On-board diagnostics (OBD) continues to be an important industry topic, technically and legislatively. Do you wonder why so many of the '96 and '97 vehicles needed to have their PCM's “flashed”? Or a better question: What's next in this arena? This article will address these concerns, and will also look at the relationship between OBD II and inspection and maintenance (I/M) programs.
OBD II is different from OBD I in that OBD II is strictly emissions oriented — it will illuminate the malfunction indicator lamp (MIL) whenever it detects a component/system malfunction that could cause emissions to exceed 1.5 times the federal test procedure (FTP) standards for that model year of vehicle. This includes random misfires causing an overall rise in HC emissions, operating efficiency of the catalytic converter dropping below a certain threshold, system detection of air leakage in the sealed fuel system, a fault in the EGR system causing NOx emissions to go up, or failure of a key sensor or other emission control device. In other words, the MIL light may come on even though the vehicle seems to be running normally and there are no real driveability problems.
Future: Beginning January 1, 2002, manufacturers will have to make the info available in a standardized electronic format. Many OEMs are already moving toward electronic information, but in a proprietary format. (For example, to use GM information you need special GM software to read the CD-ROM data files, etc.) Starting in 2002, manufacturers would have to put the information in an SAE standardized format so a single software program could be used to look at any OEM's service information. As a side note, if the EPA's service rule goes through and requires availability through the Internet, CARB will probably accept Internet capability as equivalent to putting the information in this SAE standardized format since any technician with a Web-browser would have access.
Future: Data from OBD II demonstration vehicles show that NOx emissions can be high when HC emissions reach the malfunction threshold. This means that a way of monitoring NOx emissions is coming! To maximize NOx emission reductions, CARB is proposing a NOx-based catalyst monitoring requirement for vehicles in addition to the HC-based requirement. Monitoring technologies, such as the use of a NOx sensor, might be used to meet this proposed requirement. Another method is to evaluate the light-off characteristics of the catalyst using a catalyst temperature sensor. CARB proposes that the NOx efficiency monitoring requirement be phased in with the introduction of vehicles meeting the LEV II standards (2004-2007). The most likely method (expected way for all manufacturers to meet this requirement) is to continue to use dual O2 sensors and monitor the ability of the catalyst to store and release oxygen. This dictates the ability of the catalyst to convert HC or NOx emissions.
Future: Misfire monitors are sensitive; a car misfiring for a minute or two can be enough to turn on the MIL and it can be difficult to recreate that actual misfire and repair. To address this sensitivity, CARB changed the requirement from exceeding the misfire rate for any one 1000 rev block to any four 1000 rev blocks in a single driving cycle (i.e., the misfire has to be around for four times as long as it used to be). The '96, '97 and some '98s use any 1000 rev block while most '98 and newer models use any four 1000 rev blocks.
Future: With the accelerated catalyst light-off strategies, the idle speed control system must function properly during cold conditions to achieve the emission standards. If higher idle speed cannot be achieved due to a fault that is present only during cold operation (such as a sticking or slow idle air control valve), catalyst light-off will be delayed, and emissions can be excessive with no indication to a driver or repair technician. Therefore, CARB believes that confirmation of the accelerated warm-up conditions is necessary.
There is a proposed requirement to monitor the engine operating conditions such as engine speed, air/fuel ratio and ignition timing to verify that the commanded conditions were achieved. The regulatory language would thus specify that the emission/engine control system shall be monitored for achieving the required engine operating conditions necessary for accelerated catalyst warm-up. A manufacturer may develop a diagnostic strategy using catalyst temperature sensors to verify that the conditions to accelerate catalyst light-off were achieved or catalyst light-off temperature was reached within a specified amount of time.
