AFTERMARKET COMPONENT TEST STANDARDS FOR THE NORTH AMERICAN MARKET

Transcription

1 AFTERMARKET COMPONENT TEST STANDARDS FOR THE NORTH AMERICAN MARKET Martin O Hara 1 and Rick Goodwin 2 1 Trafficmaster, University Way, Cranfield, MK43 0TR, England 2 Solectron Invotronics, 365 Passmore Ave, Toronto ON M1V 4B3, Canada Abstract: This paper looks at the minimum testing required, the basic testing recommended and what is essentially good design practice for any automotive component to be sold in the North American market for aftermarket fit. The SAE EMC standards (SAE J1113), that are the usually quoted standards for the North American market, are compared with their international equivalents and recent updates discussed. Some of the SAE standards that have no ISO or CISPR equivalent are also examined in brief, particularly where these have an OEM test origin. The paper will conclude with a matrix of recommended SAE standards that are good practice to apply to aftermarket automotive components intended for the North American market and some guidelines on which to apply and what severity standards might be considered due diligence. Introduction Europe has a well developed and well known minimum EMC standard for automotive components sold as aftermarket products; the Automotive EMC Directive 95/54/EC [1] (to be replaced by 2004/104/EC [2]). In North America the situation is less well defined for aftermarket automotive components especially. The applicable EMC standards for North America are usually covered in the FCC regulations, however, the usual FCC part 15 standard [3] appears to specifically exclude components for use in the automotive environment; Section Exempted devices (a) A digital device utilized exclusively in any transportation vehicle including motor vehicles and aircraft. This appears to be a get-out-of-testing card for US aftermarket vehicle products, although the manufacturer is strongly recommended to meet the technical requirements of FCC Part 15. Consequently some testing may be advisable, if not mandatory, and not necessarily via a fully complaint test facility. However, given the litigious nature of North America society in general, surely some proof of due diligence would be a good idea? Society of Automotive Engineers (SAE) Standards Although most people first think of the FCC or CSA when considering EMC for the North American market, the automotive side of EMC has been practically ignored by these standards bodies in favour of the SAE. The SAE standards originally came together to try to tie up the disparities and some contradictions in the EMC specifications of the Big 3 American motor manufacturers (General Motors, Ford and Chrysler originally) and some of the international standards available (mainly CISPR-12, CISPR-25, ISO 7637 and ISO 10605). The standards of the Big 3 were not always easy to obtain (some are still difficult) and the SAE were able to put together, in an easily obtainable form [4], a series of standards that brought together some of the most common global test methods with variable limit levels to suit most vehicle manufacturers (VMs). The SAE also had good support form the American automotive industry and the standards quickly became the acceptable norm for EMC testing of automotive products. Being independent of the VMs and the international standards bodies allowed the SAE to introduce methods sooner than other standards might adapt, for example the SAE introduced ESD testing of components not fitted to a vehicle before the ISO update to reflect handling conditions (and for the 9 years this standard remained a technical report only). This also brought some problems in that the SAE standards were also often slightly behind any significant changes to the major international standard such as CISPR-25 update in 2002, but these were quickly incorporated into the SAE documents [5]. The SAE have also gone significantly further than many VM or other international standard bodies in their work, encompassing many whole vehicle tests outside of the usual EMC remit (e.g. low frequency magnetic fields) and even having standards for testing IC s for automotive use (SAE J1752 series). Automotive EMC Conference 2005 Page 1 11 th May 2005

