4 introduces these 3 limit curves: Conducted emission 150 Ω, limit for supply pin emission (Supply Conducted), • Conducted emission 150 Ω, limit for port pin emission (I/O Conducted), • Radiated emission, limit for Mini TEM cell emission (Radiated). dBµV • 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 I/O Conducted Supply Conducted Radiated 0,1 1,0 10,0 100,0 1000,0 Frequency/MHz Figure 4 BISS limit curves in logarithmic scale If the measured emission stays below the respective limit, the measured supply or signal net is treated as “clean”. If the measured emission violates the respective limit for one or more frequencies, some more care must be taken for an EMC-friendly PCB layout. Infineon strongly recommends to use all microcontroller hardware settings provided for reduction of electromagnetic emission (as long as your required system performance permits): • Reduce system clock frequency • Disable unused clocks (e.g. CLKOUT [automatically disabled after reset]) • Reduce output pad driver strength (up to 6 settings available for all port pins) [1] • Consider Infineon’s general and product-specific PCB design guidelines which propose optimized power supply layout and decoupling concept [2] [3] Test Report 8 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 4 Microcontroller Operation during Test To get a realistic impression of the microcontroller’s emission potential, so called “application-typical” settings have been applied during the tests. This means: • CPU and all functional modules are active • CLKOUT is disabled • High-speed interfaces are active (e.g. SPI @ 5 MBit/s, ASC / LIN / CAN @ 500 kBit/s) • Other I/Os operate at lower data rates • All output pad driver strengths are selected according their data rates, driving 22pF load. Since most pins cannot be configured individually, some more pins of the same ports are also forced to stronger state even if this would be desired only for one or two port pins. This is the ratio in percent of driver settings (All GPIOs without ADC channels) used for the emission tests: Driver strength PG-LQFP-64 PG-LQFP-100 PG-LQFP-144 Weak 57 % 50 % 32 % Medium 0 0 0 Strong-soft 0 14 % 9% Strong-medium 0 0 0 Strong-sharp 40 % 34 % 58 % Extra-strong 3% 2% 1% • External memory bus (if available) is not used • Execution from on-chip flash memory In addition, a worst-case operation scenario is measured for reference purpose: • CLKOUT is activated for worst-case operation. • The external memory bus is never enabled since it is not recommended to be used because of significant occupation of ports and execution slow-down by wait-states. The power supply system is in all cases configured as follows: • VDDP and VDDP1 supplies are 5.0 V • VDDI and VDDIM are supplied by internal voltage regulator The crystal frequency is 8 MHz in all cases. Test Report 9 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 5 Application-Typical Emission Tests General remarks: The microcontroller is running in “application-typical” mode. The general hardware settings are described in chapter 4. The I/O supply voltage VDDP is set to 5.0 V. Any emission peaks rising from lower noise floor in the 900 MHz range result from GSM transmitters and are not related to microcontroller operation. Summary of emission behaviour in application-typical operating mode: 1) VDDI reflects the emission resulting from the core clock tree, i.e. the harmonics are multiples of the system clock frequency. However, since all measured microcontroller derivatives use an internal volage regulator, the VDDI net has no connection to an external voltage regulator. The only component connected to VDDI is the decoupling capacitor. Thus no effective antenna structure to radiate emission from VDDI should be provided on the PCB, and theVDDI emission should not be system-relevant. 2) Both power supplies VDDI and VDDP show highest emission on the smallest device (PG-LQFP-64). Reason is that this package provides no e-pad for central VSS solder connection. From this result, it is advisable to solder the VSS e-pad when using PG-LQFP-100 and PG-LQFP-144 devices to reduce emission. 3) Core noise coupling from VDDI to VDDP is efficiently suppressed in PG-LQFP-100 and PG-LQFP-144 devices, but some coupling happens on the PG-LQFP-64 package which is again due to the missing VSS e-pad, which connects all VSS pins (from core and I/O domain) with very low impedance and thus offers a better common reference point for the VDDI and VDDP noise. 