Future: Beginning in the 2000 model year, manufacturers are required to phase in monitoring for small leaks equal to or greater in magnitude than a 0.020 inch diameter orifice. Full compliance with the 0.020 inch requirements on all vehicles is scheduled to take place in the 2003 model year. Manufacturers are performing the 0.020 inch diameter leak check of the fuel and evaporative control system using either a vacuum or pressure strategy. Under both of these methods, the pressure inside the system is monitored over an interval of time. If the pressure or vacuum changes toward ambient at a significant rate, a leak is considered to be present. Between the two methods, pressure-based technologies are generally more reliable in detecting a 0.020 inch leak and have a lower probability of signaling false malfunctions. Therefore, to reliably detect a 0.020 inch diameter leak under reasonable monitoring conditions, vacuum-based technologies must set the malfunction threshold at a level significantly lower than 0.020 inch (maybe as low as 0.015 inch) while the pressure-based technologies can set the threshold at a level closer to 0.020 inch. Unlike the vacuum-based check where ambient air enters the system when the monitor is operating and a leak is present, a pressure check causes the air-vapor mixture in the evaporative control system to be expelled from the system when any size leak is present. In other words, performing the pressure check causes increased evaporative emissions to be released to the ambient if a leak is present. If there is a leak equivalent to or larger in magnitude than 0.020 inch present, then the leak is identified and fixed. However, if there is a leak smaller than 0.020 inch present, then the pressurized leak check would not identify the leak, but in the process of performing the check, excess evaporative emissions would be generated.
However, some manufacturers only use general fault codes to simply indicate the component or system that is malfunctioning. The repair technician is then required to use the manufacturer's service information or a fault tree to pinpoint the problem. In many cases, the diagnostic system actually detects different root causes (e.g., sensor shorted to ground or battery) for a malfunctioning component/system, yet the manufacturer only uses one fault code to identify all the different malfunctions.
Future: Many codes are designated with a P1xxx format as opposed to the generic, SAE-defined fault codes of P0xxx. Because these codes are manufacturer-defined, generic scan tools are unable to display the text label that corresponds to the fault code. Thus, technicians are only given the numeric fault code and a message such as “manufacturer-defined fault code.” CARB is seeking industry feedback to ensure that the majority of technicians have reasonable means available to identify the fault associated with a P1xxx fault code.
Similar to the practice described above, some manufacturers use limited fault codes to only indicate the malfunctioning component or system. More detailed fault information or analysis, known as symptom bytes, is then provided through a manufacturer-specific scan tool or “dealer” tool. In this situation, the dealer technician is provided with electronic information that is not available through the generic scan tool. If the information is available in the diagnostic system and useful to a dealer technician, the information must be provided to the generic scan tool.
Future: The diagnostic system may monitor the output voltage, response rate, primary oxygen sensor correction (if applicable), and other parameters that can affect emissions, of all primary (fuel control) oxygen (lambda) sensors for malfunction. It shall also monitor all secondary oxygen sensors (fuel trim control or use as a monitoring device) for proper output voltage and response rate. Response rate is the time required for the oxygen sensor to switch from lean to rich once it is exposed to a richer than stoichiometric exhausts gas or vice versa (measuring oxygen sensor switching frequency may not be an adequate indicator of oxygen sensor response rate, particularly at low speeds). This means that the vehicle will now be performing the “rate of change” test that we've been doing with our lab scopes.
The diagnostic system will store a fault code and the MIL shall illuminate no later than the end of the next driving cycle during which monitoring occurs provided the malfunction is again present.
As OBD II checks are added to I/M tests, there is concern that a significant portion of vehicles would be rejected for an I/M test because they do not have complete readiness indications for all major monitors. And what about consumer reaction should even small numbers of vehicles with incomplete readiness indications be rejected in an OBD II-based I/M program? Data from pilot I/M test programs show approximately 2 percent of test vehicles have incomplete readiness indications. An analysis of the data shows a large portion of the incomplete readiness indications are for the evaporative system monitor.
In Davis County, Utah, 65,927 vehicles were tested between January 1997 and July 1999. Of these, 91.33 percent tested “good,” 4.76 percent tested “not ready,” 2.26 percent had DTCs, .74 percent had their MIL on, and 1.48 percent were unable to communicate.
Doug Decker of the Colorado Department of Public Health and Environment referred to a DCPHE study to explain his views on the effectiveness of OBD II as a tool for identifying emission failures. The study states: “The results of this study indicate that OBD II is a very effective strategy for identifying vehicles with either high emissions or potentially high emissions. However, the data generated from this study indicate that if OBD II MILs alone were used to predict FTP pass/fail, the false identification rate would be 60 percent; i.e., 21 of the 35 study vehicles that failed OBD II (MIL illuminated) passed the FTP.”
