BA calls into question leak down tests - Excellence Mag

pbs911 · pbs911's Garage

I got the new issue of Excellence in the mail yesterday and began reading the Tech Tips. Someone wrote in about problems with leak down numbers. BA went on the comment that the leak down test is not an accurate accessment of our air cooled motors. Apparently there are problems of where the piston is positioned in the cylinder when the test is perfromed. He goes on to say that he has seen motors with broken rings give very good leak down numbers and motors that are in great condition provide poor results. BA suggests that a compression test is a much more accurate indcator of a healthy 911 motor.

Did anyone else read this or have comments? Sorry I don't have the article in front of me and I don't have a scanner, so maybe someone can post it.

Jim Sims · 08-14-2002, 08:55 AM

One can't depend on one test; to fully assess the health of an engine one probably needs it on a dyno with a scope on it. I have an old Chevy Blazer with great compression and low oil consumption but terrible power. Why is the power output so poor? The cam shaft lobes were soft and wore prematurely. Compression is okay but volumetric efficiency is terrible. However a compression test and leak down test combined with some thought and further checking can tell a lot. Cheers, Jim

arcsine · 08-14-2002, 09:02 AM

I found Bruces' opinion interesting as well. Certainly throws a lot of things up in the air. If the scans are not readable, email me and I will send a better version.

pwd72s · 08-14-2002, 09:27 AM

If I were to ever pay for a PPI? This article only made be believe more that BOTH leakdown and compression tests should be performed.

Early_S_Man · 08-14-2002, 09:52 AM

I found nothing questionable about what Bruce said. Many people have been taken advantage of by shady mechanics using leakdown numbers.

Taken into account with other assessments ... that you do, not just depending on others who may have a financial incentive to 'fudge' the test ... leakdown testing can be very useful in a diagnostic mode after a problem is known.

Placing heavy weighting of a pass/fail nature on a single test is absurd, and can be very costly if a purchase decision is made on bad information that the seller knows the buyer 'wanted' to hear! On late model cars ... condition of the head studs is easier to check, and actually more important to know than some 'questionable' leakdown numbers!

Uniform compression and 24 good head studs gives me more confidence in an engine's health ... than some vague leakdown numbers!

widebody911 · widebody911's Garage

I usually get a compression test and leakdown when I have the car in for any major servicing - I like to keep track of these things. In my 2.7L days, I took it to a shop in Granite Bay (by Roseville). They spent 2 weeks *****ing with the car, racking up a tidy bill replacing tons of stuff chasing down a rough idle problem. I'd also asked them to do the compression checks, but when I got my paperwork back, the data wasn't there. I asked about it, and he says "It was 160 across the board - I forgot to write it down" Riiiight. The chances of a 150k+ mile engine that has the same compression #'s in all cylinders is right up there with me getting a call from the Vatican next week 'cause I've been nominated for pope. More importantly, I had 2 sets of data that were significantly lower than 160#. The guy was full of *****, charged me for work he didn't actually do, then lied about it. Needless to say I won't go back there, even though it's more conveniently located to my work.

So, that being said, it's much more helpful if you can establish a history, rather than a single sample over which there could be some conflict of interest.

boyo · 08-14-2002, 02:02 PM

Thom - is that the shop on Douglas? Good to know, I've taken a Boxster there and they did good work, but I haven't/don't dare take the 911 there. The shop used to be run by a Porsche factory trained (in the 60's!) guy, but I think he's retired and his son is running the show now. He seems about as slow as 85W oil, but he wasn't the guy working on my car so I didn't worry.

-Boyo

Tinker · 08-14-2002, 07:18 PM

I ve got to agree with BA and some of the prior posts. You can not rely on one test telling you everything. I have never had good luck with leak down tests on air cooled motors. I tend to focus more on a compression test, especially how the numbers build/rise during the test. I read an article a while back in Flying magazine that came to the same conclusions. They also stated that a sampling of air cooled aircraft engines with excessive leak down numbers still developed their specified horsepower output numbers despite having poor sealing.

Food for thought.

Tinker