Leak-Down vs Compression Test

Compression screens, leak-down diagnoses. They are not two ways of doing the same job, and the order matters: the compression test tells you whether a cylinder is down, and the leak-down test tells you where the air is going.

If you have no compressor, that decision is made for you.

A note on this comparison: this rests on what each test measures and how each is performed, not on tests run here. See how we test.

The short version

Compression test. The engine makes the pressure while cranking. You read a peak figure per cylinder. Cheap, no compressor, quick — and it tells you that something is wrong without telling you what.

Leak-down test. You supply regulated air into a stationary cylinder at top dead centre and measure what percentage escapes. Needs a compressor. It tells you where the air is going, because you can hear it leaving.

What actually separates them

The listening is the whole point

This is what people miss about leak-down: the gauge is only half the instrument. With the cylinder pressurised, where the air emerges names the fault.

• Exhaust pipe → exhaust valve not sealing
• Intake or throttle body → inlet valve not sealing
• Oil filler or dipstick tube → rings or bore wear
• Bubbles in the coolant header tank → head gasket or a crack

No pressure figure from a compression test can distinguish those. That is the entire reason to own the more awkward tool.

Both read relative, not absolute

Neither test has a universal pass mark that is worth much.

A compression figure varies with cranking speed, battery condition, engine temperature, altitude and compression ratio. A leak-down percentage varies with engine design and with the tester itself.

What is diagnostic in both cases is the spread across cylinders. One cylinder well below its neighbours is a finding. All of them moderately low is usually an engine with mileage on it, and occasionally a technique problem.

The wet test bridges the gap for free

If a compression test finds a low cylinder, a small amount of oil down the plug hole and a re-test tells you something before you buy anything.

If the figure rises noticeably, the oil has temporarily sealed the rings — so the leak was past the rings. If it does not change, the leak is somewhere the oil cannot seal, which points at a valve or the gasket.

That is a genuinely useful narrowing at no cost, and it is covered alongside the tool in the OTC 5605 review.

Both mean removing every plug

Worth saying because it is the actual labour. On a modern engine with coil-on-plug and an intake in the way, plug removal can be most of the job — and doing it badly is its own expensive outcome.

Spark plug tools before you start covers not damaging anything on the way in.

The decision, plainly

• Screening an engine you are thinking of buying → compression
• A misfire that codes have not explained → compression, then leak-down if a cylinder is down
• Oil consumption with no external leak → leak-down, listening at the oil filler
• Suspected head gasket → combustion leak test first, it is cheaper and more specific
• No compressor → compression, plus a vacuum gauge
• You want one tool for occasional use → compression

This page contains affiliate links. If you buy through them, Ugly Fixes may earn a commission at no extra cost to you. Products are selected independently.

A compression tester suits nearly everyone, and it is the one to own if you own only one. Check the adapter threads cover your engine.

A named option with verified specs, if you would rather not choose blind — the OTC 5605, reviewed.

A leak-down tester suits anyone with compressed air and a cylinder already known to be down. Dual gauge is easier to interpret than single.

Spark plug tools suit both tests equally, because both start with removing every plug.

The mistake to avoid

Buying a leak-down tester without a compressor. It is the more interesting tool and the more informative test, and it is completely inert without a continuous air supply. That constraint is easy to overlook on a listing and it is the single reason these end up unused.

The second mistake is treating either figure as a verdict on its own. A low compression reading on a cold engine, cranked on a tired battery, with the throttle closed, can look alarming and mean nothing — and before any of this, a vacuum gauge on the inlet manifold reveals valve, timing and exhaust restriction problems without removing a single plug, which is why the old-school diagnostic is still the cheapest first move.