A second way of looking at OBD II is to determine if the stored DTCs actually identify an abnormality. For regular-mileage vehicles, OBD II DTCs identified system or component problems that were in need of repair or that could have eventually resulted in high emissions 94 percent of the time (30 of 32 vehicles). For high-mileage vehicles the identification rate was 67 percent (six of nine vehicles). Using this criterion, OBD appears to be a reliable tool with which to identify high emitters or potentially high emitters. However, when OBD II DTCs indicated component or system problems, 18 of the 31 identified regular mileage vehicles and five of the seven identified high mileage vehicles passed the FTP. As a result, the DTC false identification rates with regard to federal In-Use FTP standards were 58 percent and 71 percent for regular and high mileage vehicles respectively.
Decker emphasized a problem scenario - a vehicle that failed OBD II for a lit MIL in the I/M station. “This vehicle is now in your shop for repair, and its emissions levels are cleaner than when it was new; and approximately 10 times cleaner than the I/M standard,” said Decker. “There is a problem with the car, but it may cost the motorist $400 or more for a negligible emissions reduction (and likely no change to driveability or fuel economy).”
There are a number of reasons why readiness indications may not be set at the time of inspection. If a vehicle with the MIL illuminated was repaired shortly before an I/M test and had fault codes cleared subsequent to the repair, the vehicle may not have been driven sufficiently to exercise all of the major monitors before being taken to the I/M station. In some cases, vehicle operation in extreme ambient conditions will prohibit the monitors from running and setting readiness indications.
“When a tech repairs an OBD II problem and clears codes, the readiness monitors need to be reset before the car can be retested. Who will do this? ... It may take a couple of days of 'normal' driving to reset monitors,” said Decker. “What about the procrastinator whose registration expired yesterday and is now driving around trying to reset monitors and gets a ticket? Will the motorist see the tech's unwillingness to reset monitors as poor service? Supposedly, some manufacturers have already established a key cycle test that flashes the MIL to indicate whether the monitors are set or not. Service advisors will need to have a script to work from - like an FAQ - when dealing with OBD II and MILs,” Decker added.
In summary, a statement from the CARB Web site appropriately addresses the status of OBD II:
“One of the primary goals of the OBD II program is, and always has been, to improve the availability of service information to the aftermarket repair industry. As such, the OBD II regulation contains several requirements for standardization of diagnostic connectors, communication protocols, fault codes, engine parameter data and test equipment. Additionally, staff has proposed new amendments improving the availability of diagnostic and repair information for all emission-related repairs. These requirements will allow independent repair shops to use a single diagnostic tool to access all of the information generated by the OBD II system for any manufacturer's vehicle.”
OBD IIIHave you heard anything recently about OBD III? Michael McCarthy of CARB shared some of the latest news about what's happening in this area.“We paid a contractor to build up a couple of mock cars with remote transmitting systems and demonstrate that it was technically feasible to have a system that sends out a remote transmittal when the check engine light comes on,” said McCarthy. “A pretty basic contract since there are already systems like On-Star and LoJack.” This prototype system was built by GM Hughes Electronics, and uses a roadside transmitter to interrogate vehicles as they pass by. The system is reportedly capable of retrieving information from eight lanes of bumper-to-bumper traffic whizzing by at speeds up to 100 mph! “The concept is to stop requiring smog checks for every passing car and only test the failing cars. It would likely be a voluntary system - when you buy the car you could choose whether to go to smog check every two years or pay $xxx and never have to go to smog check,” said McCarthy. “You would, however, have to push a button on the dash once every three months that would send a signal that identifies your vehicle and the status of the check engine light. If you forgot to push the button you would probably get a letter in the mail telling you to press it or bring it in for inspection. If you pressed the button while the MIL was on, you would probably get a letter in the mail saying you have 60 days(?) to correct the problem and press the button again. If the MIL came on and you got it fixed before you pressed the button, you would never get any notice in the mail. There would not be a continuous signal identifying the location of the car or anything like that.” |
AutoInc. Magazine ® Vol.XLVIII, October 2000
E-mail: asainfo@asashop.org, Web Site: http://www.asashop.org
Copyright © 2000 Automotive Service Association (ASA). All rights reserved.