2 There are two problems with the SAE standards in their implementation, firstly they are developed primarily to permit testing for vehicle fit components installed by the VM, they are not aimed at aftermarket automotive components suppliers. Secondly, but most importantly, these standards have no authority in law, they are not necessarily recognised by the FCC or CSA as suitable product standards for demonstrating EMC compliance. Consequently, despite their wide use and almost universal acceptance by all suppliers of electronic components into the North American market, the SAE standard have no legal or regulatory standing. This situation is a deliberate result of ANSI delegating the automotive EMC standards writing to the SAE rather than the FCC (an undoubtedly sensible decision) and a long standing agreement between the FCC and US automobile manufacturing industry to use the SAE standards [6,7]. This was a deliberate policy to reduce the amount of regulatory requirements of an industry that had been successfully self-regulating much longer than the majority of the electronics market place. This does provide the aftermarket and overseas supplier to the US market with a rather convenient EMC loophole. A Pragmatic Approach Although a supplier may be able to escape without any EMC testing, the litigious nature of the North American market and the ultimate desire of most automotive component suppliers to eventually be VM line fit should ensure that most suppliers take a more pragmatic approach to EMC testing. The following are what these authors would recommend as a basic pre-requisite test set for most electronic products for sale into North America as aftermarket automotive components. Complying with FCC Part 15 The need to comply with FCC part 15 is quite explicitly exempt for all transport electronics. That might sound pretty final but if a retailer in the US decides that they will only sell FCC part 15 compliant products the route to market for non-compliant product (regardless of the actual legality of the requirement) will be blocked. Therefore if selling through retail (i.e. primarily for user fit) then FCC compliance may be an essential requirement for market access 1. Initially the testing to FCC Part 15 will seem a little alien to anyone only used to automotive EMC testing, firstly there is no ground plane on the table and secondly the LISN/AMN used is the commercial 50uH/50Ù type (not the usual automotive 5uH/50Ù). It is consequently very difficult to gauge from any automotive testing (radiated emissions or conducted emissions) probability of passing the FCC part 15 test. Figure 1: Radiated Emissions Testing to FCC-part 15 (20MHz-1GHz) on Rotating Turntable (photo courtesy of Trafficmaster/3C Test Ltd). Figure 2: Results of FCC part 15 (green) and CISPR-25 (blue) Radiated Emissions testing on the same product with FCC Class A limit Line shown, 1 One of the authors has come across such retail restrictions in the European market with a self-contained battery operated product that an automotive components retailer refused to sell without an e mark. Automotive EMC Conference 2005 Page 2 11 th May 2005

3 Test harnesses and supply lead need to be 1.5m long, which is common with many automotive tests. The device under test (DUT) and support equipment are rotated and height scanned for maximum levels during radiated emissions testing (figure 1), certainly different from the static test nature of most automotive component EMC tests. One other significant different is the DUTto-antenna distance; 3m or 10m for FCC part 15 compared to 1m for most automotive testing. The results tend to be normalised for emissions (presented in dbuv/m), however, this is not a true levelling otherwise the 3m and 10m limits would be identical. The FCC part 15 3m results are, however, relatively close in overall magnitude of the signal levels to those obtained via CISPR-25 [8] tested units received signal levels (figure 2). Consequently a general indication of compliance may be obtained from a quick comparison of an automotive radiated emissions test with the 3m FCC part 15 limit lines overlaid, but this would not be adequate for a compliance test. In most instances it should be possible to avoid having to test for conducted emissions under the proviso that the DUT is only intended for vehicle use and can never be directly connected to the AC mains supply (figure 3). However, if necessary the results are not dissimilar to those obtained under a CISPR-25 test condition for power line conducted emissions (figure 4). The results obtained under an FCC part 15 test are generally lower than the CISPR-25 results due to the shorter power harness used for CISPR-25 testing (0.2m compared to 1.5m for FCC part 15). The LISN should have minimal impact since this should present a similar impedance to the DUT and test network, it is only between the battery and DUT that there is an increased impedance, but this will have negligible impact on the results as a vehicle battery has relatively poor transient response to the higher frequencies where this might otherwise impact the measured data. Radiated Emissions Figure 4: Conducted Emissions test data for ESA tested to FCC part 15 (green) and CISPR-25 (blue). Although we have already specified that radiated emissions under FCC part 15 may be a commercial requirement, for a true automotive product testing to SAE J (CISPR-25) would be the preferred method, even for aftermarket products. The main differences have already been outlined above, and although similar results may be obtained, separate testing should be considered, especially if wanting to submit for approval to a VM at some point as any VM will not consider FCC part 15 results as having any relevance to the vehicle environment. Conducted Emissions Figure 3: Conducted Emissions Testing to FCC-part 15 (150kHz- 30MHz) via AC AMN (50uH/50W) The SAE standard SAE J is identical to CISPR-25 test method for both conducted and radiated emissions. This would be the preferred choice for an automotive component above the FCC part 15 test discussed above, the arguments for this as for radiated emissions, have already been discussed. Although the standard includes both power and control line emission testing, for an aftermarket device power line only is more typically tested. Automotive EMC Conference 2005 Page 3 11 th May 2005

