Engine people: What is this telling me?

Compression: 68, 78, 78, 78

It tells me that you haven’t measured the other four cylinders! :smiley:

Just kidding. Those are pretty low numbers. Did the engine sit for a while? The rings might be stuck. Can you get a leakdown tester to check things out? Personally, I trust them more.

Dumb question, but all of the plugs were removed and the throttle was kept at WFO for the test, right?

Not my engine, but mid-70s are normal for this type. Bore: 5.125", stroke: 3.875", displacement: 319.8 cubic inches, generally runs around 2,400 rpm. I’ve noticed that other engines I’ve happened to see have three cylinders with the same or very close numbers, and one cylinder that is lower.

Snark! I just spewed Diet Coke all over the keyboard. That reply really made my night. Thanks for the laugh!

I’m glad to see that you’re still working on British cars.

It’s been way too long since I was a mechanic. Barring any other issues I wouldn’t worry too much about it. All of the readings are within 10% of the average.

Those numbers are too low for the engine to even run much. Do the test after putting a few drops of oil in the cylinder. If it’s still very low, then you have leaky valves.

Lycoming engine (you could just tell us what it is instead of being dodgy?) what compression ratio?

Well, so much for the British car idea (and at 320 cu in, I really wasn’t paying attention!) I still say that it’s not too abnormal. The rings might be a little stuck, or something like that. What’s the whole story? If you’re buying, then would this be a good time to get an A&P involved?

No, I’m not buying. It’s just that this is something I’ve noticed and was curious about, and thought it would apply to all multi-cylinder engines. Or is there something about boxer engines that makes one cylinder have lower compression than the other three or five, and why?

Leakdown test matters so much more…

no, they should all be consistent. at least within 10%. if one is low, it could be something like a head gasket leak or a worn/burned valve.

I have doubts we’ll find a definitive answer to this one. There are many factors that affect compression readings, and compiling all the relevant data from a sufficiently large sample of various engines would be a daunting task.

It’s even possible that the variance in readings is not rooted in the mechanical condition of the engine, but rather in the testing method. For example, in the first cylinder tested the piston has usually just sat there before measuring compression, whereas all the subsequent cylinders have moved up and down during the testing of the first one. That can distribute oil from the rings in a way that didn’t occur in the first one, possibly resulting in a small change in the rings’ sealing properties.

As for the engine itself, one very specific example is the old air-cooled VW Beetle engines. It became standard practice to adjust the valves of cylinder #3 with a couple thousandths more clearance than the other three cylinders. This was because #3 ran hotter than the others due to some of its cooling airflow being blocked by the oil cooler. Extra valve clearance resulted in more time for the valves to seat between each opening, allowing more heat transfer from the valves to the cylinder head and thus reducing the likelihood of burning a valve. The extra valve clearance and lessened open time for the valves also causes less air to be inducted, which could lower the compression reading somewhat.

I also wonder if confirmation bias comes into play. Occasionally I would expect variations like one cylinder reading a bit higher than the other three, or two reading higher than the other two. If the “one lower than the other three” instances somehow caught your attention, the other scenarios may not have stuck in your mind so much.

I’d wondered if this was something to do with air-cooled engines, like on VWs and older Porsches. This is just something that caught my eye, and I haven’t been looking at any liquid-cooled engines.