Managing the EVAP maze

Evaporative emissions (EVAP) systems have been around since the early 1970s. When an emission system has that much tenure on a vehicle, you would think diagnosing problems would be as easy. Creating a great degree of frustration is the over proliferation of unique systems that work in various manners requiring a wide variety of diagnostic approaches. The names and types of systems boggle the mind: Intrusive, Non-Intrusive, ENOV, NVLD, DM-TL, LDP, EISM and on and on.

It all adds up to a frustrating path we must maneuver in the service bay that reminds me of one of those hedge row or corn mazes that cause even the smartest explorers to get lost in their complex choice of paths to follow. It doesn’t have to be a maze of confusion. EVAP Diagnostic Trouble Codes (DTCs) fall into the top 10 items to turn on your customer’s MIL so being competent and confident in your diagnostics of EVAP systems can put money in your pocket.

Your Approach to Keeping It Simple (KIS)
There almost seems to be as many different types of EVAP systems as there are vehicle models on the road today. Let’s take a look at some ways to simplify the complex world of evaporative emission systems so you can find your way to the finish point of a properly repaired vehicle. It really all boils down to three things:

1. Knowing the System

2. Knowing the EVAP DTCs

3. Knowing your EVAP diagnostic equipment.

Knowing the system you are working on always is first and foremost to success in diagnosing anything on a vehicle. Pre-OD II days saw a plastic canister of activated charcoal in place to perform the process of the adsorption of fuel vapors in the fuel tank. Adsorption of vapors is a process somewhat comparable to the more common term “absorption” of liquids. It is important to remember that the vapor adsorption process is limited (an EVAP canister can only hold so much) and creates heat.

If you ever want to check this out, fill a gas tank and point an infrared pyrometer at an EVAP canister sometime. You’ll be amazed at the temperature increase observed – about 150° F on the outside surface. Because heating and cooling cause expansion and contraction, the canister is equipped with a screen wire like grid that, with the help of a spring, keeps the activated charcoal packed tightly in place throughout the various thermal cycles.

Because we know that a canister is not simply a plastic box filled with charcoal, it’s easier to appreciate how chuck holes and off road driving can take a toll on canisters. Internal fractions can allow the charcoal to escape into the purge line. If the spring gives up the ghost or breaks, the charcoal particles will have too much space between them and no longer be effective as well as move around and become pulverized. We’ve all seen those little black “coffee ground” looking particles plugging up purge valves.

Inside that more complex than you might have imagined canister also might be an enormous amount of dirt and sand if the fresh air filter has been neglected. This will cause the canister to be restricted for proper purge flow. One Asian OEM had a Technical Service Bulletin (TSB) that instructed technicians to remove the canister and weight it on a scale as part of a diagnostic process. A canister too heavy (from dirt) meant it’s time to replace. This condition can cause a DTC P0446 for excessive tank vacuum/restricted canister vent.

Because the canister can only hold so many vapors the Engine Control Module’s (ECM) job is to control the EVAP purge solenoid. Part of the “maze” of EVAP diagnostics is the ever-changing terms that apply to various components which essentially do the same job. While GM calls the solenoid that controls purge operation a canister purge valve, Toyota might call it a VSV (Vacuum Switching Valve) and Ford may call it a VMV (Vapor Management Valve). Thinking generically, just call it the “Front Door,” because it is the front door to allowing vacuum from the engine into the evaporative system in order to pull vapors from the canister.

If there is a front door to allow vacuum into the canister there must be a “back door: to allow fresh air to move through the canister in order for the canister to release the vapors it stores. So along with the obvious things like the fuel tank, filler neck and gas cap we see that every EVAP system has all of these items in common in one way or another. Prior to OBDII, you might remember vehicles that had canisters that were vented open all the time.

From the beginning of computer controlled engine management days in the 1980’s, the ECM would apply a varied Pulse Width Modulated (PWM) duty cycle to the front door and upon assumption that there were vapors in the canister to be burned in the engine, backed off the base pulse width in order to compensate for those burnable vapors and maintain a stoichiometric mixture. Early OBDII systems called “non-enhanced” went beyond this compensation to test for actual operation, with a purge vacuum switch and/or an ECM algorithm that allowed for O₂ / short term fuel trim to determine if the all mentioned components were doing their jobs.