4 The SAE also have a conducted transient emissions standard (J ) that utilises the transient emission test methods of ISO (although not explicitly referenced in the SAE document). Although this method is relatively well adopted by the automotive industry, it is not a popular test for aftermarket devices, although it was incorporated by the recent updated EU Automotive EMC directive (2004/104/EC) so will be a requirement in Europe for such products. The SAE J method utilises a fast oscilloscope (>400MHz analogue BW) to catch transients and analyse these in the time domain. The technique is relatively low cost and simple to perform and lacks much of the detail normally observed from the more familiar CISPR-25 power line conducted emissions test. As a low cost test to ensure basic compatibility it may be adequate, but again if wishing to gain VM acceptance the more generally accepted SAE J (CISPR- 25) power line conducted emissions test would be preferable. Immunity to Electrostatic Discharge (ESD) Figure 5: Conducted transient emission in the time domain. The SAE standard for ESD (SAE J ) is virtually identical to the latest IEC equivalent (ISO 10605:2001 [9]). The test method covers both powered (i.e. in-situ) and unpowered (i.e. handling) conditions. The test is relatively straightforward and covers both air and contact discharge to all metal connectors and component seams on a product. It is well respected, well used within VM EMC standards, although nearly all quote the IEC version. While it should be expected that most automotive components are tested for ESD, a review of the data suggests that very little is actually performed on aftermarket products. In Europe even the EU Automotive Directive does not cover this aspect, leaving it to the VMs to again self regulate. The general North American electronics market also seems to be somewhat blasé about ESD testing. Figure 6: Automotive component ESD test (photo courtesy of BRE Testing Ltd) A low severity test for handling in particular would seem to be an obvious requirement for an aftermarket automotive product but it would appear to be a little used test outside of line-fit components. Immunity to Conducted Transients As with ESD, the SAE standards for transient immunity (SAE J : Power lead and 12: Coupled) are similarly identical to their IEC equivalents (ISO and 3 respectively [10]). These standards are again well respected and almost universally adopted within the VM standards. Automotive EMC Conference 2005 Page 4 11 th May 2005

5 The tests contained in these standards, particularly the power lead tests, are basic electrical compatibility with a vehicle electrical environment. These are the most essential of tests that an aftermarket product should have to comply with, if a product can not meet these tests it is unsuitable to be connected to the power network within a vehicle. The ISO 7637 standards contain a series of tests, so unlike some of the other standards are not a single one-shot type test. Consequently set-up time is relatively long and some of the tests require extensive repetitive pulses (5000 minimum is some cases). As a result it may be tempting to take shortcuts with some of this testing, reducing the number of pulses or duration of testing to minimise cost. However, this could be a false economy as failure to comply with these tests are likely to result in large scale customer returns and warranty claims that are most likely not going to be observed from any other test described here. There are also standards produced by the SAE covering conducted immunity to continuous ripple voltage over the frequency ranges 30Hz-250kHz (SAE J1113-2) and 250kHz to 500MHz (SAE J1113-3). The ripple voltage applied is relatively low (up to 3V peak-peak) and these have some correlation to ISO [11], although not derived from this IEC standard. These tests are not particularly difficult to actually pass, however, few aftermarket products would be expected to have been tested to this level and the expense of these tests are difficult to justify for most aftermarket components. The SAE also have 60Hz AC magnetic and electric field tests that are not derived from an IEC based standards (SAE J and 26 respectively). These tests are relatively benign but again would be difficult to justify the cost for most aftermarket devices. Immunity to Radiated Fields The SAE standards for radiated immunity testing (including BCI) are similes of the ISO series of standards [12]. The single standard that would be familiar to most test engineers and technicians is probably SAE J Radiated Immunity absorber lined chamber (equivalent to ISO ). This will seem similar to both the CISPR-25 and current EU Automotive Directive (95/54/EC) test set-up for radiated immunity. Consequently as a first choice of test method to employ this would be preferable for radiated immunity. The frequency range covered by this standard is exceptionally wide (10kHz to 18GHz), however, most other standards for this type of testing (including VM standards) generally cover from 30MHz up to 3GHz, some still restrict to 1GHz at the high end. A significant Figure 7: Immunity to power line transients test set-up (photo courtesy of 3C Test Ltd/Trafficmaster) Figure 8: BCI Testing of Automotive Component (photo courtesy of 3C Test Ltd) difference in testing occurs at approximately 800MHz-1GHz where the modulation scheme usually changes from amplitude modulation (AM) to pulsed modulation (PM) scheme to reflect the differences in operation of product in these bands; analogue radio below 800MHz and primarily digital transmissions Automotive EMC Conference 2005 Page 5 11 th May 2005

