Leak down test.....feedback please
03-21-2011, 05:16 PM
#1
Instructor
Thread Starter
Join Date: Mar 2007
Location: Albuquerque, NM
Posts: 104
Likes: 0
Received 0 Likes
on
0 Posts
Leak down test.....feedback please
Hello, I'm new to the forum and while I have desired a 911 for more than 30 years I have finally got serious about buying one. I have been lurking around here for a while learning what I can...
I have located a 88 3.2 that I'm interested in. The owner says he does not believe that the car has ever had any top end work done. He has owned the car for 5 years and has good records during that time but prior to is less clear. When he bought the car it had 88K on it now it has 104K.
Leak down test resulted in #1 -4%, #2 -7%, #3- 3%, #4-10%, #5-4% and #6-4%
he also stated that the car is using less that 1 qt of oil per 1000 miles.
My question is this uncommon ( given the valve guide issues) for a car of this mileage to not have had the top end gone through? What are considered acceptable #'s for a leak down test? Total speculation I understand but what might the near future hold for this engine given these numbers?
Thanks to everyone in advance for any insight or guidance
regards,
Mike
I have located a 88 3.2 that I'm interested in. The owner says he does not believe that the car has ever had any top end work done. He has owned the car for 5 years and has good records during that time but prior to is less clear. When he bought the car it had 88K on it now it has 104K.
Leak down test resulted in #1 -4%, #2 -7%, #3- 3%, #4-10%, #5-4% and #6-4%
he also stated that the car is using less that 1 qt of oil per 1000 miles.
My question is this uncommon ( given the valve guide issues) for a car of this mileage to not have had the top end gone through? What are considered acceptable #'s for a leak down test? Total speculation I understand but what might the near future hold for this engine given these numbers?
Thanks to everyone in advance for any insight or guidance
regards,
Mike
03-23-2011, 01:44 PM
#3
Rennlist Member
I know guys that are running on engines that have 80% plus on 1 or more cylinders. Not that this is recommended at all, but I would say for the right price it might be worth the gamble. Sometimes these numbers are nothing more then a stuck valve that you can free up with a tank or 2 of Techron or some fuel cleaner.
GL either way.
GL either way.
03-23-2011, 05:00 PM
#5
Race Car
Leakdown numbers can be eronious. They respond to temperature, even the source of compressed air can throw them off.
leakdown is a great diagnostic tool, but a compression test should be done first. If you have a serious compression issue, then a leakdown can tell you what it is.
I see no problems with the figures stated, and would think that you have another 150K miles left in that car if all is taken care of.
leakdown is a great diagnostic tool, but a compression test should be done first. If you have a serious compression issue, then a leakdown can tell you what it is.
I see no problems with the figures stated, and would think that you have another 150K miles left in that car if all is taken care of.
03-23-2011, 05:14 PM
#6
Drifting
Thanks whalebird. I see so much misinformation posted when it comes to compression and leakdown numbers. Ideally - you want both your compression and leak down to be consistent. As you said - compression usually will tell you better the health of the engine - leak down will tell you where the problem is. (at least where to look) Cold leakdown vs warm leakdown can show dramatically different results in Aluminum engines.
I always cringe when people post "zero leakdown" - are they sure the equipment was hooked up?
I always cringe when people post "zero leakdown" - are they sure the equipment was hooked up?
03-23-2011, 05:21 PM
#7
Rennlist Member
Compression: 1) 180 2) 190 3) 180 4) 180 5) 190 6) 185
Leak down: 1) 4% 2) 4% 3) 4% 4) 2% 5) 3% 6) 3%
The air-cooled master tech told me that you better grab this car before the word gets out. REst as they say was history
Trending Topics
03-23-2011, 05:52 PM
#8
Team Owner
the first thing you shouid do after leak down test .. is a leakdown test ... if the numbers are consistent you are doing pretty good .. I think those numbers are fine .. and 100K is nothing on these engines .. I think that even has a G-50 in it . if the price is right grab it ..
03-23-2011, 06:21 PM
#9
Rennlist Member
At 100K, there's at most 50-60K left on the guides. And likely not that if the engine will ever see severe use. There's no getting around that. Why do guys forget the amount heat carried away from the valves through the guides is huge?
