Testing piston rings with water

I am trying to get every cylinder to leak 20% or less. That's my target. I head up to 45% is ok but besides this 75% that I am trying to fix, none come even close (15-20-25%).

 

That's with water poured into them or leakdown test? One will not be comparative to the other.

 

I've heard different answers on this but the canonical I think is NO -- Toyota and Haynes want you to remove the diff to get the upper oil pan out. Some people here told me they were able to weasel it out without removing the diff by loosening other things but I can't find the post now.

 

Leakdown test, see the picture I posted above. That is scientifically accurate. The water test is relative to other cylinders that are not leaking.

To be super safe, I measured it like 5 times, each time spinning the crank just to make sure I don't have any valves open. I was super OCD about it, laboratory-level cautious

 

Yeah, i meant anything else besides the diff? I would assume the diff should come out if you want to do a good job with the rest of the work.

 

I'm pretty sure the diff is the only obstruction

 

Yeah, i follow all of that. What I am saying is if the cylinders all leak 20% of the water over the same amount of time, how do you know if that is good, bad, ok? If one leak 30% and the others 10%, is the 30% one bad or the others really good? If they all leak everything in 5 seconds, are they all bad? My point is, if you have no reference, it's just an exercise for experimentation sake. Just assuming based on the leakdown test of the other cylinders being good that one will be worse than the other 3? I guess it won't hurt to try if you recover most of the water but water under no pressure is not the same as 80+ psi of air and might not show you anything, or might mislead you. In any case, I wouldn't quantify anything based on the water leak test, just relative to each other.

 

I found online that up to 40% is okay but I don't think that's normative.

 

That is why the water test takes MUCH LONGER -- I give it 24 hrs.

E.g. for the intake valve test, I put 100 ml of water in the manifold neck and put the head on bricks with a plastic dish underneath. Let it sit 24 hrs and the problem cylinder leaked 40% of the water. I am now doing it with the other two cylinders that the leakdown test did not fail to confirm that they are leaking much less water over the same period of time.

I am pretty darn aware that the pumped up air and water do not compare on the same scale -- however they are correlated pretty much 100%. The water test takes much more time (inconvenience)

 

Up to 20% leakdown is acceptable for normal vehicles

1-10% leakdown is acceptable for performance/ racing engines

 

For water leakage or leakdown test with air? Leakdown tests with air on the engines I have experience with, 40% would be poor results. 80/60 is where I usually draw the line.

 

Leakdown with air

 

that's why i am shooting to be around 20 max. I do have a couple that are 25 and some 15 but oh well. I'm not going to drive the thing forever

 

I wouldn't use that to quantify your ring leakage test with water. I don't think they are comparable, especially with a arbitrary time of testing.

Anyway, GL with this project. You may not drive it forever, but it would be nice to have the option!

 

Correct and I am not. I tried to explain that while the air/compressor leakdown test is normative and absolute, the water test is relative to other cylinders HOWEVER the results should be highly correlated.

Like basically, the cylinder that is leaking 75% of pumped up air (100 psi) should also, over a 24 hr period, drain more water than the other cylinders that are leaking 15-20%

 

That would be my dream life -- I have no one to impress with a brand new truck.

However, the reality...

 

Just curious did you even do a compression test prior to tearing apart and if so what where those numbers? What were the initial symptoms?

 

I took it apart for a different reason (wings broke on the left VVT phaser) but then the reason I took the head off was because a cylinder on the left side was leaking 75%. I didn't do compression but I reckon it can't be too good if it leaks 75%. The reason I didn't do it is because I didn't own a compression tester but I think I will now get one

 

You would normally do a compression test, to see if you have good compression...

And a leakdown test, to see how it measures. There should be factory specs.
And if all cylinders are similar to each other.

You either have symptoms and a problem to fix, or you don't.
Problems include oil burning, which can be caused by other things...

rings can be cheap. If you are in doubt, you can just replace them.
There are ring cleaning products to run in the engine that claim to work, but they may or may not.
Or things like Marvel Mystery piston soak, running diesel fuel for oil for a little bit or ATF, and so on. Typically to cure a problem on a junker, not something you try for no reason.
Depends on the factory design. Does the truck have a reputation for poorly designed rings causing problems? Not really.
If you neglect the fuck out of oil changes, that will contribute to coking up the rings with carbon.
As in wrong filter, wrong oil, not changed often, and so on.

typically you have a problem, diagnose the problem, then fix the problem. Re-assemble the engine to run it and move on. Further curiosities are generally for a different day, if you even have the time.
In your case, a broken VVT unit.
Nobody changes a front brake pad and then goes "well while I'm here, I wonder if my rear diff works properly, even though I have no issues. Let me fully disassemble it to look inside."

 

How did you know it was leaking with no compression test? That’s what I’m trying to figure out. If the truck was running fine? If you had no symptoms why yank the head off? Curious?