Called a Loaded Canister Test, the typical ECM approach is to gradually ramp up the purge valve duty cycle while watching Short Term Fuel Trim (STFT) for excessive shift due to the hydrocarbons (HC) being pulled from the canister. If the ramp up of purge is too fast (ECM not seeing much difference in trim numbers) the conclusion is an insufficiently loaded canister. Today’s ECMs still might do this purge test although it’s unclear exactly which manufacturers are still running this test (if any) with enhanced systems.

On some vehicles today, if the canister is no longer effective, a DTC may or may not set. Some of the early OBDII non-enhanced systems such as those found on GM used a purge switch to give the ECM feedback that purge vacuum was being applied to the line running to the canister. So there was only the potential for flow codes as these systems had no ability to detect leaks. The ECM also has, of course, the ability to detect any opens/shorts and over current conditions in most EVAP related outputs such as purge and vent solenoids just as it has on other electrical outputs. Keeping these facts in our minds helps us to remember that EVAP DTCs are not exclusive to leakage problems.

How And When
Systems with enhanced diagnostic capabilities for reporting leaks can be broke down into two major categories of how leak testing is done and when leak testing is done. How EVAP system on board leak testing works can be broke down into two categories: pressure and vacuum. Pressure systems are mainly the Chrysler LDP (Leak Detection Pressure) and DMTL (Diagnosis Module Tank Leakage) systems used in BMW, Mazda, Land Rover and Kia. Vacuum systems are most everything else.

Vacuum systems have been around since the beginning of enhanced EVAP systems and don’t have any extra parts like pumps and the like. Just a front door (normally closed purge solenoid) and a back door (normally opened vent solenoid) along with a Fuel Tank Pressure (FTP) sensor is all that is required for leak detection operation. The FTP sensor on GM is biased at ambient pressure (sea level) at around 1.5 volts with positive pressure pulling the sensor lower and negative pressure (vacuum) pulling the voltage above the 1.5-volt mark.

Ford bias its FTP just the opposite with pressure causing a move up the voltage scale from their 2.5 ambient pressure starting point and voltage running lower than 2.5 volts when vacuum is sensed in the system. Typical leak detection monitors require quite a bit of enabling criteria in hopes of avoiding false codes. And being a two trip code, it’s no wonder that EVAP monitors are the hardest monitors to run.

The leak test portion of the monitor requires an initial look at the voltage when the front door (purge) is opened along with the back door (vent) being open to determine flow problems. A slight vacuum is expected to occur. When vacuum is introduced into a system with normal restrictions such as the canister and vent filer, there should be a lowering of pressure. Too much means too large of a restriction through the canister and vent filter or something plugging the purge line or even a lack of purge valve operation. This is followed by a gradually opened front door (purge) and closed back door (vent) to determine if the system can indeed create a vacuum or drop in pressure within the system. If we can’t pull the system down in the purge/system seal mode, then there must be a gross leak – just like when you can’t pull a vacuum in an air conditioning system with your vacuum pump.

Finally the front door (purge solenoid) is closed to see if the vacuum we created with a sealed system during purge will hold for a bit. If there is a decay of vacuum then there is a leak, which again affords itself to the air conditioning vacuum leak test when the pump is turned off on your RRR machine. Simple enough?

Going forward to more modern 0.020-inch leak standard, those tests can be misleading (spell that false codes) due to fuel sloshing and fuel temperature changes. This ushered in a host of key off systems during the mid-2000s. Toyota turns on a vacuum pump about five hours after the engine is turned off, first pulling the vacuum through a .020-inch orifice inside the vacuum pump and then pulling the vacuum against the closed system.

The whole idea with Toyota is to get the fuel cooled off and thereby stable for pressure testing. If the outside air temp is over 95 degrees F. the wait time could be extended to seven to nine hours. GM’s EONV system works with the key off like Toyota, but that’s where the similarities end. GM uses an immediate test to capture the pressure increase that occurs with hot fuel in the fuel tank from the previous trip. As the fuel cools, the tank pressure drops thereby creating a natural vacuum that can be watched for a decay which indicates a small leak. There are numerous enabling conditions unique to this system which can take up to four days to complete the number of successful tests needed to complete the monitor. E85 and high RVP fuel can make a radical initial pressure spike prior to the cool down pressure drop/natural vacuum and inhibit the monitor from running.

