Compression and Leak-Down give VERY different results
 

My 1988 911 seemed to lack some power. I did a dry compression test on it and got 180 psi on all cylinders except #1 which was 165. (I suspect I may have gotten some dirt in there during the test so was planning on re-testing).

I brought the car to the Porsche dealer. They did a leak-down test and found 90% (yes, ninety percent) leakage in cylinder #4.

I understand the tests are different but can someone give me an explanation for the wide variance in readings? From my brief conversation with the service department (I didn't speak with a technician), it sounded like he was saying you can have a leak that comes out slower which wouldn't show up on the compression test. He thought it might be a valve seat or valve guide. However, from what I've read, 90% leakage indicates a hole in your compression chamber. Also, he told me that you can't detect where the air is escaping while the engine is in the car (I thought you could).

Thanks for any explanation you can offer.

Ken K.

Hi Ken:

A few thoughts,...

1) Dry (cold) compression testing is frought with errors and inconsistencies. It really tells you nothing constructive due to all the variables.

2) IMHO, Leakdown testing is the only useful way to help determine engine condition since it not only gives you a hard number, it allows one to pinpoint the SOURCE of the compression loss.

If the technician observed a true 90% leak, he should be able to tell you WHERE it was coming from. If he heard air escaping from the air cleaner, that indicates bent intake valves. If he heard air escaping from the tail pipe, that indicates bent exhaust valves. If he heard air coming from the oil tank filler tube, that indicates ring and/or piston problems.

These engine tests can easily be performed with the motor both in and out of the car,.........been doing just that since 1978. :)

I would ask them to do it again and preferably by a competant technician knowledgable & experienced with these procedures.

Ken,
Welcome to The Forum SmileWavy,
You will find a lot of help here.

Steve and I are in agreement for 35 years or so.

Never depend on a single measurement. I recommend you go drive your 911 for an extend period and some “sporting” driving. Then repeat both tests. Depending on the results, you may want to repeat several times. It is far better to do more testing than make decisions on less-than-perfect information.

I differ with many in that I will utilize several “cold” cylinder leak measurements. I don’t think they have to be done hot (warm) to give very useful information.

In your situation, I suspect the measurement. It is very unusual to have consistent compression among those cylinders and that high leak on one. It is possible the leak is at the hose to the sparkplug port. The technician should report where the leak is.

Best,
Grady

Steve and Grady - Thank you very much for the detailed response. I'm brand new to the forum and to Porsche, so I appreciate the support.

I've got a call back into the Porsche dealer on getting better info where the leak is coming from. I didn't talk directly to the tech, so I'm assuming he knows the answer.

When I said a "dry" compression test, I meant without putting a bit of oil into the cylinder (to see if it is the rings). I may have used the wrong terminology. The car was at operating temperature during the test (although I did forget to have the throttle wide open during the test... not sure how much difference that makes).

Grady, I noticed you advised driving the car and re-doing the test. If there is truly 90% leakage, would it be possible to make matters much worse by driving the car?

Thanks again.

Ken K.

Ken,

I don't believe there can be acceptable (and consistant) cranking compression yet 90% leakage. I suspect the leak is an improper test.

You have been driving your 911 with no adverse effects other than you sense a decrease in power. You seem to indicate it is running on all six. If a cylinder were 90% leak, that cylinder wouldn't be running.

I strongly agree that an engine with damage shouldn't be stressed or even run.


Perhaps a better course of action would be to repeat both tests and make a decision based on the new results. To get accurate results, the car needs to be driven and stressed. The most accurate measurements are just after running with no intermediate start-ups.

You will find my proceedures posted a few years ago.

I just spent an hour+ on the phone with Steve solving these and other Porsche and World issues. :D

Best,
Grady

Thanks, Grady. Porsche just called me back and said that they can hear the air leaking, but can't tell where the leak is coming from.

They did indicate that the tech could hear a misfire happening in that cylinder, but (being inexperienced as I am), I only noticed that it seemed to lack some power. The engine didn't sound like it was misfiring to me and overall drove well.

I think I will do both tests again either myself or at a different Porsche specialist.

Thanks again for your input and for helping solve Porsche and world issues.

Ken K.

Ken,

Good plan. Repetitive tests inspire confidence in the results. Science 101. Having someone else perform the tests is very wise.

Depending on the technician, “hearing” the air escape can be from a problem (like head seal) but it can also be from a poor seal at the sparkplug port with the leak test hose. With 90%, the latter is possible.

I’m sorry. You can’t tell which cylinder is missing by listening. That is only possible with electronics (a scope and other) or by deactivating ignition to each cylinder selectively.

Best,
Grady

If a tech told me he could hear air leaking but couldn't tell from what location, I would take my car to a different shop.

It's not rocket science.

A small piece of carbon can cause a valve to get stuck open and leak during a test. Take it out and drive it like Grady suggested, then take it to a reliable independent Porsche shop.

Ken,

Where are you located? If you can let the community know that we can steer you to a mechanic who knows how to do a leakdown test.