4) I/O emission is – in contrast to supply emission – visible up to 1 GHz. Reason is that there is no available on-chip space to design decoupling capacitors for the VDDP system. Also I/O pins cannot be decoupled on chip nor on PCB. This is the reason why the BISS limit curve for I/O emission is defined 10 dB above the supply limit curve. Please note that all I/O pins where emission is measured are inactive (i.e. not switching) pads which are set to state “output low”. From the emission shape, special care on PCB layout might be necessary to prevent I/O noise from being coupled to any antenna structures on the PCB. Advisable PCB layout rules are: a. Keep signal traces short b. Route signal traces as micro-strip or stripline (ground shielding) c. Avoid signal trace vias through power or ground planes d. Use lowest permissable data rates e. Use weakest permissable pad driver settings For further recommendations please refer to Infineon’s application notes listed in chapter 7. 5) Summarized, the emissions of all package versions stay below the BISS limits and thus should not cause any severe problems for low-emission PCB design. More care on power routing and noise decoupling should be taken for PG-LQFP-64 derivatives. 6) Radiated emission stays far below BISS limits – there is no indication of any significant direct radiation from chip or package, independent from the package size. 7) The part of conducted and radiated emission which is caused by switching I/O noise will decrease when the VDDP voltage is lowered from 5.0 V to 3.3 V. In this case emission peaks may be reduced by ca. 6 dB. Test Report 10 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on core supply net VDDI; system clock is 20 MHz. PG-LQFP-64 CE VDDI 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 5 PG-LQFP-64; Application; VDDI conducted; 20 MHz PG-LQFP-100 CE VDDI 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 6 PG-LQFP-100; Application; VDDI conducted; 20 MHz PG-LQFP-144 CE VDDI 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 7 Test Report PG-LQFP-144; Application; VDDI conducted; 20 MHz 11 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on core supply net VDDI; system clock is 80 MHz. PG-LQFP-64 CE VDDI 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 8 PG-LQFP-64; Application; VDDI conducted; 80 MHz PG-LQFP-100 CE VDDI 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 9 PG-LQFP-100; Application; VDDI conducted; 80 MHz PG-LQFP-144 CE VDDI 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 10 Test Report PG-LQFP-144; Application; VDDI conducted; 80 MHz 12 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on I/O supply net VDDP; system clock is 20 MHz. PG-LQFP-64 CE VDDP 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 11 PG-LQFP-64; Application; VDDP conducted; 20 MHz PG-LQFP-100 CE VDDP 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 12 PG-LQFP-100; Application; VDDP conducted; 20 MHz PG-LQFP-144 CE VDDP 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 13 Test Report PG-LQFP-144; Application; VDDP conducted; 20 MHz 13 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on I/O supply net VDDP; system clock is 80 MHz. PG-LQFP-64 CE VDDP 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 14 PG-LQFP-64; Application; VDDP conducted; 80 MHz PG-LQFP-100 CE VDDP 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 15 PG-LQFP-100; Application; VDDP conducted; 80 MHz PG-LQFP-144 CE VDDP 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 16 Test Report PG-LQFP-144; Application; VDDP conducted; 80 MHz 14 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on I/O net at upper package edge; system clock is 20 MHz. PG-LQFP-64 CE P10.14 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 17 PG-LQFP-64; Application; P10.14 conducted; 20 MHz PG-LQFP-100 CE P1.4 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 18 PG-LQFP-100; Application; P1.4 conducted; 20 MHz PG-LQFP-144 CE P9.5 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 19 Test Report PG-LQFP-144; Application; P9.5 conducted; 20 MHz 15 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on I/O net at upper package edge; system clock is 80 MHz. PG-LQFP-64 CE P10.14 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 20 PG-LQFP-64; Application; P10.14 conducted; 80 MHz PG-LQFP-100 CE P1.4 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 21 PG-LQFP-100; Application; P1.4 conducted; 80 MHz PG-LQFP-144 CE P9.5 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 22 Test Report PG-LQFP-144; Application; P9.5 conducted; 80 MHz 16 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on I/O net at left package edge; system clock is 20 MHz. PG-LQFP-64 CE P6.1 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 23 PG-LQFP-64; Application; P6.1 conducted; 20 MHz PG-LQFP-100 CE VAREF 