Leakdown and compression test results question

[Tom '74 911]

Hi -
The motor in question is a 2.7L CIS. I was wondering if it is possible to have good compression results and bad leakdown results from the same cylinder. Here's some numbers:

cyl./leakdown/compression
#1 64% / 119
#2 8% / 118
#3 9% / 118
#4 6% / 118
#5 5% / 120
#6 5% / 125

So obviously the numbers are very consistent except for the leakdown on #1. I/we could not for the life of us figure out where the air was escaping to. I took out the injector and air cleaner and it did not seem to be coming out the intake. There did not seem to be a noticeable flow out the exhaust either - I assume w/that much air moving, it'd be pretty obvious? So that leaves the rings - or a faulty gauge I suppose as it was the last cylinder I tested? Could the compression be as consistent w/faulty rings or an otherwise compromised combustion chamber? Both tests on this cylinder were done multiple times and back to back - to try to get the results of the leakdown better and to diagnose the issue. If there were a piece of dirt or carbon stuck in one of the valves and hindering the leakdown number, would the compression test be as good still?

Thanks,
Tom

[MURF]

To check your rings..listen inside oil tank with leak down pressurized..to check for bad rings on compression test,you put oil inside cyld and pressure will go up..same with leak down oil will fill up the rings and reading will be lower.....could just be ring gaps rotated into alignment....

[JFairman]

Take a section of hose and with one end held to your ear while pressurizing the combustion chamber with 100-120 psi air, take the other end and put it in the oil tank filler or breather hose to listen for ring leakage, the throttle body while open for intake valve leaking and if your car has an oxygen sensor remove it and put the end of the hose in there to listen for exhaust valve leakage. You might hear some hiss at the tailpipe if there is no oxygen sensor and the exhaust valve is leaking.

With 64% leakdown you should hear a hiss somewhere.

[RichHawk]

A better question: Why were you doing the leakdown and compression tests? What symptom were you chasing?

[OldTee]

Go run it then recheck #1 cylinder leak down. Probably some carbon under a valve.

[Tom '74 911]

Originally Posted by RichHawk

A better question: Why were you doing the leakdown and compression tests? What symptom were you chasing?

I'm really just looking for an excuse to upgrade to a 3.2 or 3.6. The motor really runs well actually. A little puff of smoke on startup every now and then, a few weepy areas (not enough to actually drip though), needs a new muffler, that's about it. I was running a bit roughly a few weeks ago, but I think I may have filled up with a bad tank of gas because it was fine before and it's fine again now after another fill-up.

"Go run it then recheck #1 cylinder leak down. Probably some carbon under a valve."

I think this will be my plan for now.

Thanks,
Tom

[Ericv]

I have a very similar issue. A compression test reveals that all cylinders are around 140-150 except for cylinder #1. This cylinder is at 70. I hear a distinctive hissing sound coming from that cylinder. If it's a valve issue, can the headwork be completed for cylinders 1, 2, and 3 only since 4, 5, and 6 reveal good numbers? And, if it;s a ring issue as long as the ring gap is within spec can the rings for that cylinder be replaced? Or would it be better to replace the rings on cylinders 1, 2, and 3 so they would have the same wear? I've already resigned my self to the fact that the left side of the engine needs to be torn down to the pistons but I didn"t know if I needed to do the right side as well. Sorry for the hijack.

[Scrappy]

Since the head heights are set evenly so the cam tower will bolt flat and seal, you could feasibly do one head if the height is not screwed with. Cheap way out, but could work.

I had to replace one so I took three to the machine shop ... 2 off the same bank and the new used one. They matched the used one to the others. This was on a 3.0L.

The gasket kit is about $270 (?) to do the heads. If you want to rering all the pistons the gaskets are already paid for.

[Walt Fricke]

Tom, with 64% leakage you really ought to hear where it is escaping. If it is carbon under a valve, you'd still hear it leaking out.

