View Single Post

Grady Clay
Grady Clay is offline
Registered
 
Grady Clay's Avatar
 
Join Date: Dec 2003
Location: Arapahoe County, Colorado, USA
Posts: 9,032
Hey Dude (no pun intended),

Kurt and Wayne are right on with their advice.

A Cylinder Leakage Tester is one of the best diagnostics for the mechanical performance of an engine. A cranking compression gauge (one that screws into the sparkplug hole) is useful also but tells you different things and should be used as secondary to a leak tester. The other really good instrument is a bore scope that fits the plug hole and will fit past the open valves.

Some cylinder leak test basics:

The test is done with low pressure compressed air, say 5 PSI, and the tool measures the small amount of air flow. The tool has a regulator that brings shop air pressure (70-200PSI) down for safe use.

The test is done on each cylinder in firing order sequence with that piston at TDC compression. I take that plug out only when the piston is in place as some carbon on the plug could get under the exhaust valve and prevent an accurate measurement. The piston has to be at exactly TDC as the compressed air can turn the engine. Don’t leave a wrench on the crank bolt and be prepared for the engine to inadvertently spin.

The beauty of this test is that it both tells you if there is a problem and, if so, where the problem is. My gauge reads from 0% to 80% air flow. I disconnect the gauge from the engine and set the regulator to 0%. I then reconnect to the engine (it has a homemade screw-in-the-head quick disconnect to fit 911) and take a measurement. I disconnect and re-connect to take several measurements and to insure the “zero” hasn’t changed.

On my particular instrument, a normal, perfect running, best-you-can-get reading is about 1%. Most read 1-2%. That is all past the rings (you put your ear to the oil filler.) If I get readings approaching 4% past the rings, I put the plugs in and drive on the highway for a while and then re-test. If the numbers come back down, to say 3% or less, I know it need to be driven more.
If there is any detectable leakage past the valves (you put your ear to the intake and exhaust), again go drive the car on the highway and re-test. If there is still leakage past a valve then measure the valve clearance next. If it has opened up since your last valve adjustment or is much looser that the other valves, the next place to look is something between the valve and the seat or UGH a bent valve. I have had skilled mechanics tell me that when they get to this point they hammer on the rocker adjusting screw while pressure is on the cylinder. I don’t like that approach but some seem to think it works.
If there is any leakage past the head gasket, stop and torque the head nuts. If you find one (some) that turn to take torque, note which ones, and test drive, and re-torque. The critical issue here is they need to hold torque, particularly on mag-case engines.

Some put shop air to a cylinder. I don’t think that is ever a good idea. A 100mm cylinder and 125 PSI air is putting 484 pounds force on the static piston and trying to lift the head by the same amount or more. You do the math.

FYI, I have two archaic (circa ‘60s) Sun Electric Corporation, Model 228, Cylinder Leakage Testers. They work great, are repeatable and agree with each other within their resolution (about 0.2%). I went on line and tried to find the company to no avail. I’m sure they have long since been bought out.

What are the current good testers?

Best,
Grady
__________________
ANSWER PRICE LIST (as seen in someone's shop)
Answers - - - - - - - - - - - - - - - - $0.75
Answers (requiring thought) - - - - $1.25
Answers (correct) - - - - - - - - - - $12.50
Old 01-20-2004, 12:22 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #5 (permalink)