The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Fuel Injector Clinic
Please Support STM Tuned

Compression and Leak-Down Testing

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

I see countless threads in the forums where someone has recommended a compression test for various problems. A leak-down test is mentioned much less often, but in many cases it would actually be more beneficial. Although they may seem similar the two tests are actually quite different, and used together they can tell you a lot about what is happening inside your motor. In this article I'll try to explain the difference in the two, explain how to do them and interpret the results, and maybe shed some light on a few other related topics as well.

When you look past the shiny IC pipes, front mounts, BOV's, and all the other bling surrounding our little engines, they are really nothing more than big glorified air pumps with a single purpose in life:

1. suck in as much air as possible
2. mix fuel with it
3. compress it
4. ignite it to create power
5. get rid of what's left over through the exhaust

Most people are familiar with this concept and understand how to test for intake boost leaks, see whether they are getting fuel or not, and check to see if there is spark. But when it comes to knowing what is going on in the middle of the process when the air and fuel are being compressed and burned, things seem to get a bit confusing. To help clarify this, I've thrown together this little refresher on how the 4-stroke combustion process works.

So what's the difference between a compression test and a leak-down test?

In very basic terms, a compression test tells you about your engine's ability to generate cylinder pressure, and a leak down test tells you about it's ability to hold pressure. Another way to think of it is that with a compression test you are seeing how much pressure the engine creates, and with a leak-down test you are seeing how much pressure it loses.

There are a few things worth noting here about an engine's ability to create and hold pressure. For one (although it doesn't really affect the test results), an engine will develop higher cylinder pressure when it's running rather than when you are just cranking it over for the test. This is due to several things, including:

1. When the engine is running, the higher speeds of the pistons don't allow as much time for pressure to bleed past the rings.
2. While the engine is running, a lot of oil is coating the cylinder walls, which helps seal the rings.
3. When the engine is running, pressure changes due to "scavenging"; a process caused by the valve overlap mentioned above.
4. Higher temperatures in the cylinder will create higher pressures when running

It's always important to have the engine at operating temperature when doing any type of pressure tests. Rings, pistons, head gaskets, cylinder walls, etc. all expand by different amounts at different temperatures. To help minimize the affect of this on testing, you always want the engine at operating temperature.

Compression Test

This type of test is most useful for identifying major problems (such as a hole in the piston), or leaking piston rings. The tester itself is simply a special air pressure gauge with a built-in check valve, and a hose that threads into the spark plug hole along with some adapters for different spark plug threads. These testers are fairly cheap and available at almost any auto parts store, Walmart, Harbor Freight, etc.

You must be logged in to view this image or video.


Note: Most of these testers are "sample and hold" types. The check valve mentioned above allows air to enter the gauge, but it can't leak back out until you press a small button on the side to release it. This is why a compression test will tell you nothing about whether or not you are leaking cylinder pressure. In fact, with a compression tester that is in good condition, you may be able to disconnect the tester from the plug hole completely and it will still show pressure.

You perform a compression test by first attaching the pressure gauge to each cylinder's spark plug hole one at a time, and you then crank the engine over and note how much pressure each cylinder generates. The idea is to record the pressure reading for each cylinder, and then compare the readings to each other and to the recommended value. The actual steps for this test are listed below.

Compression Test Procedure

1. Make sure the battery is fully charged and the car is completely warmed up (drive it for a few minutes). This is important! If the battery is weak, the engine will turn over slower and the readings will be inaccurate.

2. Pull the MPI fuse on the positive terminal of the battery. This keeps your injectors from spraying fuel during the test.

3. Unplug the coil connector on the back/right side of the block.

4. Remove all 4 spark plugs, being careful to not allow anything to fall into the cylinders. This is also a good time to inspect them for signs of trouble.

5. Make sure you have the correct adapter, and screw the compression gauge into the spark plug hole. If the tester has good o-rings, you can usually get by with finger tight; but you need to make sure that air doesn't leak from the connection or your results will be off.

