VAPOR TRAILS: FINDING & FIXING EVAPORATIVE SYSTEM FAULTS BY BOB PATTENGALE Current OBD II systems can detect evap system leaks too small to see, or even smell. Finding and fixing these and other system faults requires a carefully considered approach. Evaporative emissions systems have been around for many years and for the most part have required virtually no maintenance or repairs. The only time we checked or repaired the older systems was when a customer smelled fuel vapors, when the purging vapors were causing an engine performance issue or when the tailpipe emissions exceeded recommended limits. The situation changed with the introduction of stricter emissions regulations, enhanced evaporative emissions monitoring systems and the illumination of the malfunction indicator light (MIL) when a fault existed. Here s a little history on the evolution of evaporative emissions systems: The Clean Air Act of 1970 required a 90% reduction in vehicle emissions by 1975 and established the U.S. Environmental Protection Agency (EPA) to manage and enforce the new regulations. In a very short period of time, the EPA recognized that a significant amount of hydrocarbon emissions were coming from the fuel vapors in the gas tank. Evaporative emissions standards were established in 1971 and the first charcoal canisters were introduced to trap gasoline vapors. The concept of trapping fuel vapors in a charcoal canister and purging them while the engine is running has basically stayed the same for more than 35 years. Starting in 1996, vehicle manufacturers were required to begin a phase-in of the EPA enhanced evaporative emissions standards. The new standards required the on-board diagnostic (OBD) system to check for evaporative emissions leaks greater Photos: Bob Pattengale than.040 in. The phase-in of the new standards began with 20% of the vehicles in 1996; 100% had to be in compliance by 1999. California took this a step further and required 20% of the vehicles to be capable of checking for leaks greater than.020 in., beginning in 2000. The final 100% phase-in was set for 2002. What does this mean for vehicle manufacturers and technicians? It means that in order to accurately 28 February 2006
catch all leaks greater than.020 in., the on-board diagnostic system needs to check for leaks smaller than.020 in. The threshold to set a diagnostic trouble code (DTC) might be as small as.010 in. Finding leaks smaller than.020 in. can be a real challenge. If you re using a smoke machine to locate the source of a leak, very little smoke will be visible if the leak is under.020 in. We ll discuss options to locate small leaks later in this article. The first rule in diagnosing any evaporative emissions problem is: Don t touch anything! Do not fall into the trap of tightening the gas cap and then beginning your testing. What if the gas cap was the problem and you run an in-bay evaporative system leak test or perform a leak test with a smoke machine and no leaks are found? At this point, you won t know if the leak is intermittent or if it was actually the cap. In the early stages of your diagnosis, perform as many nonintrusive tests as possible. After retrieving and recording any DTCs, look up the exact description and recommendations for your specific code. Don t forget to check and record the freeze frame data. The key freeze frame data parameters (PIDs) to check are engine coolant temperature (ECT) and vehicle speed sensor (VSS). The ECT will let you know if the engine was cold or warm when February 2006 29
Inches of Water Gauge Smoke Machine Connection Point Vacuum Test Tee 1. This photo of a DaimlerChrysler StarSCAN screen was taken during the Purge Vapors System Test. The unit is displaying data related to the performance of the purge solenoid, O 2 sensors and fuel trim. Monitor the scan tool data while performing a wiggle test on the related circuits and components to check for intermittent issues. the DTC was set. The VSS might help you decide if vehicle vibration might be a contributing factor in the case of an intermittent leak. In some cases you might have two trouble codes present: a P0xxx generic DTC and a P1xxx manufacturerspecific code. For example, a vehicle with a generic DTC P0455 (Evaporative Emissions Control System Leak Detected [Gross Leak]) might also indicate a P1456 (Evaporative Control System Leak Detected [Fuel Tank]). In this example, the factoryspecific trouble code points you to the fuel tank area for the source of the leak. In this case it might be a loose or faulty gas cap. The next step in the diagnostic process is research specifically, checking for technical service bulletins (TSBs), recalls or powertrain control module (PCM) reflashes. If you want to save time and money, spend a few minutes researching the problem; you might find a solution prior to testing. For example, if you were working on a 2002-04 Honda CR-V with a P1456 (Evaporative Control System Leak Detected [Fuel Tank]) and you checked for TSBs, you d find TSB 04-002, which describes a condition where the threads in the fuel fill pipe bind, not allowing the gas cap to fully seat, thus creating a leak condition. If this TSB applied to the vehicle you were working on, following the recommendations might point you right to the source of the problem. In any case, you still need to verify the source of the leak. It would be a good idea to connect a smoke machine and see if smoke is coming from around the gas cap. Another benefit of accessing the TSB is that it lets you 3. An evap-approved smoke machine like Vacutec s EVAPro can also be used to check for restrictions in the evap system. Watch the position of the flow check ball (arrow) to determine if a problem exists. 