What are good leak down test numbers?

[zracer22]

I recently purchased a 1.8T from a 2003 VW GTI with 65K to drop into my Jetta race car. I started to do a leak down test on it, but stopped after a bad result on the first cylinder. The problem is that the engine is cold and dry , so i don't expect to see 2-5% numbers. What would be good numbers on a cold and dry motor? I've also read that the ring gaps on turbo motors are wider and the leak down numbers will be a little higher. Is that true?

Thanks in advance, Mark

[Team SSR]

If the engine isn't installed in the car yet it is usually easy to determine where the major leaks are just by listening. If the air is leaking into the pan it is blowing by the rings, if it's leaking into the intake or header it is the corresponding valve. If it is a valve that is leaking it is sometimes because of carbon or 'ick' that will clear up when the engine is fired (or the valve is tapped with a plastic mallet to open and close it rapidly). If you don't like the leakdown numbers you are seeing, try to determine a pattern. If all the cylinders are leaking into the oil pan, it probably needs a rebuild. If all of the exhaust valves are leaking, it may have been overheated or had valve to piston contact. If only one cylinder has a problem look for obvious problems - evidence of water in the cylinder, sparkplug condition, etc.

[zracer22]

I'm aware of how a leak down test works, my question is how does a cold engine without any oil in the ring grooves effect the results of a leak down.

[Quickshoe]

I'd "guess" that if it is all through the rings you need less than 10% cold to be good. My motor is around 2-3% warm and is still less than 10% leakage cold...it is aircooled so I don't know how much difference there is with piston/cylinder clearances cold vs. hot and ring gap. YMMV. Just don't expect huge number improvements when hot over cold.

Try your lawnmower or street car hot and cold see what kind of change you get.

[Team SSR]

The difference in a hot vs. cold motor is not oil in the rings, rather it's the shape and size of things when they are at temperature. Cylinders, pistons, and valve seats all change shapes and sizes when at temperature. Each motor is a little different. Engines that run at higher temperatures generally have wider ring end gaps to accomodate for the greater expansion which is why a turbo engine usually requires a larger gap. An exception to this would be an engine with aluminum cylinder walls that expand similar to the piston.

If your engine hasn't been run for a while it may not have a good leak percent. Small amounts of rust and carbon, etc... may keep the rings and valves from sealing - when the engine fires it will likely clear up. This is why I hesitate to recommend a definite percentage number to call 'good' or 'bad'. Rather, if you have a cylinder that you think is bad try rotating the motor over several times and see if it gets better. If it does it will probably be OK when the engine runs. On the other hand if it doesn't get better I would consider investigating the problem now, before the engine is installed.

Having said all that, an honest to goodness race engine that leaks 10% (at 100lbs. cold) will probably not be noticeably down on power from new, at 15% it may be. I suspect a turbo engine may be slightly less affected by the engine seal-up because of the increased cylinder pressures and the 'push' the intake charge gets. It still comes down to a judgment call that is usually weighed out in leak% vs. time/money available.

[Andy Bettencourt]

You also need to do a little investigation as to were the leakdown is. With a little creativity and a quiet shop, you will be able to isolate on each cyl where the air is coming from - intake valve, exhaust valve or rings...

[Charlie Broring]

A "good number" varies from tester to tester. The calibration is determined by the size of the air orifice and isn't really standardized. An engine should be warmed up and preferably run hard prior to testing for the most meaningful results. As SSR says, I wouldn't trust the leak down readings for your motor that has been sitting unused for a while.

I don't trust leak down testers to evaluate ring seal. I use a "blow by gauge" which is a flowmeter attached to the otherwise sealed up crankcase to measure gas flow past the rings while running.

Charlie

[bobqzzi]

For a 1.8T that has been sitting around for a while and is cold, I wouldn't be concerned until it got over 20%. Even then, if it is a bent valve it will be 100% and 1.8Ts seldom have ring problems.