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 24 PG-LQFP-100; Application; VAREF conducted; 20 MHz PG-LQFP-144 CE VAREF 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 25 Test Report PG-LQFP-144; Application; VAREF conducted; 20 MHz 17 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical conducted emission on I/O net at left package edge; system clock is 80 MHz. PG-LQFP-64 CE P6.1 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 26 PG-LQFP-64; Application; P6.1 conducted; 80 MHz PG-LQFP-100 CE VAREF 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 27 PG-LQFP-100; Application; VAREF conducted; 80 MHz PG-LQFP-144 CE VAREF 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 28 Test Report PG-LQFP-144; Application; VAREF conducted; 80 MHz 18 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical radiated emission; system clock is 20 MHz. PG-LQFP-64 RE 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 29 PG-LQFP-64; Application; radiated; 20 MHz PG-LQFP-100 RE 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 30 PG-LQFP-100; Application; radiated; 20 MHz PG-LQFP-144 RE 20MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 31 Test Report PG-LQFP-144; Application; radiated; 20 MHz 19 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Application-typical radiated emission; system clock is 80 MHz. PG-LQFP-64 RE 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 32 PG-LQFP-64; Application; radiated; 80 MHz PG-LQFP-100 RE 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 33 PG-LQFP-100; Application; radiated; 80 MHz PG-LQFP-144 RE 80MHz APP 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 34 Test Report PG-LQFP-144; Application; radiated; 80 MHz 20 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 6 Worst-Case Emission Tests The microcontroller is running in “worst-case” mode. This means that in addition to the “application-specific” settings, CLKOUT function is activated on port pin P2.8. The toggle rate on CLKOUT is equal to the system clock (20MHz or 80MHz). The CLKOUT pin is set to the strongest driver setting “extra-strong”. If CLKOUT is used to drive slower clocks, its P2.8 driver should be configured to weaker settings which significantly reduce the electromagnetic emission; figures 35 and 36 show the emission reduction potential by pad driver scaling for the two cases VDDP=5.0V and VDDP=3.3V. dBµV COMPARISON: XC2267/87, Conducted Emission Measurement at VDDP0 fsys=var., fosc=16MHz, VDDP0=5.0V, Cload=10pF EXTCLK (P2.8) toggles at var. Frequency / Driver set to var. 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 EXTRA-STRONG 66MHz STRONG-SHARP 40MHz STRONG-MEDIUM 20MHz STRONG-SOFT 20MHz MEDIUM 2MHz WEAK 500kHz 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 35 Emission reduction by pad driver settings for VDDP = 5.0V dBµV COMPARISON: XC2267/87, Conducted Emission Measurement at VDDP0 fsys=var., fosc=16MHz, VDDP0=3.3V, Cload=10pF EXTCLK (P2.8) toggles at var. Frequency / Driver set to var. 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 EXTRA-STRONG 66MHz STRONG-SHARP 40MHz STRONG-MEDIUM 20MHz STRONG-SOFT 10MHz MEDIUM 2MHz WEAK 500kHz 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 36 Test Report Emission reduction by pad driver settings for VDDP = 3.3V 21 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line The weakest possible driver setting for a port pin depends on: • VDDP voltage • External capacitive load • Data rate • Edge-to-period ratio of the data signal • Ambient temperature Figure 37 explains how to use the decision graphs provided in the “Scalable Pads Application Note” [1]. The title of each diagram indicates the conditions (edge/voltage/temperature) for the shown values: Edge is either “T/4” (meaning the rise time and the fall time take each 1/4 of the data rate period) or “T/6” (meaning the rise time and the fall time take each 1/6 of the data rate period). Voltage indicates the I/O pad supply voltage VDDP and is either 3.3 V or 5.0 V. Temperature indicates the ambient temperature and ranges from -40°C up to +125°C. The maximal clock/data rate to meet good signal integrity is given in [MHz] for capacitive loads of 10, 15, 22, 33, 47 pF and driver selections of weak, medium, strong soft/medium/sharp/extra-strong. In Fig. 35, the resulting maximum data rates for the strong-medium driver are marked with red circles: If the strong-medium driver is loaded with 10 pF (which means actually 13 pF including the oscilloscope probe load), it is able to drive a clean 37 MHz signal (under the above mentioned edge/voltage/temperature conditions). An 18 pF load (15+3pF) can be driven at 31 MHz; a 25 pF load (22+3 pF) can