Leak down test results

[Bala]

Did the cylinder leak down test and got the following results (bought the tool from harbor freight).

Air compressor pressure out was at 80psi
The pressure guage on the tool seemed to be at a constant 25-28psi

The loss guage showed the following:
Cylinder 1 - 3%
Cylinder 2 - 5%
Cylinder 3 - 2%
Cylinder 4 - 1%
Cylinder 5 - 7%
Cylinder 6 - 1%

Seems pretty good for a car with 92k but would like to make sure that I did this right before I celebrate. Here's what I did:

Removed the coils,
loosened the plugs and re-tightened them just enough,
re-installed the coils without the bolts,
warmed up the car to operating temp,
disconnected the battery,
removed the coils and plugs,
set the engine at TDC and locked it,
started test with cylinder 6 and moved up.

The tool instructions says to leave the oil dip stick and coolant cap out. I did hear a faint hiss thru the dip stick hole but the coolant fill hole was quiet.
I did have to rotate the engine 360 for cylinder 4, 3, 2 & 1. Otherwise the pressure wouldn't hold. The instructions by Wayne are for a 911 so I hope the above is how it needs to be done.

I could not hear any leaks on the exhaust either. I didnt have access to the top side but no audible noise that I could tell.

Could the experts please chime in? Appreciate any comments. thanks

[Jake Raby]

Cylinder 2 and cylinder 5 are questionable. These are the hottest running cylinders and show signs of deformity before any of the others. I
'd consider your value on cylinder 2 to be borderline and cylinder 5 is out of my spec. There are no factory values for this, so like everything else we have had to write our own book.

[shadrach74]

A leak down check is a single data point and can often change one day to the next. What was the impetus for doing a leak down test? Is the car using a significant amount of oil? Is it down on power? Oil fouled plugs?

In the aviation industry we do compression and leak down checks annually at a minimum and in many cases every 100hours of tach time. The readings go up and down from check to check but should stay within a certain acceptable range. In aviation, more than 12.5% (70/80) gets a second look as does a variance of more than 7 percentage points. I'd be willing to bet that you could find factory new auto engines that have leak down values in excess of 10% (hopefully not Porsches). I'm sure Jake has his reasons, but his numbers seem a bit conservative to me, but then I've never built a Porsche engine, just aero and motorcycle engines...

If the engine is running reasonably, I'd not tear it down solely based on the numbers you've posted...

[Bala]

Plugs are not oil fouled but 2 and 5 seems a bit darker than the rest. I wouldn't be too concerned about that. However there is loss of power. My maf went bad and I replaced with a eBay one. Not happy. Could be due to that as well so I'm going the Bosch replacement route. It was one main reason to do the test. Others would be because car is at 92k miles and of course the most common reason being that I need to tinker around.

Given what you and Jake said I think I'll run another test when cold and see what comes up. No plans to tear down the motor - maybe not yet. Just doing some health checks while I'm under the car checking some shifting issues.

[Steve Tinker]

Doing the tests with a cold engine will not prove a thing - an engine @ normal operating temps will always give the most accurate results.
After saying that, at 92,000 miles I think the results are not perfect but not too far off the optimum......
Theres a lot of owners out there would love those leak down resulta at that mileage.

[Bala]

Originally Posted by Steve Tinker

Doing the tests with a cold engine will not prove a thing - an engine @ normal operating temps will always give the most accurate results.
After saying that, at 92,000 miles I think the results are not perfect but not too far off the optimum......
Theres a lot of owners out there would love those leak down resulta at that mileage.

Was curious what the difference would be.

Cylinder Leak-Hot Leak-Cold
1 3% 7%
2 5% 6%
3 2% 5% (this I happened to leave for an hour and it was @ 10%)
4 1% 4%
5 7% 5%
6 1% 6%


Something tells me I may not be doing this right. Especially the numbers for 2 & 5. Can someone confirm these are the right steps below?

