compression test procedures

[Regal2800]

here is a question for you guys; The detroit manual specifically states that the engine should be running while conducting compression tests on each cylinder. However it seems to be industry standard to conduct compression testing on diesel engines without the engine running. (just cranking) It would seem that a running engine is the most accurate way to conduct compression testing as detroit diesel recommends. Can anyone explain why most other diesels are tested WITHOUT running the engine?

[MikeP]

"Can anyone explain why most other diesels are tested WITHOUT running the engine?"

It's easier!

I have been told that some mechs do a "standard" comp check on DDs, just like you would do on a car, using the starter to spin the motor. I would expect that the numbers might end up different based on which comp check you did - engine running or starter only but I've never done one with the starter so I can't say that for sure. Since DDs specs in the service manual are based on running the engine, unless there is a reliable conversion chart in existence, the numbers could be misleading.

Obviously the thing to do would be to do one with the starter motor and do that cyl again with the engine running. If they come out the same then I'd say it doesn't matter. For that matter, even if they don't, you could then come up with a "correction" factor and do the rest with the starter if you wanted to.

[SeaEric]

They need to be warm and running for a proper compression test result.

[Boatsb]

"Can anyone explain why most other diesels are tested WITHOUT running the engine?"

It's easier!

I have been told that some mechs do a "standard" comp check on DDs, just like you would do on a car, using the starter to spin the motor. I would expect that the numbers might end up different based on which comp check you did - engine running or starter only but I've never done one with the starter so I can't say that for sure. Since DDs specs in the service manual are based on running the engine, unless there is a reliable conversion chart in existence, the numbers could be misleading.

Obviously the thing to do would be to do one with the starter motor and do that cyl again with the engine running. If they come out the same then I'd say it doesn't matter. For that matter, even if they don't, you could then come up with a "correction" factor and do the rest with the starter if you wanted to.

We got the start of something here.

after mapping out all the cylinders we could average the correction factor and test other engines to see if the average correction factor is constant. Compare 53, 71, 92 and other series and determine if it's across the board or specific to each series. Also NA vs turbo.

I wonder if intercooler or after coolers will effect the results?

Someone needs to start a spreadsheet so we can start gathering get the first 1000 samples.

[Regal2800]

ya, i have always done my compression tests with a running engine as well. But now i have to do on my 7.3L powerstroke and everyone is giving my numbers and procedures that require cranking the engine only. I guess it doesn't matter either way so long as the specs you have specify whether it is cranking or running. I was just curious as a running engine seems to be the most accurate.

[Finalee]

We got the start of something here.

after mapping out all the cylinders we could average the correction factor and test other engines to see if the average correction factor is constant. Compare 53, 71, 92 and other series and determine if it's across the board or specific to each series. Also NA vs turbo.

I wonder if intercooler or after coolers will effect the results?

Someone needs to start a spreadsheet so we can start gathering get the first 1000 samples.

That's absurd. 500 samples is enough.

[krush]

That's absurd. 500 samples is enough.

Really? I think 5,000 is required to create a valid normal distribution curve with fair uncertainty.

[Boatsb]

ya, i have always done my compression tests with a running engine as well. But now i have to do on my 7.3L powerstroke and everyone is giving my numbers and procedures that require cranking the engine only. I guess it doesn't matter either way so long as the specs you have specify whether it is cranking or running. I was just curious as a running engine seems to be the most accurate.

Why do you need to check compression on the 7.3?

They run forever.

[captddis]

For one thing is the cylinders are warm and the metal has expanded. Doing it cold will give a lower reading. The DD compression acceptable limits are taken at 600 rpm. The liners are very sensitive to temp. About 10 years ago we were fitting liners on a 71. Had them all marked and went to double check . Number 5 and 6 almost slid in on their own. We looked at each other wondering what happened as we had a tight fit before. We realized that those two liners were in the air cond stream. Let them sit off to the side and they fit tight again.
Most tests regardless of engine say to check comp with a warm engine.

[Bob Bradley]

Why do you need to check compression on the 7.3?

They run forever.

Roger that. 300k and going strong!

Truck good shot.jpg