Blogs Car Info Our Show Deals Mechanics Files Vehicle Donation

Can someone explain leakdown test? And WHEN it is appropriate?

Do you only do a leakdown if the compression test is bad?
Any basics on this? They are just words to me.
When do you do on, not the other, or both?

Leakdown test is how well the cylander holds pressure. Pressure test is compression while the engine is working. Both can be used to evaluate head gasket and or cylandar performance. Outboard motors have been my experience with leakdown tests, cars usually seem to navigate towards compression tests.

A leak-down test confirms one thing. If there’s a leak in a certain cylinder.

If you apply compressed air into a cylinder while it’s at TDC,. that compressed air allows you to determine if there’s a problem with an intake valve when air comes out the intake system, or if there’s a problem with an exhaust valve because air comes out the exhaust pipe. Or if there’s a problem with the rings because air comes out the dip stick tube. Or if bubbles start forming in the coolant because of a head gasket breach.

A compression test is a dynamic test where the pistons move up and down quickly where it won’t indicate a problem in these areas. But a leak-down test is a static test. And will show the real problem over a compression test.


Here’s an article:

Do you do them in a certain order?
Are they mutually exclusive?

Like if the compression test is good,
do you bother with a leakdown?

Since a compression test is quicker, I would do that first, if the results are good I wouldn’t bother with a leakdown test.

Is that just preference, or is that hard logic?
If compression is good, there is no logical need for a leakdown?

Yes, as far as I know.


Does one of your engines have a problem which you are trying to diagnose and resolve?


It’s like if you thought your bike tire had a leak, but not sure. So you connected a bicycle pump (which has a built in pressure gage) to the bike tire, pumped up the tire, then left the pump connected. If the tire has a leak, the gauge will slowly move down. “Leak Down” in other words. No leak, the gauge will remain steady.

If you are trying to find a problem a compression test is quicker. If you want to find the condition of your engine, a leakdown check is best. The percent leakdown you get is indicative of the engine. A well broken in engine can leakdown as little as 3-5%. A well worn motor may be showing 10%. Over 20 on any cylinder, or all tells you a rebuild is in order.

@Mustangman - if there’s not any apparent problem, and a compression test looks good (even pressures within spec) is there a reason to do a leakdown test?

A compression test won’t always show a problem.

I had a Jeep Wrangler with a 4.0 liter engine come in with misfire codes for cylinders 3 & 4. After checking the obvious things that can cause a misfire I did a compression test. All readings were within specs. So then I did a leak-down test. When compressed air was introduce into cylinder 3, air started coming out the spark plug hole for cylinder 4. And when air was introduced into cylinder 4, air came out of the spark plug hole for cylinder 3.

Here’s the head gasket that was replaced.


Great photo @Tester . Instructive why doing a leak-down test can prove a very useful diagnostic.

How did you know to take the #4 spark plug out?
Are all plugs out during a leak down?

Yes all plugs are out to make turning the crank by hand easier.


Don’t take this the wrong way, please

I can recommend some good automotive textbooks, if you’re interested

@texases, Yes, there are reasons to go ahead and run the leakdown test even if the compression test is OK. @Tester had a good example although I’m scratching my head as to why a compression test didn’t catch the head gasket. The weakness in the compression test is the variability you get with it. The acceptable range is wide because atmospheric pressures affect the result, engine temp (hot when you do #1, warm when you get to #6) and battery condition (how fast the starter spins the engine. The leakdown eliminates most of these, except engine temp plus it gives you a nice defined % leakdown number that can be compared to an absolute. When you run a compression test and all is well but the problem persists… get out the leakdown tester!

Sure, what books? I’ve already read the Reader’s Digest book from the 80s.