6 above 800MHz (e.g. GSM, Bluetooth). A continuous wave (CW) field is also usually applied across the complete frequency range. The popular BCI method is also covered in the SAE standard J (ISO ), covering the frequency range 1MHz-400MHz. BCI is popular for several reasons; it s relative low cost, it s ability to be performed on-vehicle easily and it s relative simplicity (although the calibration can be awkward). This method is used for the lower frequency bands hence with CW and AM only. The SAE have also have taken on-board tests from several VM standards that incorporate the reverberation chamber test method into SAE J (based on IEC [13]). Outside of the VMs, this is one of the first automotive specific test standards to recognise the reverberation chamber method. The technique has been gaining momentum over the last few years especially for generating high field strengths (e.g. 600V/m or more) and for testing large objects (e.g. whole vehicles, missiles, aircraft). This would be slightly excessive testing for a purely aftermarket product but is becoming more common for VM line fit item, particularly safety critical components. The SAE standard limits the technique to 2GHz, but other practitioners have used this technique at significantly higher frequencies, the main difficulty is obtaining field uniformity and repeatability (site-to-site) at higher frequencies. For aftermarket components BCI might be considered as a suitable method for single test technique, or coupled with anechoic chamber testing to cover say 1MHz to 3GHz. Reverberation testing though is still in its infancy and few test facilities can offer this method, hence would not be expected to be used on an aftermarket device. Severity Levels Figure 9: Reverberation Testing of Whole Vehicle Radiated Immunity (photo courtesy of ETS-Lindgren) Most of the SAE (and their IEC equivalents) have a severity level for each test that can be selected by the supplier. This might seem a little strange to some, but there is sense in not over testing non-critical devices when using automotive test methods, hence the higher severity levels are primarily for safety critical equipment (e.g. ABS, engine management) and the lower severity for comfort and entertainment features (e.g., in-car entertainment, satellite navigation, HVAC). It is relatively frightening, but not unheard of, to find people fitting aftermarket ABS and ignition circuits (if not full engine management) to their vehicles. Any components supplied for this sector clearly have to meet the most stringent levels of safety and personally it would be suggested that only VM approved parts should ever be used for these functions. The VM approved devices should have gone though the most stringent EMC testing applicable and usually are designed and tested to line-fit levels. The majority of aftermarket devices fit into the comfort and entertainment section, hence the lower severity levels can be applied. When testing aftermarket components for emission levels to the above standards, it is probably worthwhile performing the tests and then determining which severity level the product can meet (i.e. overlaying the limit lines on the test results and deciding what level the device is capable of meeting). Immunity testing is more difficult and can be expensive to continuously test until a failure occurs, therefore most aftermarket device (when using these test) go for the lowest levels of severity. Automotive EMC Conference 2005 Page 6 11 th May 2005

7 SAE or IEC Standards? This is always a difficult issue as it varies depending on which side of the Atlantic you re having the testing done. Outside of North America few test houses will be sufficiently familiar with the SAE standards and nomenclature to be able to confidently perform test to the SAE standards, even though they are identical to many IEC methods. Similarly in the USA many test houses are more familiar with the SAE testing and slight nuances between an SAE and IEC test may cause an issue. It should always be possible to hedge your bets with these tests and have both standards quoted in the report. If not most competent VM EMC Engineers at least will be familiar with the cross references hence it will not be a barrier to VM acceptance within North America to have the SAE standards listed. However, outside of North America the IEC standards are undoubtedly more universally acknowledged and consequently if looking for a global market for your products the IEC standards would be the most obvious choice when specifying the test. Summary The tables below summaries the relevant SAE standards and where applicable their IEC equivalents. Table 1: SAE Component EMC Emissions Standards SAE Part Test Type Frequency Range Comparable IEC Standard J On-Board Antennae 150kHz - 1GHz CISPR-25 J Conducted Transients Not applicable ISO Table 2: SAE Component EMC Immunity Standards SAE Part Test Type Frequency Range Comparable IEC Standard J Introduction Not applicable ISO J Conducted Immunity 30Hz - 250kHz Not applicable J Conducted Immunity 250kHz - 500MHz ISO J Bulk Current Injection 1MHz - 400MHz ISO J Power Lead Immunity Not applicable ISO J Coupled Immunity Not applicable ISO J Electrostatic Discharge Not applicable ISO J Radiated Immunity 30MHz - 18GHz ISO J AC Power Line - Magnetic Field 60Hz Not applicable J Radiated Immunity - Strip-Line 10kHz - 200MHz ISO J Radiated Immunity - TEM Cell 10kHz - 200MHz ISO J Radiated Immunity - Tri-Plate 10kHz - 500MHz Not applicable J AC Power Line - Electric Field 60Hz Not applicable J Radiated Immunity - Reverberation Chamber 500MHz - 2GHz IEC The SAE are intending to drop parts 23 and 25 within the next release of their EMC standard manual [7]. The complete tests above are slightly excessive test regimes for an aftermarket component and the table below summaries the level of testing such a component might expect to undergo if supplied by a professional product supplier and one looking for eventual approval by a VM or adoption as a line-fit item at some time in the future. Automotive EMC Conference 2005 Page 7 11 th May 2005