How is it possible to have good compression on a cylinder with 90% leakdown? Well, it's only possible when the mechanic doesn't have a valve closed all the way-- the timing mark must be aligned with the notch on the blower housing, and held there when the air is turned on. Sounds like they pulled the plugs, put the air hose adapter in the hole, then slapped the air on, moving the engine, then claimed it was 90% leakdown.

That's the only explanation that makes sense to me. Something's really not accurate.

John - Thanks for the response. I am located in the Chicago area (western suburbs). I went to a Porsche dealership thinking that it would be the best option, but I haven't been impressed with this dealership. Their conclusion was a $12,000-$14,000 rebuild. If the problem was the test itself, then I will be..... well, happy. But unhappy with them.

I would love recommendations on reliable Porsche mechanics in the Chicago area. I did a little research and think I found one in Barrington, Illinois (Fischer Motors) who said they could do a repeat leak-down test and a test drive for me at a reasonable cost ($175). They said that what I was told was inadequate in terms of an informative test result.

Any recommendations from anyone would be very much appreciated.

Thanks again for the help!

Ken K.

Ken:

Go see Midwest Eurosport,..... :) :)

My car showed one cylinder down at 160-165 on compression and down to 65% for leak-down, all the rest were 180-185 and 95%. Turned out to be a broken compression ring.

I agree with Cliff. The testing isn't hard to do but it does require some skill/experience to interpret the results. Dieters Porsche in San Diego has a pretty good selection of technical articles. One is called "Compression test vs. Cylinder leak down How, why and what do they mean?" The Valvoline website also has a good article called "Compression Check-Up" which explains the compression testing process pretty clearly. Even if you don't perform the test(s) yourself, it's always valuable to know what the technician is doing so you can better understand if what he/she is telling you is correct.

So far, the responses include phrases like, "must be done right," "require skill", etc. These phrases and the one I'm going to add: ("Your mechanic should inspire confidence in his ability and the fact that he's not going to rip you off" ) do not equal dealership.

Please go to an independent mechanic who comes highly recommended. Just consider yourself lucky that your car is out of warranty and you don't HAVE to bring it to a dealership.

I may get flamed for making such a broad statement, but in my experience dealerships more often than not make up problems and mis-diagnose issues, not to mention show utter disrespect for your vehicle.

Just a quick story: At 50,000 miles my Audi allroad was diagnosed by the dealer as needing a new clutch. I brought it to an independent Audi mechanic who informed me that there is no way the dealer could diagnose that due to the clutch design, especially since they didn't even drive the car. Anyway, 30,000 miles later the clutch is still acting as it should. I'm sure I spent the $2,600 the dealership wanted on something frivolous anyway!

Oh, one more thing: a pre-1999 Porsche might as well be a spaceship to a Porsche dealer. It says Porsche on the hood - that's about all they will recognize.

Valve seat maybe. But contrary to popular belief, bad valve guides aren't going to show up in a leakdown or compression test... unless there is so much "wobble" that the valves aren't seating properly.

Thanks everyone.

Steve - Thanks for the tip on Midwest. I've checked further and found others saying good things about them, so that was helpful.

Crosby - Yes, I have been restraining myself on speaking bad about this dealership, but the service has been awful. The car has been in there 9 days. They actually don't return phone calls as promised. The car also needed a front condensor blower motor for the a/c. I was told by them that this is a "tricky repair" (90 minutes) requiring some work from underneath the car which I believed. I checked in Bentley and found that the repair required removing the battery, the spare tire, disconnecting a housing and removing 4 bolts to slide the motor out, all from within the luggage compartment. I called them back and asked them if I was looking at the right motor and they insisted that this is a tricky repair. They are billing me 90 minutes for this repair. I am a new Porsche owner and really wanted to like this dealership. I am definitely going the route of an independent Porsche specialist from now on. OK, done with my rant. I'll be nice now.

Ken K.

I took my first car a 1990 c2 to the local dealership for a PPI. I specifically asked them to make sure the mating surface on the cylinder to head surface was sealing properly - a early 964 common issue that I learned from reading on Pelican Parts -

They did the PPI - said that the engine was leaking a little oil near the O ring for the power steering pump, that this was the only source of oil - gave me a quote of $800 to fix - which I used to bump the purchase price down.

3 months later and I am cleaning the engine trying to figure how hard the o-ring is to replace - and bam - wouldn't you know it - the cylinder to head is leaking and bubbling - and it's carmel colored - like it's been doing it for years.

I called the dealership - and the service manager tells me that it wasn't doing it when they looked at it. :) -

Well, I finally have my car back. It is running a bit rougher than before, so something has happened since I dropped the car off. The paperwork on the leakdown test gave no details other than "serious air leak from cylinder #4" (although they verbally told me that leak was 90% in that cylinder and all other cylinders were at specifications).

I think I will do both my own compression test and also try my first leakdown test. The car is running rough enough that I am nervous to drive it. If the problem is getting worse now, I should probably just pull the engine (yikes). That's a topic for another thread though.

Thanks again everyone.

Ken K.

Midwest Performance Cars in the West Loop also does good Porsche work.

www.midwestperformancecars.com

-Mark

They just did mine. Tom didn't like what the first compression test showed, he didn't think the numbers were accurate, so he did another one without even me asking just to be sure of the results. I was impressed.