- Warmed up the car to about 180 on the guage
- by the time I got to starting the test (jacking up, removing wheels etc) it was a good 20 mins. Engine was hot but it wasn't 5 mins wait like Wayne says in the instructions.
- The tool instructions says to remove oil dipstick, coolant fill cap and open throttle body. I did oil dip stick and coolant cap only.
- If I remember correctly, the TDC set is good for cylinder 5 & 6. Need to rotate 360 for cylinder 4 & 3. Once again turn over for 1 & 2. I could be mistaken but if the engine was not at TDC then the air leaked severely and did not hold. So I knew I was not at TDC for that cylinder. Hope that was the procedure.
- Pressurized each cylinder for a min of 5 mins and max 10 mins
- air comp outlet guage was around 75 psi and the air pressure guage on the tool itself was around 25 psi.

Sorry If the information is already somewhere. I just couldn't find it for the Boxster e-gas. What I found were all related to 911. Also Wayne's instructions do not emntion taking out the oil dipstick and the collant cap etc. So am confused about that as well.

[paulv]

Bala,

You're almost there. It is imperative that both valves in the cylinder under test are closed -- so TDC or BDC. It is better to test with TDC but some people have safety concerns and use BDC. I always go with TDC and leave the car out of gear (or in park). haven't had the need to do this with my Box, but helped out a couple of friends with their junk cars.

Here are the items required for your testing:

Throttle body open -- allows for hearing leaks from a bad intake valve(s).
Exhaust system (obviously open) -- allows for hearing leaks from a bad exhaust valve(s).
Remove oil dipstick -- allows for hearing leaks into the crankcase from blowby caused by bad rings.
Coolant fill cap -- allows for seeing bubbles in the coolant caused from cracked head/headgasket/etc.

Regards,
paul...

[Bala]

Originally Posted by paulv

Bala,

You're almost there. It is imperative that both valves in the cylinder under test are closed -- so TDC or BDC. It is better to test with TDC but some people have safety concerns and use BDC. I always go with TDC and leave the car out of gear (or in park). haven't had the need to do this with my Box, but helped out a couple of friends with their junk cars.

Here are the items required for your testing:

Throttle body open -- allows for hearing leaks from a bad intake valve(s).
Exhaust system (obviously open) -- allows for hearing leaks from a bad exhaust valve(s).
Remove oil dipstick -- allows for hearing leaks into the crankcase from blowby caused by bad rings.
Coolant fill cap -- allows for seeing bubbles in the coolant caused from cracked head/headgasket/etc.

Regards,
paul...

Have to do this again.

Did not leave the throttle body open for the hot test. But, for the cold test I could not hear anything from the throttle body.
Did hear air hissing out of Oil dipstick but manual says some leakage is expected.
No leak what so ever from coolant - good new I guess.
Nothing of significance from exhaust. Could hear a very faint echo like air noise.

Let's see how this goes again. Maybe I'll write up a DIY article on this while I'm at it.

Thanks Paul et all.

[The Radium King]

i thought compression test was the first step. if compression is equal across all cyls then you are ok. if it isn't then a leakdown test to try and determine why.

[paulv]

Originally Posted by Bala

......................... Maybe I'll write up a DIY article on this while I'm at it.
.............

You can use this guide (I forgot I had it) as a basis for your Boxster-specific DIY article.

Regards,
paul...

[Bala]

Did the warm/hot engine test again. This time with throttle body, oil dipstick and coolant cap open.

None of the cylinder showed any air noise from throttle body or the coolant fill (no bubbles either). Oil dipstick had a slight air hiss as before. I understand this is normal. Exhaust had a air noise echo. Very faint but could hear it like before.

Leak percentages were same as before but cylinder 2 and 5 were lower. This time the engine was much hotter than earlier though. Others were about same or lower. So am good? No engine teardown? I was reallt hoping for a drop and tinkering. Guess it has to wait for another day.

I guess i don't need to do a DIY. What PAULV attached is pretty straight forward.

[Bala]

From what I have read and understood, leak down test is more comprehensive and better indicator of issues than a compression test. You may skip the compression test and perform the leak down instead.

[JFP in PA]

A leak down test gives more detailed information about the condition of the engine than a simple compression test. In some cases, an engine that passes a compression test can acutally fail a leak down.