6. While keeping the gas pedal pushed to the floor, crank the motor over for 5 full revolutions. Each time the motor turns over, the needle on the gauge will jump up slightly less than the time before. If the needle doesn't stabalize, crank the motor for up to 10 seconds until it does. If it still doesn't stabalize, check your connection.

7. Record the last number that the gauge rests on. This will be the compression reading for that cylinder.

8. Repeat steps 5 through 7 for each cylinder. Once completed, go back through all 4 cylinders again and get an average; (I like to do it 3 times). If one the readings for any cylinder is way out in left field, you may have a loose connection or flaky gauge and will need to re-test that cylinder.

9. Once you have good average numbers for all 4 cylinders, compare them to each other and to the factory specs. They should be within 15 psi of each other, and above the factory service limit. If not, procede to do a leak-down test to help identify the problem and verify the results. The list below shows the factory specs for compression on several DSM' engines; if you have aftermarket parts, these numbers will probably vary. If anybody has standard and limit compression numbers for various aftermarket combinations, please PM me and I'll add them to this list.

Note: These numbers were taken from a VFAQ page. If there are errors, please let me know.

1G (90-94) T/E/L 1.8L NT engine
Compression ratio 9.0:1
Standard compression 185 psi
Service limit 131 psi

1G (90-94) T/E/L 2.0L NT engine
Compression ratio 9.0:1
Standard compression 192 psi
Service limit 145 psi

1G (90-94) T/E/L 2.0L Turbo engine
Compression ratio 7.8:1
Standard compression 164 psi
Service limit 121 psi

2G (95-99) T/E 2.0L NT 420A engine
Compression ratio 9.8:1
Standard compression 170-225 psi
Service limit 100 psi

2G (95-99) T/E 2.0L 4G63 Turbo engine
Compression ratio 8.5:1
Standard compression 178 psi
Service limit 133 psi

2G (95-99) T/E 2.4L NT Spyder engine
Compression ratio 9.5:1
Standard compression 192 psi
Service limit 146 psi

EDIT: My 2.3l 6-bolt stroker with 8.8:1 pistons reads 187psi.

"Wet" Compression Test

The idea of the wet compression test is that the addition of some oil will help the rings seal. In theory, if the compression numbers are better with the wet test than they were dry then you have worn rings.

The problem is that for one thing, the wet test is dependent on the profile of your pistons. If all the oil remains at the center of a dished piston instead of out at the rings, then the results won't change much. Another problem is that adding too much oil can effectively lower the displacement of the cylinder and naturally increase compression, leading to false assumptions.

If you find that the average for one cylinder is 15-20 psi lower than the others, you can add a tablespoon or so of oil to that cylinder through the plug hole and repeat the test. If the compression comes back up, you may have worn rings on that piston; but keep the points mentioned above in mind and take the new results for what they are worth.

Leak-down Test

While similar in some ways to the compression test described above, a leak-down test is quite different...in fact, some people call it a compression test done in reverse. The idea is to inject pressurized air into the cylinders, measure the amount that is leaking, and then listen for where it is leaking from.

To do a leak-down test, you'll need a special tester that consists of two gauges, a metering orifice, a pressure regulator, and a hose with adapters for attaching it to the spark plug hole. This usually all comes assembled together in a case as a complete tester, and can be purchased at most auto parts stores, Harbor Freight, etc. It's a bit more expensive than a compression tester, but still under $40 from some places. There are also ways to make a DIY tester if you have the right parts, but IMO you might as well just spend $35-$40 for a dedicated tester and get the storage case as well. You'll also need an air compressor to inject pressurized air with.

You must be logged in to view this image or video.


Once a pressurized air supply is connected, one gauge on the tester measures the pressure of the air entering into the cylinder and the other measures the percentage of the air escaping (leaking) from the cylinder. The percentage of air that is lost will indicate the condition of the cylinder and overall condition of the engine.