be driven at 30 MHz; a 36 pF load (33+3pF) can be driven at 28 MHz; a 50 pF load (47+3 pF) can be driven at 26 MHz. Figure 37 Test Report Explanation of pad driver selection diagram 22 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Figures 38 to 41 show decision graphs for some variations of these parameters. It is assumed that rising edge and falling edge occupy ¼ of the signal period. For a full description and suggestions please refer to [1]. Frequency Limits, Edge=T/4 (5V) -40°C 160 150 140 130 120 Frequency [MHz] 110 100 90 80 70 60 50 40 30 20 10 0 XST 47 SSH 33 SME Driver SSO 22 MED C load [pF] 15 WEA Figure 38 150-160 140-150 130-140 120-130 110-120 100-110 90-100 80-90 70-80 60-70 50-60 40-50 30-40 20-30 10-20 0-10 10 Pad driver selection for VDDP = 5.0 V at -40°C Frequency Limits, Edge=T/4 (5V) 125°C 140 130 120 110 130-140 120-130 110-120 100-110 90-100 80-90 70-80 60-70 50-60 40-50 30-40 20-30 10-20 0-10 Frequency [MHz] 100 90 80 70 60 50 40 30 20 10 0 XST 47 SSH 33 SME Driver SSO 22 MED WEA Figure 39 Test Report C load [pF] 15 10 Pad driver selection for VDDP = 5.0 V at +125°C 23 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Frequency Limits, Edge=T/4 (3,3V) -40°C 130 120 110 100 Frequency [MHz] 90 120-130 110-120 100-110 90-100 80-90 70-80 60-70 50-60 40-50 30-40 20-30 10-20 0-10 80 70 60 50 40 30 20 10 0 XST 47 SSH 33 SME Driver SSO 22 MED WEA Figure 40 C load [pF] 15 10 Pad driver selection for VDDP = 3.3 V at -40°C Frequency Limits, Edge=T/4 (3,3V) 125°C 70 65 60 55 65-70 60-65 55-60 50-55 45-50 40-45 35-40 30-35 25-30 20-25 15-20 10-15 5-10 0-5 Frequency [MHz] 50 45 40 35 30 25 20 15 10 5 0 XST 47 SSH 33 SME Driver SSO 22 MED WEA Figure 41 Test Report C load [pF] 15 10 Pad driver selection for VDDP = 3.3 V at +125°C 24 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line General remarks: The microcontroller is running in “worst-case” mode. The general hardware settings are described in chapter 4. The I/O supply voltage VDDP is set to 5.0V. Any emission peaks in the 900 MHz range rising from lower noise floor result from GSM transmitters and are not related to microcontroller operation. Summary of emission behaviour in worst-case operating mode: 1) I/O noise is efficiently coupled to VDDI, especially in the smallest PG-LQFP-64 package. VDDI emission from CLKOUT is less in PG-LQFP-100 and PG-LQFP-144, where QFP144 shows the lowest CLKOUT coupling. 2) I/O noise is massively seen as VDDP emission. Here, the smallest package QFP64 shows the best damping towards high frequency > 600 MHz. VDDP emission from CLKOUT is less and comparable in PG-LQFP-100 and PG-LQFP-144 with significant high-frequency peaks above 600 MHz. 3) I/O pin emission is highest between 500 and 700 MHz. Reason is that crosstalk from VDDP to the port pins becomes more efficient at higher frequency. Above 700 MHz, the natural damping occurs. Please note that all I/O pins where emission is measured are inactive (i.e. not switching) pads which are set to state “output low”. From the emission shape, special care on PCB layout might be necessary to prevent I/O noise from being coupled to any antenna structures on the PCB. Advisable PCB layout rules are: b. Keep signal traces short c. Route signal traces as micro-strip or stripline (ground shielding) d. Avoid signal trace vias through power or ground planes e. Use lowest permissable data rates f. Use weakest permissable pad driver settings For further recommendations please refer to Infineon’s application notes listed in chapter 7. 4) Summarized, the activation of any high-speed clock or data driver with strongest settings rise real layout challenges for low-emission PCB layout. Thus high-speed signals should be avoided. If this is not possible, they should be limited by clock rate and driver strength. Hints for the expected emission reduction have been provided above. In case of strong high-speed signals, the smallest package PG-LQFP-64 shows lowest emission on I/O signals; minor differences are visible between PG-LQFP100 and PG-LQFP-144. The bigger packages show significant high-frequency emission on I/O nets. 5) Radiated emission exceeds the BISS limits. The level can be significantly decreased when operating I/Os at lower speed and select weaker driver strengths, as discussed above. All packages show similar emission, where the PG-LQFP-64 stays ca. 6dB below the bigger ones. Test Report 25 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on core supply net VDDI; system clock is 20 MHz. PG-LQFP-64 CE VDDI 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 42 PG-LQFP-64; Worst-case; VDDI conducted; 20 MHz PG-LQFP-100 CE VDDI 