8 Conclusion To sell aftermarket automotive components into the North American market a supplier can get away with out any formal EMC testing. However, it is not recommended and radiated emissions to FCC part 15 may be a minimum barrier to market access. As a basic test of compatibility with the automotive electrical environment a low severity pass of SAE J (ISO ) should also be performed. These 2 tests would be the absolute minimum that any supplier should perform before considering supplying to the market. Table 3: North American EMC Standards Suggested for Aftermarket Component Testing Conducted Radiated Emissions SAE J FCC part-15 SAE J Immunity SAE J SAE J ESD SAE J Those tests shown in bold are the suggested minimum test set for an aftermarket product. If wishing to show a more professional approach, and wanting at some stage to gain VM approval or line fit, then the additional tests listed above should also be performed. The testing may sound quite extensive for an automotive aftermarket product, but the test costs are actually relatively low compared to most commercial testing, and given the litigious nature of the market place any sign of not taking due diligence will cost more if brought to court. References [1] Commission Directive 95/54/EC of 31 October 1995 adapting to technical progress Council Directive 72/245/EEC on the approximation of the laws of the Member States relating to the suppression of radio interference produced by spark-ignition engines fitted to motor vehicles and amending Directive 70/156/EEC on the approximation of the laws of the Member States relating to the type-approval of motor vehicles and their trailers [2] Commission Directive 2004/104/EC of 14 October 2004 adapting to technical progress Council Directive 72/245/EEC on the approximation of the laws of the Member States relating to the suppression of radio interference produced by spark-ignition engines fitted to motor vehicles and amending Directive 70/156/EEC on the approximation of the laws of the Member States relating to the type-approval of motor vehicles and their trailers [3] FCC Part 15: Radio Frequency Devices, July [4] SAE Surface Vehicle Electromagnetic Compatibility (EMC) Standards Manual, SAE, 1999, ISBN [5] SAE Surface Vehicle Electromagnetic Compatibility (EMC) Standards Manual (HS-3600), SAE, 2004, ISBN [6] A Review of Today s Automotive EMC Standards, Poul Andersen and Kimball Williams, EE- Evaluation Engineering, [7] IEEE 2004 Automotive EMC Event, Ford Events and Conference Center, Dearborn, MI, 3-4 February Automotive EMC Conference 2005 Page 8 11 th May 2005

9 [8] CISPR-25: Limits and methods of measurement of radio disturbance characteristics for the protection of receivers used on board vehicles. [9] ISO 10605: Road vehicles electrical disturbances from electrostatic discharges (2001) [10] ISO 7637: Electrical disturbance by conduction and coupling Part 1: Definitions and general considerations (2002) Part 2: Electrical transient conduction along supply lines only (2004) Part 3: Electrical transient transmission by capacitive and inductive coupling via lines other than supply lines (tba) [11] ISO 16750: Road vehicles - environmental conditions and testing for electrical and electronic equipment Part 1: General Part 2: Electrical loads Part 3: Mechanical loads Part 4: Climatic loads Part 5: Chemical loads Part 6: System integration [12] ISO 11452: Road vehicles electrical disturbances by narrowband radiated electromagnetic energy component test methods Part 1: General and definitions (2001) Part 2: Absorber-lined chamber (1995) Part 3: Transverse electromagnetic mode (TEM) cell (2001) Part 4: Bulk current injection (BCI) (2001) Part 5: Stripline (2002) Part 6: Parallel plate antenna (1997) Part 7: Direct radio frequency (RF) power injection (1995) [13] IEC Testing and measurement techniques Reverberation chamber test methods (2003) Biographical Details Martin O Hara: Electronic Hardware Designer. A graduate of Durham University (Applied Physics and Electronics) and holder of a Masters degree from Manchester Metropolitan University (Instrumentation), chartered electrical engineer (CEng), member of the IEE (MIEE), SAE and registered European Engineer (Eur Ing). Has worked in the Electronics industry for the past 18 years for companies including Ferranti, GEC-Marconi, Newport Components and Motorola. Currently involved in the design of in-vehicle telematic hardware utilising GPRS, GSM, GPS and multi-processor technologies. Author of EMC at Component and PCB Level, contributor to Power Specialist's App Note Book and numerous technical papers, particularly on EMC, automotive electronics and circuit simulation. Rick Goodwin: EMC Test Technician. Employed at Invotronics for 20 years. Performing EMC testing for GM, Ford and Daimler Chrysler for 14 years. Invotronics are currently the only supplier lab in Canada certified to do Chrysler EMC testing, and working towards A2LA AEMCLAP accreditation for the other VMs. Automotive EMC Conference 2005 Page 9 11 th May 2005