2. This particular vehicle did not have an evap system test port, so a vacuum tee was installed for testing purposes. The photo illustrates alternative methods of testing for purge restrictions and system leaks when no test port is provided. know the fuel fill pipe needs to be replaced, not just the gas cap. I m guessing that without the TSB, most of us would replace the cap, which would be a temporary fix. Checking for PCM software calibration updates or reflashes can correct issues that cannot be fixed any other way. For example, some GM vehicles need a calibration update to correct a DTC P1441 (Evaporative System Open Purge Flow) from setting on a hot restart. I know it s been mentioned many times before, but don t overlook checking www.iatn.net for good suggestions. Other techs may have already reached successful diagnostic conclusions under similar or identical conditions. Evaporative emissions problems fall into three basic categories: component and circuit DTCs, purge flow DTCs and leak detection DTCs. The type of code present will determine the diagnostic path. Component and circuit DTCs normally point to opens, shorts and component failures. For example, a DTC P0443 (EVAP Purge Solenoid Circuit) can occur if an open or shorted condition is detected in the purge solenoid circuit. Your diagnostic method will depend on the tools available. The purge 30 February 2006
4. This disassembled purge solenoid was taken from a Nissan vehicle with a clogged evap system. As you can see in the smaller inset photos, the charcoal pellets were able to travel the full length of the evap system. control circuit can be checked with a basic voltmeter or scan tool. The scan tool would be my first choice, if bidirectional controls are available for this component or circuit. For example, to quickly check the function and operation of the purge solenoid, use the scan tool to activate the solenoid and listen for a clicking sound. This checks the PCM, wiring and component all at one time. If the solenoid clicks, the circuit is currently working, and you might be dealing with an intermittent problem. At this point, perform a wiggle test on the related wiring and connection points and repeat the test. If the solenoid did not click, the problem may be with the wiring, the purge solenoid or even the PCM. The next step is to use a voltmeter to check power and ground wires and to repair as needed. If the circuit checks good, then the problem is most likely with the purge solenoid, which can be checked with an ohmmeter or manually tested with a set of jumper wires. If the scan tool displays specific PID data related to this circuit or component, you can use the PID data for a quick reference. Photo 1 on page 30 shows a screen capture taken from the Chrysler StarSCAN showing Actual Purge Current in milliamps (ma). If the circuit were open or shorted, the current values would exceed recommended limits. You can also watch the scan tool data while performing a wiggle test on the related circuits and components to check for intermittent issues. Purge and vent solenoids are generally easy to diagnose. If you re dealing with pressure sensors or leak detection pump (LDP) circuits, it will be important to refer to specific manufacturer diagnostic recommendations. The next area of the evap system is purge flow diagnostics. Possible DTCs for this problem include P0441 (Insufficient or Excessive Flow Detected During EVAP Operation), P1447 (Evaporative Control Purge Flow Monitoring) and other codes. Honda vehicles have a manufacturerspecific DTC P1457 (Evaporative Control System Leak Detected - EVAP Canister System) that could create some confusion. The code description might have you looking for a leak in the system when the problem is actually a restricted or clogged evap system. There are a variety of methods employed by vehicle manufacturers to detect purge flow. A widely used option is to monitor oxygen sensor or fuel trim values when the purge solenoid is activated. The PCM expects to see a change in either of these values when the purge solenoid is opened. If you plan on using this method to verify purge operation, keep in mind that when the purge solenoid is commanded on, the values 32 February 2006
5. The vent solenoid must have a fresh air inlet. In this case, it s via a hose attached to a hollow frame rail. To quickly check for leaks without electronically closing the vent solenoid, disconnect the hose, as shown in the inset photo, then plug it. may go rich or lean. In most cases, the O 2 sensor will read rich and short-term fuel trim (STFT) will go negative, due to the buildup of fuel vapors in the evap system. However, there are times when the content of the evap system is mostly air. In this case, the O 2 sensor will read lean and STFT will go positive. Another option available to vehicle manufacturers is to monitor changes in the fuel tank pressure (FTP) sensor when the purge solenoid is activated. In most cases, the PCM will command the vent solenoid closed and the purge solenoid open, purging the vapors in the evap system and drawing the system into a slight vacuum condition. The PCM expects the FTP sensor reading to decrease. Once again, your diagnostic approach will depend on the equipment available. If you have a scan tool with an in-bay evap test option, this test can verify that a purge problem currently exists. If the scan tool confirms a purge restriction, you ll still need to locate the source of the restriction. My first choice would be an evaporative emissions system-approved smoke machine with a flow gauge. This test will work on most evap systems with a vent solenoid that s normally open. Connect the smoke machine to the evap test port. Removal of the Schrader valve is recommended. You might find that the evap test port is missing. Some manufacturers have chosen to remove the evap test port to save money. Photo 2 on page 30 shows the installation of a vacuum tee and inches of water gauge for testing purposes. After the test gauge is removed, a vacuum cap can be installed on the tee for future testing purposes. If you re using one of the newer smoke machines, select the TEST option and insert the probe into the test port. Turn the smoke machine on and watch the flow meter check ball (photo 3). If there are no restrictions in the evap system, the check ball should be close to the top. If you want to be more precise, connect an inches of water gauge, as shown in photo 2. If the gauge