Ray’s annual check-up: How to determine engine health
The modern farm has a myriad of equipment that employs internal combustion engines of many types. Gasoline, diesel and natural gas designs abound. They are essential tools that need to be reliable.
When a problem does crop up it is prudent to get to the cause as quickly and easily as possible so a proper business decision can be made.
All of the tests described here can be employed on any engine but on some diesels, it may be arduous to gain access to what is required — an injector is not as easy to pull out for a compression or leak down test as a spark plug.
The best way for any farmer to keep track of engine health is to perform routine testing and document the results. This way as hours accumulate you can accurately track the internal condition of the engine not only as a confirmation of all being well, but also as a trigger to locate a potential issue before it becomes a major problem. This data is also valuable when it comes time to trade-in or sell the equipment.
The compression test establishes the engine’s ability to pump air and build cylinder pressure in the bore. It checks the sealing of the cylinder under pressure and the piston rings ability to keep the compression gases working against the piston crown instead of escaping into the oil pan. It also is an indicator of the ability of the valve to seal against the seat and make an air tight bore.
Most are familiar with a traditional cranking compression test but there is also a running compression test that is very important.
Leak down test
While the compression test is dynamic (the piston is moving) the leak down has the piston static (not moving). Instead of the piston building pressure in the bore by moving up and down, the leak down test applies shop (compressed) air to the cylinder while determining the amount that exits. The leak down test is read as a percentage of total cylinder pressure inputted. It is a differential test.
The standard protocol is to fill the cylinder with 100 psi of air and then measure the loss. If 94% of the pressure remains the cylinder is said to have 6% leakage.
This test is used to determine the health of a head gasket in the region between two cylinders. There are occasions when either a compression or leak down test does not show a minor leak between the cylinder sealing surfaces.
It is performed by applying 100 psi of air pressure into the bore while placing the hose from a compression or leak down tester in the adjacent cylinder’s spark plug/injector hole with one end in a can of water. If there any sealing issues bubbles will be seen in the water.
To be accurate, it should be done with each cylinder pressurised since there may be an occasion where the gasket may leak only in one direction. Though this condition is rare it must to be understood that the purpose is to diagnose a problem and no assumptions or short cuts can be taken.
There is no pressure reading with the bubble test only a visual confirmation of air in the water in the can or pail.
Vacuum test (spark ignition only)
The vacuum test is very useful for diagnostics. It is very easy to perform since it only needs to be connected to a manifold (full-time) vacuum port on the engine. It can determine late camshaft or ignition timing (low vacuum signal) a burned valve (rapidly shaking needle) or an exhaust restriction (vacuum decreases from idle to 2,500 rpm).
As with any test protocol there needs to be a logical approach and a determination of when to do each type.
The traditional cranking compression test is the first to be called upon for a poor running engine that has no ignition or fuel system problem. If the engine is in such bad shape that a running or idle problem is present you will usually find it with a cranking compression test.
If the results of the cranking compression test are inconclusive a running compression test should be performed.
With this procedure you ground the spark plug wire on the test bore (or remove and plug the injector and fuel line) and take a compression reading with the engine idling.
In some instances, a loose or pinched/cracked ring or a valve that is not sealing properly or is slow to return it may be masked by the slow speed of the engine while cranking.
As with any test procedure error can be introduced if not done correctly or as in many cases sloppily. To perform an accurate cranking compression test on a spark ignition engine the following needs to be done:
- Remove all spark plugs from the engine
- Connect a booster-style charger to the battery to maintain cranking speed
- Pin or hold the throttle wide-open (make sure the choke is open)
- Crank the engine over five pulses of the gauge
There is debate as to whether a cranking compression test should be performed with the engine hot or cold. As the author I do not feel that it makes a difference as long as all of the cylinders are tested with approximately the same temperature. You would not want to start the test with the engine hot and then come back the next day and finish with it cold.
Another trick that some like to do on a cylinder with low compression is to squirt a little oil in the bore and if the reading comes up, the rings/bore are worn. There is value to that in eliminating the valves as a source of sealing. But if you have a leak down tester I believe that would be more accurate than some oil. There is always the possibility the oil does not seal the bore completely.
If the cranking compression test proves inconclusive then do a running compression test. The only stipulation with a running test is that you only idle the engine. You do not want excessive upward force on the piston without an expanding flame on the top. You do not want to check the running compression at 3,500 rpm for example. But idling is no problem.
If you perform a cranking and running compression test and all looks good the next step would be a leak down test. If the cylinder is not sealing properly the engine will lose power and run poorly.
A leak down test should be performed with the piston at TDC for that bore with confirmation that both valves are closed. You simply read the tool and determine the leak rate.
The important thing about a leak down test is that all cylinder leakage does not occur past the rings. If the leak value is high you need to listen for where the air is going.
If the air is entering the oil pan by listening through the dipstick hole then the rings or cylinder wall is the problem. Air escaping from the induction system is an intake valve. If the noise is heard at the exhaust pipe it is an exhaust valve. This is why the leak down test is a compliment to the compression test. It determines where the leakage actually is.
Analysing the results
Good data and a history are very important.
It is important to recognise that in almost every instance you are more concerned with uniformity between the cylinders for any test over the actual number. A piece of farm equipment with a high leak down rate or low compression that is even in all bores can still provide excellent service for many years without any investment. A cylinder that has an extremely different reading is the trigger for a problem.
Thus, it is best to look at the results as a percentage between the worst and the best cylinder. Divide the two to determine the percentage especially when studying either cranking or running compression. Look for a drastic difference.
For the leak down test do not be alarmed if you see 20% to 30% or more on a diesel with a big bore and a heavy forged piston that expands when heated. Keep in mind that neither the leak down nor compression test will determine the condition of the oil ring on the piston.
Depending on the amount of use the equipment sees it would be a good program to perform these tests every 1,000 to 1,500 hours to track wear and tear. Simply record the results along with the date and hour meter reading.
The data can be more valuable than you first realise. If you are contemplating updating one of two pieces of equipment, easily confirming the internal engine health can often make that decision clearer and more informed. The numbers do not lie.