03-23-2011, 07:04 PM
#11
Race Car
I've seen 3.2s go for well over 250k miles with the motors factory original. No it's not the norm, and those cars saw tremendous highway use. My comment was made in light of (lack of) any additional insights as to the cars condition/usage/maintenance etc. I would hesitate to predict a valve guide meltdown, but drive it until it needs it. Could be sooner could be later.
w00ts post shows how comp/leak figures can diverge. What he didn't post was the cumulative compression figures. A successful comp test will have multiple figures per cylinder as it "inflates", not just the max pressure achieved. In short, as the piston cycles the compression stroke successively, it may achieve max after 3 compression strokes, where it's first was 95. other cylinders may achieve the max (same)pressure but after 5 strokes and started at 70 psi. A consistent trend across all cylinders, as well as each bank of cylinders is the optimum.
w00ts post shows how comp/leak figures can diverge. What he didn't post was the cumulative compression figures. A successful comp test will have multiple figures per cylinder as it "inflates", not just the max pressure achieved. In short, as the piston cycles the compression stroke successively, it may achieve max after 3 compression strokes, where it's first was 95. other cylinders may achieve the max (same)pressure but after 5 strokes and started at 70 psi. A consistent trend across all cylinders, as well as each bank of cylinders is the optimum.
03-23-2011, 07:29 PM
#12
Rennlist Member
I've seen 3.2s go for well over 250k miles with the motors factory original. No it's not the norm, and those cars saw tremendous highway use. My comment was made in light of (lack of) any additional insights as to the cars condition/usage/maintenance etc. I would hesitate to predict a valve guide meltdown, but drive it until it needs it. Could be sooner could be later.
w00ts post shows how comp/leak figures can diverge. What he didn't post was the cumulative compression figures. A successful comp test will have multiple figures per cylinder as it "inflates", not just the max pressure achieved. In short, as the piston cycles the compression stroke successively, it may achieve max after 3 compression strokes, where it's first was 95. other cylinders may achieve the max (same)pressure but after 5 strokes and started at 70 psi. A consistent trend across all cylinders, as well as each bank of cylinders is the optimum.
w00ts post shows how comp/leak figures can diverge. What he didn't post was the cumulative compression figures. A successful comp test will have multiple figures per cylinder as it "inflates", not just the max pressure achieved. In short, as the piston cycles the compression stroke successively, it may achieve max after 3 compression strokes, where it's first was 95. other cylinders may achieve the max (same)pressure but after 5 strokes and started at 70 psi. A consistent trend across all cylinders, as well as each bank of cylinders is the optimum.
I have posted a number of times about how a leak down test can be done wrong, and the one 10% cylinder on this car might be a perfect example. If all six plugs are removed, and then the test is performed, there is a very real chance that a tiny piece of carbon/debris will fall off the tip of a spark plug, and land on the valve seat of an open exhaust valve.
During the test the engine is manually placed on a cylinder to be tested, but in many cases if debris gets caught between the valve and seat, the valve spring tension might not be sufficient to crush the debris. The result can be a reading of 10% (the max allowable) or worse. A quick drive around the block, and a re-test of that cylinder, might produce a 4% number.
Basically, there's a right way and a wrong way to do just about everything. Compression tests can be incomplete if not done by two people with the "inflation" rate recorded, and a leak down test can produce erroneous numbers simply by removing a spark plug while the engine is on a different cylinder.
So, with the OPs car, 10% is probably not a bad number, but for peace of mind it should be re-checked after the car is driven a few miles. Unfortunately, regarding oil consumption, the honesty of the seller is about all that a buyer has.
03-23-2011, 09:37 PM
#13
Rennlist Member
You're probably good, as long as your numbers aren't like these-this is known as the "death formula":
Compression: 1) 180 2) 190 3) 180 4) 180 5) 190 6) 185
Leak down: 1) 4% 2) 4% 3) 4% 4) 2% 5) 3% 6) 3%
Compression: 1) 180 2) 190 3) 180 4) 180 5) 190 6) 185
Leak down: 1) 4% 2) 4% 3) 4% 4) 2% 5) 3% 6) 3%
03-23-2011, 10:22 PM
#14
Pro
Join Date: Dec 2007
Posts: 615
Likes: 0
Received 0 Likes
on
0 Posts
Hello Peter, good to hear the Masters Voice.
There are several ways to approach compression readings. The inflation concept is something I have not paid a lot of attention to and I am happy to be aware.
I have liked both static and dynamic readings, while the engine is hot and cold, for elderly aircraft engines that I was going to fly behind. It takes longer but sometimes gives variations that may need attention.
