This article is one in a series that have been released in conjunction with Wayne's new book, 101 Performance Projects for Your Porsche Boxster. The book contains 312 pages of full color projects detailing everything from performance mods to changing your brake pads. With more than 950+ full-color glossy photos accompanying extensive step-by-step procedures, this book is required reading in any Boxster owner's collection. The book is currently available and in stock now. See The Official Book Website for more details.
Today's modern cars have a tremendous amount of vacuum hoses and boots contained with the engine compartment. To the uneducated eye, the engine compartment can easily look like the insides of an oil refinery with all the hoses running in and out. As these hoses and boots age and are constantly exposed to hot and cold temperatures, they tend to break down and develop cracks which can then cause vacuum leaks. Unfortunately, when the fuel injection system develops a vacuum leak it will tend to confuse the fuel injection computer's sensors, and the car will cease to run properly. You may get decreased gas mileage, rough idling, misfires, and sometimes a check engine light (CEL) on the dashboard.
Whenever someone contacts me and indicates that their car is running rough, I almost always tell them to check the entire system carefully for vacuum leaks. Without the proper tools, this can sometimes be very difficult. Old rubber boots have a tendency to crack and leak in spots that are not visible to the naked eye. Sometimes squeezing them will show a crack that you can't normally see when the boot is in its initial resting position.
One poor man's way to check for vacuum leaks is to artificially create a leak and see how the engine reacts. There are two primary vacuum systems on the Boxster engine: crankcase vacuum and intake manifold vacuum. You can test for proper crankcase vacuum by removing the oil cap while the car is idling. The engine should change idle and begin to run a bit rougher with the oil cap off. If there is no change in the running of the engine, then you might have a crankcase vacuum leak somewhere (sometimes caused by a failing air-oil separator: see Pelican Technical Article: Air / Oil Separator). To induce a vacuum leak into the intake manifold, you can disconnect one or more of the hoses that connect to the intake. One example would be the hose that connects to the air-oil separator. If you crack one of these hoses open just a bit, and the engine RPM doesn't change, then you might have an intake vacuum leak somewhere.
By far, the best way to test for vacuum leaks is with a smoke machine. Although these are somewhat expensive at about $650, you can rent and/or borrow them from some shops. The machine generates smoke and then blows it through your engine's intake and crankcase. All you need to do is sit back and watch for little puffs of smoke where there is a leak in the system. This can save many, many hours of random troubleshooting and guesswork. For the project car for this book, I wanted to make sure all of my custom-made hoses that were required for the installation of the 3.4 engine into the Boxster were leak-free. I ran the car through the smoke machine and confirmed that everything was airtight and there were no troublesome leaks.
The smoke machine runs on standard household baby oil, and generates smoke by heating the oil. The smoke generated is very similar to the type created in model railroad steam engines. The machine also needs to be connected to a shop air compressor. The compressed air is mixed with the smoke and then funneled through the intake system of the engine. In the case of the Boxster, I removed the intake air filter and used a cone adapter that comes with the machine to seal the smoke hose to the intake pipe. The car should be completely cold when you are testing it for leaks, as the rubber hoses and boots are most likely to leak when they are cold and contracted. You should also remove the mass airflow sensor from the system so that the smoke doesn't build up on the sensor and affect it's operation. Plug the hold for the sensor with some masking tape.
After you power up the smoke machine, it will take a minute or two to fully smoke out the car. If your car is completely air tight (a good thing), then you might not see any smoke. You can check to see if the smoke machine is operating correctly by removing the oil cap in the rear trunk. You should see a steady plume of smoke exiting out of the oil filler (this is normal). On the Boxster, you may also see some very tiny plumes of smoke exiting out of the resonance flapper bearings: this is also normal and doesn't affect the operation of the car or indicate a major vacuum leak. If you do see steady plumes of smoke exiting out of the engine compartment, then investigate further. On the 3.4 conversion engine, there weren't any major leaks, but there were a few minor ones that were found (one of the breather hoses that connected to the radiator tank needed to be tightened).
Although you might think a smoke machine is a limited use tool, it's highly versatile for solving other problems as well. Basically any system that contains air can be tested. You can use the system to check climate control systems, leaky headlamp housings, exhaust systems, A/C lines and compressors (although most A/C leaks are very small and difficult to detect with just a smoke machine). You can also bench test components like radiators prior to installation to make sure that they are factory perfect.
You can even use the smoke machine to detect leaks from door and window seals. First, rollup all windows and seal the car. Then turn on the fresh air fan motor to the maximum setting (do not set the system to recirculate). Using the smoke machine with a diffuser (a wider nozzle that will slow down the flow of smoke), move around the outside of the car and blow the smoke onto the area you think might have a leak. The fresh air fans inside the car will create a positive pressure environment that will push air out through any leaks. By slowly blowing smoke on these suspected areas, you can see the smoke pattern become disturbed by the leaking air. This test of course requires that you do it inside your garage in an environment where there is very still air.
Shown here is the Redline Smoke Pro machine. This extremely useful tool is invaluable for finding vacuum leaks within your fuel injection system. The machine's air supply is provided by an air compressor (upper left) that is plugged into a pressure regulator (green arrow). The heater is powered by your car battery (upper right and blue arrow). The compressed air is combined with smoke and then pushed out through the nozzle (yellow arrow).
With the smoke machine turned on, you can see the trail of smoke that exits out of the nozzle. You don't want to "smoke out" your mass air-flow sensor (MAF), so be sure that you remove it and tape off or plug the hole prior to pressurizing the system (upper left). The Smoke Pro comes with a whole set of adapters that you can use to plug into the intake system. On the Boxster here, we removed the aftermarket cone filter and plugged the intake with the rubber cone adapter that comes with the Smoke Pro (blue arrow).
With the system pressurized, you can check to see if your engine is "fully smoked out" by removing the oil filler cap. A steady stream of smoke should exit the filler hole. This means that smoke is going from the intake, through the air-oil separator and into the crankcase. At this point, replace the cap and carefully examine your intake for smoke trails that will indicate vacuum leaks.