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 43 PG-LQFP-100; Worst-case; VDDI conducted; 20 MHz PG-LQFP-144 CE VDDI 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 44 Test Report PG-LQFP-144; Worst-case; VDDI conducted; 20 MHz 26 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on core supply net VDDI; system clock is 80 MHz. PG-LQFP-64 CE VDDI 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 45 PG-LQFP-64; Worst-case; VDDI conducted; 80 MHz PG-LQFP-100 CE VDDI 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 46 PG-LQFP-100; Worst-case; VDDI conducted; 80 MHz PG-LQFP-QFP100 CE VDDI 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 47 Test Report PG-LQFP-144; Worst-case; VDDI conducted; 80 MHz 27 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on I/O supply net VDDP; system clock is 20 MHz. PG-LQFP-64 CE VDDP 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 48 PG-LQFP-64; Worst-case; VDDP conducted; 20 MHz PG-LQFP-100 CE VDDP 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 49 PG-LQFP-100; Worst-case; VDDP conducted; 20 MHz PG-LQFP-144 CE VDDP 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 50 Test Report PG-LQFP-144; Worst-case; VDDP conducted; 20 MHz 28 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on I/O supply net VDDP; system clock is 80 MHz. PG-LQFP-64 CE VDDP 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 51 PG-LQFP-64; Worst-case; VDDP conducted; 80 MHz PG-LQFP-100 CE VDDP 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 52 PG-LQFP-100; Worst-case; VDDP conducted; 80 MHz PG-LQFP-144 CE VDDP 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 53 Test Report PG-LQFP-144; Worst-case; VDDP conducted; 80 MHz 29 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on I/O net at upper package edge; system clock is 20 MHz. PG-LQFP-64 CE P10.14 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 54 PG-LQFP-64; Worst-case; P10.14 conducted; 20 MHz PG-LQFP-100 CE P1.4 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 55 PG-LQFP-100; Worst-case; P1.4 conducted; 20 MHz PG-LQFP-144 CE P9.5 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 56 Test Report PG-LQFP-144; Worst-case; P9.5 conducted; 20 MHz 30 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on I/O net at upper package edge; system clock is 80 MHz. PG-LQFP-64 CE P10.14 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 57 PG-LQFP-64; Worst-case; P10.14 conducted; 80 MHz PG-LQFP-100 CE P1.4 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 58 PG-LQFP-100; Worst-case; P1.4 conducted; 80 MHz PG-LQFP-144 CE P9.5 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 59 Test Report PG-LQFP-144; Worst-case; P9.5 conducted; 80 MHz 31 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on I/O net at left package edge; system clock is 20 MHz. PG-LQFP-64 CE P6.1 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 60 PG-LQFP-64; Worst-case; P6.1 conducted; 20 MHz PG-LQFP-100 CE VAREF 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 61 PG-LQFP-100; Worst-case; VAREF conducted; 20 MHz PG-LQFP-144 CE VAREF 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 62 Test Report PG-LQFP-144; Worst-case; VAREF conducted; 20 MHz 32 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case conducted emission on I/O net at left package edge; system clock is 80 MHz. PG-LQFP-64 CE P6.1 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 63 PG-LQFP-64; Worst-case; P6.1 conducted; 80 MHz PG-LQFP-100 CE VAREF 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 64 PG-LQFP-100; Worst-case; VAREF conducted; 80 MHz PG-LQFP-144 CE VAREF 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 65 Test Report PG-LQFP-144; Worst-case; VAREF conducted; 80 MHz 33 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case radiated emission; system clock is 20 MHz. PG-LQFP-64 RE 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 66 PG-LQFP-64; Worst-case; radiated; 20 MHz PG-LQFP-100 RE 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 67 PG-LQFP-100; Worst-case; radiated; 20 MHz PG-LQFP-144 RE 20MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 68 Test Report PG-LQFP-144; Worst-case; radiated; 20 MHz 34 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line Worst-case radiated emission; system clock is 80 MHz. PG-LQFP-64 RE 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 69 PG-LQFP-64; Worst-case; radiated; 80 MHz PG-LQFP-100 RE 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 70 PG-LQFP-100; Worst-case; radiated; 80 MHz PG-LQFP-144 RE 80MHz WC 60 55 50 45 40 dBµV 35 30 25 20 15 10 5 0 0 100 200 300 400 500 600 700 800 900 1000 Frequency/MHz Figure 71 Test Report