I think a cold static reading of either 80 or 100 psi at tdc is the best tattle tale test for valves and rings if that is all you are going to do. It is handy because if you get a hiss you know there is a leak. Exhaust hiss is exhaust valve on that cylinder, intake hiss is intake valve that cylinder, oil fill up hole or dip stick hole hiss is rings. The only bad thing about static testing is keeping the engine from turning through when a cylinder is compressed. Static, cold, also will produce the worst reading of all the test variation.
Peter mentions a carbon chip or debris between the valve and seat to alter what would be a true reading. Not recommended for P engines but used in aircraft (and I have done it on the SC with success—but not recommending it) is Staking. This is a brass hammer tapped on the valve stem or rocker to let the valve get off the seat and snap back onto it. This technique is more wisely used on your Briggs lawn mower or Chevy/Jeep straight 6, cause it usually takes 2 or 3 whacks.
There are several ways to approach compression readings. The inflation concept is something I have not paid a lot of attention to and I am happy to be aware.
I have liked both static and dynamic readings, while the engine is hot and cold, for elderly aircraft engines that I was going to fly behind. It takes longer but sometimes gives variations that may need attention.
I think a cold static reading of either 80 or 100 psi at tdc is the best tattle tale test for valves and rings if that is all you are going to do. It is handy because if you get a hiss you know there is a leak. Exhaust hiss is exhaust valve on that cylinder, intake hiss is intake valve that cylinder, oil fill up hole or dip stick hole hiss is rings. The only bad thing about static testing is keeping the engine from turning through when a cylinder is compressed. Static, cold, also will produce the worst reading of all the test variation.
Peter mentions a carbon chip or debris between the valve and seat to alter what would be a true reading. Not recommended for P engines but used in aircraft (and I have done it on the SC with success—but not recommending it) is Staking. This is a brass hammer tapped on the valve stem or rocker to let the valve get off the seat and snap back onto it. This technique is more wisely used on your Briggs lawn mower or Chevy/Jeep straight 6, cause it usually takes 2 or 3 whacks.
03-23-2011, 11:01 PM
#15
Rennlist Member
Hello Peter, good to hear the Masters Voice.
There are several ways to approach compression readings. The inflation concept is something I have not paid a lot of attention to and I am happy to be aware.
I have liked both static and dynamic readings, while the engine is hot and cold, for elderly aircraft engines that I was going to fly behind. It takes longer but sometimes gives variations that may need attention.
I think a cold static reading of either 80 or 100 psi at tdc is the best tattle tale test for valves and rings if that is all you are going to do. It is handy because if you get a hiss you know there is a leak. Exhaust hiss is exhaust valve on that cylinder, intake hiss is intake valve that cylinder, oil fill up hole or dip stick hole hiss is rings. The only bad thing about static testing is keeping the engine from turning through when a cylinder is compressed. Static, cold, also will produce the worst reading of all the test variation.
Peter mentions a carbon chip or debris between the valve and seat to alter what would be a true reading. Not recommended for P engines but used in aircraft (and I have done it on the SC with success—but not recommending it) is Staking. This is a brass hammer tapped on the valve stem or rocker to let the valve get off the seat and snap back onto it. This technique is more wisely used on your Briggs lawn mower or Chevy/Jeep straight 6, cause it usually takes 2 or 3 whacks.
There are several ways to approach compression readings. The inflation concept is something I have not paid a lot of attention to and I am happy to be aware.
I have liked both static and dynamic readings, while the engine is hot and cold, for elderly aircraft engines that I was going to fly behind. It takes longer but sometimes gives variations that may need attention.
I think a cold static reading of either 80 or 100 psi at tdc is the best tattle tale test for valves and rings if that is all you are going to do. It is handy because if you get a hiss you know there is a leak. Exhaust hiss is exhaust valve on that cylinder, intake hiss is intake valve that cylinder, oil fill up hole or dip stick hole hiss is rings. The only bad thing about static testing is keeping the engine from turning through when a cylinder is compressed. Static, cold, also will produce the worst reading of all the test variation.
Peter mentions a carbon chip or debris between the valve and seat to alter what would be a true reading. Not recommended for P engines but used in aircraft (and I have done it on the SC with success—but not recommending it) is Staking. This is a brass hammer tapped on the valve stem or rocker to let the valve get off the seat and snap back onto it. This technique is more wisely used on your Briggs lawn mower or Chevy/Jeep straight 6, cause it usually takes 2 or 3 whacks.
Briggs lawn mower...LOL!!!