There is always some leakage past the rings. Even when both my guages read 100 (=0% leakage), I can still hear air leaking into the case.

Sounds like you aren't getting leakage out the intake valve, so listening in the exhaust end should be revealing. And the oil tank. To see what either of those should sound like with normal leakages, try putting air into one of the other, "good", cylinders and listen.

My first thought was that somehow you had missed TDC on #1 Lots of marks on the pulley in that vicinity. But neither valve is open any where near the real TDC (spark TDC), and if you aren't on TDC for the hole you are testing, the air often will rotate the crank. Now, if you actually were at TDC #2, between the nominal intake and compression strokes, that's different.

As mentioned, you almost don't need the gauges for this to be meaningful. Get at the proper TDC, hitch the air hose to the connector you have in the spark hole. and listen.

Run the engine a little, then retest #1?

Walt Fricke

[Alan L]

I wonder if the leakdown unit was sealing properly in the plug hole - being one of the harder ones to get to.
Alan

[john walker's workshop]

the compression is somewhat even, but not really good. should be around 150. it will vary, depending on the gauge, engine hot or cold, cranking speed, how many pumps per cylinder, etc. i always run the engine for a few minutes then retest any cylinders that had a high leakdown. then check that the bad cylinder has clearance on both valves.

[Tom '74 911]

Originally Posted by john walker's workshop

the compression is somewhat even, but not really good. should be around 150.

I am at about 6000' - might that make a difference in my compression numbers? I wasn't too stressed about them as they were pretty even. I haven't had a chance to retest the leakdown on cylinder #1 - hopefully in another week or two.

Tom

[Alan L]

6000 rpm?
What are you spinning the engine with?
Alan

[Tom '74 911]

Originally Posted by Alan L

6000 rpm?
What are you spinning the engine with?
Alan

Sorry - I meant 6000' in elevation. The air is less dense here so I was thinking maybe less compression?

[Alan L]

It would hardly make a difference - few % perhaps - temp variations would be about same difference - eg compression on a hot sunny day and cold winters day.
Alan

[Walt Fricke]

Well, I have always been told (by sources unremembered, but of some stature) that altitude does make a difference in these kinds of gauge compression readings, and that one should take note of that effect when interpreting the data.

After all, the air is measureably less dense at 6,000' ASL. It calls for smaller carb main jets, it doesn't cool as well, and you can run higher nominal compression with the same octane, or use lower octane with the same geometric compression. And pilots know all about resetting their altimiters. So did we mountaineers in the pre-GPS days. Barometric readings have to be adjusted to correlate with sea level for certain comparison purposes.

I use an air density meter for jetting, and yes I may change a jet size or two over the course of a day with significant temperature changes (increases, typically). But not as much as when I go to College Station or Topeka down from Colorado.

But I haven't run the math.

Walt Fricke

[Tom '74 911]

Originally Posted by Walt Fricke

Well, I have always been told (by sources unremembered, but of some stature) that altitude does make a difference in these kinds of gauge compression readings, and that one should take note of that effect when interpreting the data.


Walt Fricke

Thanks Walt. Either way, it's not the compression results that concerned me too much as they were pretty even across all 6. The low overall numbers could be altitude, the cheapo gauge I have or something else???? I tried to be consistent in the number of times I turned the engine over for each cylinder - but in the end I am only human and not too concerned with the nano or micro of things (if that makes sense).

Walt - are you planning to go to Miller in a few weekends? You were my instructor there a bunch of years ago during the first club race held there the year it opened. I'll look for you in the paddock and say hi.

Tom

[Walt Fricke]

I figured there weren't all that many Toms in Idaho with 2.7s, and yes I am finally getting back to Miller. See you there.

Walt

[Alan L]

I may have put you slightly wrong before.
At 6000 ft, you have about 20% less air density.
So you could probably add about 20% to those numbers. That will make them better.
A 20 deg C change in air temp gives you about half that value. So hot day vs cold day is about 10% variation.
Regards
Alan