Follow by Email

Saturday, 21 May 2011

Performing a compression test (Part 2)

My apologies for the longer than expected delay to part 2. Had been really busy with work recently, and still am, but I'll try to update this blog regularly.

In our last post, we stopped after we removed the plugs. The next thing we want to do is to apply some anti-seize on the threaded end of our compression tester hose.
Next we screw it in by hand into the first cylinder. It should go in quite easily if the threads are in good condition.
Next, you want to make sure you screw it in there really snug, so grab the knurled portion on the top of the hose and give it a good firm turn. Here's the knurled portion of the hose that I'm talking about.
Giving it a good turn till it'll go no more...
Next, we connect the hose to the compression gauge. For this model of compression gauge, I need to pull down the brass ring on the hose connected to the gauge side, and then slip it over the hose that connects to the engine.
In it goes, and I give it a tug to make sure it's snug.
Next, get yourself a piece of paper and a pen. Draw a table with two rows and four columns. The four columns will be for recording the compression readings of the four cylinders, one to four.

The two rows are for recording two runs of this test, one for a dry test, one for a wet test. First we'll do the dry test, and we'll talk about the wet test later.

Flick the Run button to "OFF" and then hit the starter. This turns the engine over, but the coils won't fire.

EDITED TO ADD: I made a mistake in my initial test. I forgot one very important step, and that is the open the throttle all the way. It will make a huge difference to the PSI readings, so make sure you fully open the throttle before cranking the engine over.

I generally hold the starter for 5-8 cranks, but no more than 8 cranks. There's a lot of current passing through the cable to the starter when you do this, so you don't want to hold it too long as the heat will fry the cables. After each cycle, wait awhile to let the cables cool down, before you repeat.
When you press the starter button, you'll hear the engine cranking over. Maybe something like this, "eeee........eeee.....eeee....eeee..."

You can tell I'm not to good at sounds, but you get the idea. I listen for 8 of those "eeee" sounds, while watching the needle on the gauge reading. Sometimes, you need to repeat this two or three cycles for the needle to hit its maximum reading. Now do you see why we need that battery charger?

Once the needle doesn't go up anymore, write down the reading in the table I told you about earlier. This goes under the "Dry run" row. The reading for cylinder #1 is 105 PSI.

After you're done, use the air-release valve on the hose to release the compressed air in the system before unplugging the hose.
Next you want to disconnect the hose, and then unscrew the threaded portion from the cylinder. The next test you're going to do is a wet test. To do this, you're going to squirt some clean engine oil into the cylinder before repeating the compression test again.

The purpose of doing this is so that the oil will help seal the piston rings against the cylinder walls. What we're trying to do is to find out where our loss of compression is coming from, either we're losing compression through gases escaping through the piston rings, or gases escaping through the valves because they're not sealing properly.

By squirting the oil into the cylinder, we're trying to eliminate one variable in the equation: gas escaping through the rings. If the readings increase significantly after we add oil, we know the piston rings are not sealing properly. If the readings stay about the same, most probably we're losing compression through the valves.

There are exceptions to this however, and it's important to realize this is not fool-proof. One example is if the gaps in the rings align and allow the gasses to pass through. I've seen this happen in a car engine once, so it's possible.

A more accurate way to troubleshoot where the leak is coming from is to do a leakdown test, but you'll need an air compressor for that.

Back to our wet test. If you have an oil can, squirt 2-3 squirts of oil into the cylinder. Otherwise do what I did. Just use a straw to draw up some engine oil from a bottle...
and release it into the engine.
Repeat the compression test, and take the wet reading, and write it into the table.

After this, just repeat this for all the cylinders. I'm only going to show the pictures and readings for the dry test.

Cylinder #2:
Reading for cylinder #2 is 110 PSI.

Next up is cylinder #3:
Reading for cylinder #3 is 100 PSI.

Finally, cylinder #4.
 Reading for cylinder #4 is 85 PSI. That's a bit lower than the rest.

The readings of the wet test didn't really increase compression by much, except in cylinder #2, which went up about 10 PSI, which isn't really that significant I think. However, they're all below the minimum specification. The allowable range is 142 - 199 PSI, so if these readings are accurate, that means this engine needs a rebuild

Frankly, I'd want to run this test on a bike with a known good engine and see what kind of readings it gives me, just to confirm that the gauge is reading correctly.

Well, that completes our compression test. All that's left to do now is to put all the plugs back and button up everything else.

The use of a torque wrench is highly recommended for this, because the cylinder head is a critical component and it's aluminum, so you don't want to overtighten anything in there.

Mine's a Craftsman. This model is the microtork, and it has worked fine for me so far (about 10 years of occassional use), and has a very nice action. However, avoid their "digitork" range, that one crapped out on me in weeks. It's a complete waste of money.

I torque down each of the plugs to their specified torque setting, and then I'm done.

So give it a try, and see how your engine's doing. Happy wrenching!


  1. In regards to when you said: "I torque down each of the plugs to their specified torque setting, and then I'm done." - I was wondering if you had the exact sequence and torque settings for the 1992 Honda Nighthawk 750. I have the same motorcycle of course and am having a tought time finding those settings. Thanks for your help!

  2. i just did the compression test on my '93 following the Clymer manual, it says in it that the actual values aren't as important as long as the difference between them is within 10% . mine read cyl1-135psi and cyl2 through 4- 139psi. i didnt bother with the wet test since the numbers are within the 10% delta. my understanding is that the wet test is used if you have bad compression to determine if its the piston rings or valve seals. regards, pete.