Leak Down Test
- Thread starter CarlG
- Start date
BILL WILLIAMSON
Well-known member
Carl, I "THINK" I remember that the rule of thumb is 10% loss. Uniformity is important, also. Bill W.
Jim Brierley
Well-known member
Do it with a warm engine for best results. 10% is very good. The main thing is they should be even straight across. Listen with air in the cylinder, air escaping out the exhaust pipe is from bad exhaust valves, from the oil tube is rings, etc.
oldmanandakid
Senior Member
Here is a link to the way the tractor guys do it.......http://forums.yesterdaystractors.com/viewtopic.php?t=993063
Chuck Sea/Tac
Well-known member
I made an adopter fitting to screw into a spark plug hole, but I could not stop the engine from moving from the pressure. I had it in gear and top dead center as best I could. So how do you keep the piston from moving??
Chuck
Chuck
BILL WILLIAMSON
Well-known member
Chuck, it has to be exactly on TDC. Use timing pin to get # 1 set. Then turn hand crank EXACTLY 1/2 turn & test # 2, anothher 1/2 turn & test # 4, another 1/2 turn & test # 3. HAPPY CRANKIN' Bill W.
BILL WILLIAMSON
Well-known member
Dr, J, is the "J" for "Jest"? Dog
here. Great word play on "WEAK"-test! (HO HO!) What's your leak down percentage? Ole' Bill calls his extra tall white thingy a "THUNDER-MUG" Asked him why? He mumbled something like, "Well, a loooong time ago----------in a Model A--------wasn't tied----------"
??? Buster T.
Bob Johnson
Well-known member
To keep the engine from rotating I set the #1 piston just before top dead center using the hand crank. Then I set put a wedge (short 2X4) under the hand crank to prevent it from moving.
Bob
Bob
socalplanedoc
Senior Member
See, the problem with Wikipedia is the articles are OFTEN wrong. It saddens me to think that people wil accept what they see in wikipedia as gospel without investigating it.
Aviation does not use leak-down testing. In fact, I've never seen anyone use leakdown testing for engine work.
What we use is called Differential Pressure Testing. The wiki even cites the FAA Advisory Circular AC43.13 as a source but the AC has nothing for leakdown, only differential pressure testing. The article is actually describing diffpress testing and calling it leakdown - they're not the same thing. Some of the facts they call out are wrong too..
While diffpres testing provides a clearer picture of engine health than common auto-style compression testing, it's difficult to use an cars largely due to accurately and firmly holding the engine at TDC. Airplanes have a prop you can use as a big-ol (and potentially dangerous) lever to position the crankshaft. Cars need a purpose-made custom tool. Then, there's the question about what numbers constitute good or bad results - not a lot of data for that on car engines. Still, I use it on cars largely for determining the source of a problem like others have pointed out - found water leaking into a cylinder using a diffpress tester even though there were no detectable bubbles in the coolant when the engine was running.
I feel the best thing for compression testing using any method is to strive to do it the same way, with the same tools, everytime you do it. If you're trying to compare your results with some other engine, it's important to use the same tools and same methods on that engine. Even in the aviation biz, I've seen mechanics techniques and results vary when checking the same darn engine. those variables just muck up the whole thing.
I agree with what others have said about the importance of being even across all cylinders, even if the actually number seems low. I mean, if I'm chekcing and engine and getting low numbers BUT they're all the same and it's runs OK, no way am I going to pull it apart
MikeK
Well-known member
That's why I cited the article. It explains why you can't compare leakdown % between different model and displacement engines even if you use the same leakdown setup! I hear guys say "My V8 has 6% but my 4 banger is 10%" Irrelevant, apples vs. oranges. As for the article's aviation correctness, I quit worrying about all that when I sold my Cessna 150 aerobat.
Socal, you know you could use your expertise and submit edits and corrections to Wikipedia. That's what it's all about. Thanks for reading it! I'm not so sure anyone else read it completely.
CarlG
Well-known member
This was my reason for asking my question in the first place, Thought that a Model A person might have some knowledge on using this type test on a Model A engine.... It explains why you can't compare leakdown % between different model and displacement engines even if you use the same leakdown setup! ...
And I did read it completely, it just didn't answer my curiosity.
BILL WILLIAMSON
Well-known member
Carl, fun question to ask even an experienced mechanic, "How many crankshaft revoultions does it take to fire ALL 12 cylinders on a Lincoln Zephyer"?
Most will run out of "COUNTING FINGERS" tryin' to "COMPUTE" it????------(One guy said 17!) Bill W.
Most will run out of "COUNTING FINGERS" tryin' to "COMPUTE" it????------(One guy said 17!) Bill W.
MikeK
Well-known member
Carl,
I will give you some of my own numbers, but they are certainly not any standard to judge by.
Using a setup with the correct 0.040 dia x 0.250 long x 60 degree entrance reference orifice and 100psi input:
My bone-stock 40B with 22,000 miles:
Tested after 1 hour cool-down, runs 15-W40 oil
Leakdown: 91, 93, 88, 91. Compression: 62, 62, 56, 60 (stock head)
My 160B with 6,000 miles, modern valves with 3-angle grind seated on inserts, oversize 1.75" intakes,
modern narrow rings with moly filled top,
honed with torque plate, finish plateau honed:
Tested after 1 hour cool-down, runs 20-W50 Mobil 1 synthetic oil
Leakdown: 95, 97, 94, 96. Compression: 160, 162, 160, 158 (117.5cc head, chev pistons)
Engine I ripped down, had metric pistons, mileage & oil type unknown:
Leakdown: 78, 80, 85, 12. Compression: 52, 48, 56, zero
FWIW, to keep the engine from rotating I put the trans in 3rd, hand crank to TDC of a cylinder with car moving forward, then pull the hand brake to lock the drivetrain.
I will give you some of my own numbers, but they are certainly not any standard to judge by.
Using a setup with the correct 0.040 dia x 0.250 long x 60 degree entrance reference orifice and 100psi input:
My bone-stock 40B with 22,000 miles:
Tested after 1 hour cool-down, runs 15-W40 oil
Leakdown: 91, 93, 88, 91. Compression: 62, 62, 56, 60 (stock head)
My 160B with 6,000 miles, modern valves with 3-angle grind seated on inserts, oversize 1.75" intakes,
modern narrow rings with moly filled top,
honed with torque plate, finish plateau honed:
Tested after 1 hour cool-down, runs 20-W50 Mobil 1 synthetic oil
Leakdown: 95, 97, 94, 96. Compression: 160, 162, 160, 158 (117.5cc head, chev pistons)
Engine I ripped down, had metric pistons, mileage & oil type unknown:
Leakdown: 78, 80, 85, 12. Compression: 52, 48, 56, zero
FWIW, to keep the engine from rotating I put the trans in 3rd, hand crank to TDC of a cylinder with car moving forward, then pull the hand brake to lock the drivetrain.
socalplanedoc
Senior Member
That's why I cited the article. It explains why you can't compare leakdown % between different model and displacement engines even if you use the same leakdown setup! I hear guys say "My V8 has 6% but my 4 banger is 10%" Irrelevant, apples vs. oranges. As for the article's aviation correctness, I quit worrying about all that when I sold my Cessna 150 aerobat.
Socal, you know you could use your expertise and submit edits and corrections to Wikipedia. That's what it's all about. Thanks for reading it! I'm not so sure anyone else read it completely.
ya, I tried that once.. I ran up against the "wiki-weenies" who insisted I follow a protocol that's foreign to me... I don't have enough time in a day to become a professional wiki editor.