DTML Systems
DMTL pumps have a built in DC motor that is ECM triggered. The ECM monitors the pump’s current draw as a way of detecting leaks. The DTML assembly also contains an ECM controlled change over valve that is normally open and closed by the PCM during a leak test. It essentially is the back door of these systems.

The leak test process begins with the pump applying its output of air pressure against a 0.20-inch or 0.40-inch orifice to calibrate what an acceptable leak looks like with pump current. The change over solenoid valve (back door) is then energized which seals the EVAP system and directs the pump output to pressurize the entire EVAP system. If there is a leak larger than the regulatory limit it will take less current for the pump to run. Smaller leaks or no leaks mean higher current levels and no evap leak DTCs. Because small amounts of current are involved, good connections would seem to go a LONG way in ensuring accuracy and guarding against false codes.

Chrysler LDP Systems
This system has been around for many years but seems to be the most misunderstood. The Chrysler LDP pump uses a 12-volt solenoid that when pulsed by the PCM causes engine vacuum to be applied to a diaphragm inside the pump. As the vacuum is switched on and of the diaphragm moves up and down creating a small amount of pressure against a spring that is calibrated to 7.5 in/H₂O. That’s not much pressure – a tad more than 0.27 PSI. It will expand a balloon a bit but not inflate it. The LDP pump itself is the gateway to the back door of the system consisting of a remote canister vent air filter.

Here is how the system works: first the purge is ramped on to clear out any pressure and then turns off. Next, the LDP solenoid is energized allowing vacuum to be applied to the top of the diaphragm, compressing a spring, pulling the diaphragm up. The upward movement then creates a slight pressure drop under the diaphragm. The inlet check valve opens, allowing in filtered fresh air.

Next the reed switch opens to tell the ECM the diaphragm is pulled up. The solenoid is then switched off. With pressure equal on both sides of the LDP diaphragm, the spring forces the diaphragm down physically closing the fresh air inlet valve and opening the outlet valve that allows the LDP’s pressure into the tank / canister. The magnetic reed switch state changes and the solenoid is now re-energized. The diaphragm may only move ¼ inch and then stall out while the solenoid still runs. This is normal.

The ECM looks at the fuel level and figures out how long to run the LDP. The fuller the tank the less time the pump has to run. Once the EVAP system reaches the spring pressure, (7.5 inches of H₂O) the diaphragm will simply stay in the upward position. A timer in the ECM based on fuel tank level starts and switch status is monitored. The longer the diaphragm stays up, the better that evap system can seal. Leaks will cause the diaphragm to fall and the switch sense will change back to 12 volts again. The time between LDP solenoid turn off and LDP switch closure is called the pump period. This time period is inversely proportional to the size of the leak. The shorter the pump period, the larger the leak. The longer the pump period, the smaller the leak.

Keep in mind it’s all about the tank’s fuel level though. Those factory specs greatly vary with the fuel level. If the fuel level sending unit is inaccurate (oops, I think I bent the float arm) you can actually have a leak DTC set. This can happen on most any enhanced EVAP system!

Testing LDPs
When testing the system for leaks, you must either activate the LDP with vacuum and an electrical command or pull the hose from the remote vent filter and plug the fitting at the LDP prior to using a smoke machine to create a system seal and look for smoke / leaks. A word of caution should be applied here. Failure to connect your test leads with power and ground to the correct terminals can ruin the LDP’s reed switch thereby ruining your diagnosis.

You can test your LDP with a method as simple as moving the diaphragm up with a constant ground applied to the solenoid (with a vacuum source applied to the LDP’s manifold vacuum port) to see if you can change the reed switch status (from 12 volts to 0) or you can force the pump to run using the bi-directional controls of a capable scan tool and monitor the LDP solenoid and reed switch sense lines with a dual trace scope. The choice is yours. Just don’t smoke it and call it a day. There is much more to this system than just leaks such as the ability to look for a pinched line for proper system flow during LDP operation.

Knowing Your DTCs There are a wide variety of DTCs associated with EVAP systems and even when the second digit of the DTC is a zero, meaning non-OEM specific that doesn’t mean there won’t be a slight variation as to what can cause the DTC. Some DTCs are very specific – P0456 might very well be specific meaning a small leak but what about P0440 for an EVAP malfunction? Numerous OBDII tests may run in order to set a particular DTC. These tests are often displayed in Mode $06 data and described in more detail in the manufacturer’s Mode $06 explanations.