PG-LQFP-144; Worst-case; radiated; 80 MHz 35 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 7 References These documents can be downloaded from the Infineon microcontroller internet pages: http://www.infineon.com/cms/en/product/channel.html?channel=db3a30431c69a49d011c8a7c069203dc [1] Scalable pads application notes Æ ap1612011_scalable-pads-xc2000-xe166.pdf [2] XC2000 and XE166 Family EMC guidelines for PCB design Æ AP1611621_XC2000_XE166_PCB.pdf http://www.infineon.com/cms/en/product/channel.html?channel=ff80808112ab681d0112ab6b2dfc0756 [3] General EMC guidelines for PCB design Æ ap2402633_EMC_Guidelines.pdf These documents are available on request: • Generic IC EMC Test Specification (“BISS paper”), open copyright 2004 by Robert Bosch GmbH, Infineon Technologies AG, Continental AG (former Siemens-VDO) • Electromagnetic Emission Test Report for XC2000 and XE166 Family Base Line Derivatives in PG-LQFP-64 Package, Sept. 2008 • Electromagnetic Emission Test Report for XC2000 and XE166 Family Base Line Derivatives in PG-LQFP-100 Package, July 2008 • Electromagnetic Emission Test Report for XC2000 and XE166 Family Base Line Derivatives in PG-LQFP-144 Package, July 2008 Test Report 36 V1.0, 2009-04 AP16167 Electromagnetic Emission Summary XC2000 and XE166 Family Derivatives / Base Line 8 Executive Summary 1) Conducted and radiated emissions of all XC2000 and XE166 Family Base Line Derivatives stay below the BISS limits for application-typical operation up to 80 MHz and thus should not cause any severe problems for low-emission PCB design. 2) PG-LQFP-64 derivatives show higher emission than PG-LQFP-100 and PG-LQFP-144 due to missing VSS e-pad. Thus more care on power routing and noise decoupling should be taken for PCBs using PG-LQFP-64 devices. For PG-LQFP-100 and PG-LQFP-144 it is strongly recommended to solder the VSS e-pad. 3) Since all microcontrollers use internal volage regulators, the VDDI pins should only be connected to local decoupling capacitors. Thus no effective antenna structure to radiate emission from VDDI should be provided on the PCB, and the VDDI emission should not be system-relevant. 4) Core noise coupling from VDDI to VDDP is efficiently suppressed in PG-LQFP-100 and PG-LQFP-144 devices, but some coupling happens on the PG-LQFP-64 package which is again due to the missing VSS e-pad which offers a low-impedance common reference point for the VDDI and VDDP noise. 5) I/O emission (i.e. noise propagation through inactive pads) is – in contrast to supply emission – visible up to 1 GHz. Reason is that there is no available on-chip space to design decoupling capacitors for the VDDP system. Also I/O pins can neither be decoupled on chip nor on PCB. Special care on PCB layout is advisable to prevent I/O noise from being coupled to any antenna structures on the PCB: a. Keep signal traces short b. Route signal traces as micro-strip or stripline (ground shielding) c. Avoid signal trace vias through power or ground planes d. Use lowest permissable data rates e. Use weakest permissable pad driver settings For further recommendations please refer to Infineon’s application notes listed in chapter 7. 6) Conducted and radiated emission exceeds the BISS limits if the strongest pad drivers are selected for high-speed signals. The level can be significantly decreased when operating I/Os at lower speed and select weaker driver strengths, as discussed above. All packages show similar emission, where the smallest PG-LQFP-64 stays ca. 6 dB below the bigger ones. 7) I/O noise is visible on VDDP, but also coupled to VDDI, especially in the smallest PG-LQFP-64 package. VDDI emission from CLKOUT is less in PG-LQFP-100 and PG-LQFP-144, where the PG-LQFP-144 shows the lowest CLKOUT coupling. I/O pin emission is highest between 500 MHz and 700 MHz. Reason is that crosstalk from VDDP to the port pins becomes more efficient at higher frequency. Above 700 MHz, the natural damping occurs. 8) The part of conducted and radiated emission which is caused by switching I/O noise will decrease when the VDDP voltage is lowered from 5.0 V to 3.3 V. In this case emission peaks may be reduced by ca. 6 dB. 9) The activation of any high-speed clock or data driver with strongest settings rise real layout challenges for low-emission PCB layout. Thus high-speed signals should be avoided. If this is not possible, they should be limited by clock rate and driver strength. In case of strong high-speed signals, the smallest package PG-LQFP-64 shows lowest emission on I/O signals; minor differences are visible between PG-LQFP-100 and PG-LQFP-144. Test Report 37 V1.0, 2009-04 w w w . i n f i n e o n . c o m Published by Infineon Technologies AG