reads above 3 in. H 2 O, check for restrictions. This test can also be performed using early versions of the smoke machine that have a flow meter. Once again, the flow meter can be used to verify flow. If the check ball is close to the bottom, look for restrictions. Restrictions in the evap system can be caused by pinched or kinked hoses, dirt, mud or material from the charcoal canister blocking purge lines, or other situations. In some Nissan vehicles, the charcoal canister may leak charcoal into the evap system. Nissan TSB NTB00-085 covers this issue. Photo 4 shows a disassembled Nissan purge solenoid. The lower left inset photo is a close-up of the purge solenoid plunger showing a piece of charcoal stuck to it. When I first removed the plunger, several pieces of charcoal were stuck to its surface. This would have prevented the purge solenoid from sealing. The middle inset photo shows charcoal pieces in the purge solenoid vacuum housing. The lower right inset photo shows the plunger and spring assembly. The last diagnostic topic we ll discuss is looking for leaks in the evap system. The first step is to verify that a leak currently exists in the evap system. Most leaks are pretty easy to find, but some intermittent leaks are difficult to locate. Intermittent leaks can vary with changes in temperature, vehicle vibration and sticking vent or purge solenoids. The initial leak check can be performed with a scan tool using an inbay evap test or a smoke machine. I prefer to get the smoke machine connected to verify the leak; that way, if a leak does exist, I can begin looking for it right away. Connect the smoke machine to the test port, then close the vent solenoid. You can do this with a scan tool s bidirectional control, manually connecting jumper wires to energize the solenoid or by plugging the vent hose air inlet. If you choose to manually energize the vent solenoid with jumper wires, keep in mind that most solenoids are duty cycle-controlled, which means the solenoid does not remain on for long periods of time. To prevent damage to the solenoid, 34 February 2006
6. This original Vacutec smoke machine is equipped with a flow gauge. An inches of water gauge was attached to the unit to measure its maximum pressure. The gauge reads 14 in. H 2O, which is too low to damage an evap system. 7. An exhaust gas analyzer, such as this PC-based unit from Automotive Test Solutions, can also be used to look for evap leaks. The software records gas data, which can be reviewed later if you re unable to see the gas analyzer display while sniffing with the analyzer probe. disconnect the vent solenoid after five minutes of ON time. On some vehicles, the vent solenoid inlet hose is connected to the frame rail to reduce the introduction of moisture into the evap system (see photo 5). You can plug the hose to quickly seal the system without touching the vent solenoid. If you have one of the newer smoke machines, calibrate the machine to verify the size of the leak. The older flow gauge-equipped Vacutec smoke machines can also be used to verify and look for leaks. Photo 6 shows the flow gauge and an inches of water gauge. When smoke is applied, the Vacutec smoke machine applies a maximum pressure of 14 in. H 2 O, which is safe for evap systems. Begin applying air pressure or smoke to the evap system and watch the check ball in the flow gauge. The ball will start high, then go to the bottom if no leaks are present. If the ball does not go to the bottom or exceeds the calibrated leak setting, begin looking for leaks in the system. If you re using a newer evap smoke machine, change from TEST to SMOKE. A bright light can be used to find smoke on smaller leaks. If you re unable to locate the leak with smoke, there are a few more options that might help. Photo 7 shows an ATS gas analyzer and software being used to find leaks. Move the gas analyzer probe slowly around the evap system. Hydrocarbons (HCs) are lighter than air, so be sure to check the areas above the evap system fittings and hoses to find a leak. The ATS software is helpful when looking for leaks in areas where you re unable to see the gas analyzer display while probing. The software records the HC level, so once you ve completed your search, check the recording for spikes in the HC level. If you see a high spike, refine your search or ask someone to watch the display while you recheck for leaks. Pressurizing the evap system using the TEST option on the newer evapapproved smoke machines also can help increase the output of gases at the location of the leak. If you re still having trouble locating the leak, another option is to break the evap system into smaller test sections. This should be considered an option of last resort, because you may change something in the process by disturbing the evap system, and the leak may temporarily disappear. Follow the purge hose from the front of the vehicle to the back, looking for points where you can disconnect and plug a section of the evap system. Retest for leaks. If no leaks are noted, the leak must be in another section of the system. Keep going section by section until the leak is located. Intermittent leaks are the most difficult to locate. If you can t duplicate the leak using conventional methods, try testing the evap system under various temperature conditions (cold and hot). Check the purge and vent solenoids. They may be intermittently sticking or stuck open with dirt or foreign matter. Perform a wiggle test on the evap system connection points. Keep in mind that most leak-detection methods use pressure, but most evap systems are tested with vacuum under OBD II. In some rare cases, an evap system may leak under vacuum but not under pressure. There are options to test the evap system with vacuum, but that s a subject for another time. Evap system problems are not difficult to diagnose if you follow a logical process. Take your time and make sure you spend a few minutes researching and understanding your specific evap system before you begin. Visit www.motor.com to download a free copy of this article. 36 February 2006