Before air is put into a cylinder, that piston must be at TDC on the compression stroke. There are a couple of reasons for this:

1. When a piston is at TDC, both valves of that cylinder are closed. If they weren't, pressurized air would leak out through the opened valve and into the intake or exhaust...which would be perfectly normal for an open valve. :)

2. You will be injecting 75-100psi of air into the cylinder, which if you do the math can equal about 900 lbs of force pushing down on the piston. If that piston isn't at TDC, the air pressure will easily rotate the engine. It also wouldn't hurt to use that special crank holding tool that you made for tightening your flywheel bolts...if you have one.

Leak-Down Test Procedure

1. Remove the spark plugs and rotate the engine to put the cylinder to be tested at top dead center. Tip: insert a long screwdriver or extension into the spark plug hole and turn the engine by hand with a socket on the crankshaft. When the screwdriver stops rising or falling you're at TDC. Don't turn the engine backwards if TDC is missed. Go around again.

Note: In our engines, cylinders one and four and two and three are at TDC at the same time, but you need to be sure the one you are testing is on the compression stroke; in other words... make sure all the valves are closed for the cylinder being tested.

2. Put the vehicle in gear and set the parking brake to help prevent the engine from turning when air is compressed into the cylinder. You can also use a wrench on the crank bolt, but be sure to orient it so that it won't loosen the bolt!

3. Select the correct adapter and screw the short air line into the spark plug hole (without the tester connected), and hand tighten it by turning the hose.

Note: Adding a bit of oil to the o-ring on the adapter will help it seal and keep it from wearing out as quickly.

4. With the regulator turned fully counter-clockwise, connect your compressed air to the tester's inlet. Turn the regulator knob clockwise until the percentage guage reads "0" or "Set", and then lock the regulator (usually by pushing in on the knob).

5. Connect the guage to the adapter line that's screwed into the cylinder, and record the leakage percentage for that cylinder from the second gauge of the tester.

6. Remove the oil dipstick, radiator cap, and oil filler cap and listen for escaping air.

7. Open the throttle body, pull the BOV line at the intake manifold, and listen for air leaking in the intake and exhaust.

8. Close everything back up, and repeat for the other cylinders.


No engine is perfect, and all will have at least some loss. The percentage of air lost should be consistent across cylinders: if only one cylinder has major air loss, then there is a problem specific to that cylinder. Use the following loss percentage numbers as a rough estimate of the overall health of your motor:

1. Less than 5% - You either did something wrong or have one hell of a nice tight engine.
2. 5% to 10% - Great to good. You should have no worries.
3. 10% to 20% - Your motor may still run ok, but it isn't at it's best. Parts may be starting to wear out....keep an eye on it and test it more regularly.
4. 20% to 30% - It may be (and probably is) time for a rebuild.
5. Greater than 30% - You have some major engine issues that need to be addressed.

While listening for leaking air on each cylinder, the following guidelines will help identify potential problems. This isn't 100% foolproof, but does give you a good place to start looking for issues.

1. Air whistling out of the intake filter, at the intake manifold, or through the intake piping - probable leak at the intake valve
2. Air heard hissing out of the tailpipe, turbocharger, or exhaust manifold - Probable exhaust valve leak
3. Whistling or hissing out of the PCV valve, oil filler cap hole, or dipstick tube - Worn piston rings or cylinder wall.
4. Air bubbles in engine coolant at the radiator cap or in the coolant overflow - Possible blown head gasket. May also indicate a crack in the cylinder head or cylinder wall.
5. Air escaping from the spark plug hole of an adjacent cylinder - Likely a blown head gasket between cylinders.

Wrapping it up

In addition to the two most common types of cylinder testing mentioned in this article, there are other types of tests that can be done as well. If you are really feeling adventurous, Google the phrase "running compression test" for more information. ;)
 
Last edited by a moderator:
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top