This partial list of OBDII testing strategies for EVAP leaks and flow operation are often worded very close to this sample summary in the description section at the beginning of the appropriate DTC trouble tree.

Power-up Vacuum Test: A passive test designed to detect restrictions or blockages in the vent path. When the ignition switch is turned on, the vent solenoid is opened and the purge solenoid closed. The test passes when fuel tank pressure sensor does not indicate a pressure or vacuum with the vent solenoid open.

Small Leak Test: An active test designed to detect minor leaks. The vent and purge solenoids are completely closed while a vacuum is still present in the fuel tank. The system monitors the fuel tank pressure sensor for a vacuum decay rate that is too fast. Remember; the head space (area above the fuel) in the tank dictates who quickly vacuum (or pressure on some systems) will leak off. The more the head space (less fuel) the longer the tank should take to bleed off. As stated before, tank level accuracy is super important.

Purge Solenoid Leak Test: An active test designed to detect a vacuum leak through the purge solenoid. The vent and purge solenoids are commanded closed. If the pressure sensor indicates an increasing vacuum, the purge solenoid is leaking.

Knowing Your Equipment Just as important as knowing the system and the code is knowing your equipment. Beginning with the smoke machine it’s important to remember to not tighten the gas cap prior to testing. You just may have ruined your evidence! If there isn’t an EVAP test port, improvise an entrance point for your smoke machine. Patience is required. Smoking from the front first can find leaks in the purge line, canister and vent solenoid but you’ll need to wait for the smoke to fill the tank before there is enough to come out the cap/filler neck.

Is your machine equipped with a flow meter only or pressure/vacuum gauge along with a flow meter? A flow meter is important to help you see how big the leak is. On full function machines with the flow meter and built in calibration settings for .020 and .040 leak sizes you will have the ability to induce pressure only into the system while watching the ball in the flow meter. If the ball is floating above the specified leak setting, it’s time to add smoke. If the ball is below that setting, don’t waste your smoke. The system isn’t leaking right now.

Try operating the back door (vent solenoid) using an ammeter (if possible) for 10, 20 even 30 times to attempt to catch the all too common sticky vent solenoids. On machines with pressure/vacuum gages the gauge not only shows you how much smoke/pressure you are putting into the system (back door/vent closed of course) but also allows you to look at the vacuum from purge when the engine is running.

Partnering your smoke machine up with your scan tool makes all the sense in the world, especially if the smoke machine has a pressure/vacuum gage. Commanding purge and while comparing the gauge to the FTP data PID on the scan tool can help confirm accuracy of the FTP sensor. The scanner also might be able to affect a system seal where the purge is turned on, vent closed to pull the EVAP system into a vacuum and then the purge turned off to watch for vacuum decay. If the scan tool PID for the FTP sensor is not available, you can use this gauge to watch for EVAP purge pressure drop and if you are doing a system seal you can watch the machine’s gauge to observe vacuum decay on the smoke machine.

All but one brand of smoke machine incorporates a leak detection dye in the special oil used to make smoke so don’t pass up the opportunity to use an UV light and yellow glasses (that came with the machine) to help locate hard to find leaks.

Make sure you utilize the flow control knob on machines so equipped to customize the smoke volume to the leak. Sometimes the leak is too small or too large to find with the full dose of smoke / pressure. Kinking the hose a little can work on machines that don’t have the flow control feature. It’s all about velocity. Slowing down the velocity of the smoke can GREATLY help your chances of seeing smoke come out smaller holes in the system.

Finally, realize that although smoke will find the majority of leaks, very small leaks may not allow smoke molecules to escape. Bernie Thompson, veteran mobile tech and equipment inventor at Automotive Test Solutions, came up with a solution for such small leaks after many years of battling the most stubborn small leaks. His Bullseye Leak Detector uses CO2 cartridges and an electronic CO2 detector to generalize the location of a leak along with color changing foam to pinpoint the leak. CO2 molecules can get through tiny leaks in the 0.20-inch and under range where smoke has a hard time getting out.

While your trusty smoke machine is still an MVP player in EVAP diagnostics, there is no one single tool that fixes cars for us. Navigating EVAP diagnostics requires breaking down complexities into simplicities that we can grasp. Combining that along with knowing how to use the tools you have available to the fullest extent of their capabilities will be a great help in keeping us from getting lost in the EVAP maze.