This is a real time-saving tip!

When you are doing a major overhaul, you should pause at the point of head
attachment:

With the head gasket in place and the head torqued down, BUT, before the
top end is complete (before rocker boxes, push rods, overhead cam is
installed) you can do the following cylinder leakage test to prove the
integrity of head gasket and valve seating:

Make a special tool (air injector) from an old spark plug. Gut it and
braze weld a quick release air fitting. Screw this air injector into each
cylinder and apply 20 psi. Crank position is unimportant! With the
exhaust ports, intake ports and the spark plugs open (except the cylinder
under test) you can prove the following:

Head gasket seal. Listen for leakage in adjacent cylinders, and around the
exterior junction. You can inject some penetrating oil it will make a
sizzling noise where the leakage is, if any.

Exhaust valve seating. Listen and actually feel any escaping air at the
exhuast port of tested cylinder.

Intake valve seating. Listen and feel for escaping air at the intake port
of tested cylinder.

Test all cylinders this way, then finish assembly.

This will save a LOT of hours. You can find and fix troubles before you
finish the job and find out the hard way.

This tip has never before been published, anywhere.

Nomen the megaposter

 

That's called a "differential compression". It's been active
in the small aircraft(reciprocating engines) since the 60's
that I've known about, and probably much longer than
even that. The only difference is that you need someone
to hold the propeller on an aircraft. <g>

I never have really understood why this test seemed never
to be used on engines other that aircraft. It makes an
excellent diagnostic tool. I'd never really seen it done as
part of a reassembly, as you mention. Usually, it was
performed to generally gauge an engines operational
condition with regard to "leak down" detection. Actually,
in my experience, the air inlet also shared a pressure
gauge that was used to measure the leak down to that
of a tolerable threshold. And, the pressure was much
greater than 20psi, more like 180-240psi. Since these
aircraft have mandatory rebuild guidelines, these results
are used to measure generally how close the engine is
to a rebuild specification, usually around 1,000-1,200
hours, generically speaking.

Leakage from the breather = ring seat, from the intake =
intake valve(s), and exhaust = exhaust valve(s). I've not
seen this technique used for a head gasket detection,
but the engines I'm describing have no head gaskets
anyway, since they have no removable heads.

Despite the fact that 99% of web references to differen-
tial compression" relates to data compression, I included
a few relative topics. a ctl-f search can locate the refer-
ence(s). I didn't look for any topics where automotive/
motorcycle/other engine types were mentioned, but they
may be there given the extra patience.

Thanks,

Gary

http://www.lycoming.textron.com/mai...prints/general/determineEngineConditions.html
http://www.mooneypilots.com/mapalog/M20SR_prepurchase_inspection.html
http://www.littleflyers.com/Questions/compression/bottomdifferential.htm
http://www.avweb.com/news/columns/188758-1.html

 

Nomen Nescio actually wrote something that might possibly be useful:

This might be useful for somebody installing a Wiseco big bore kit in
an older Suzuki or Kawasaki.(1)

The kit comes with a crappy copper/asbestos composition gasket. Some
people tried solid copper gaskets that would leak. But the original
equipment gaskets are usually better. Problem is, the steel fire ring
of an OEM gasket might hang into the combustion chamber a bit when a
cylinder is overbored 3 millimeters or so.

So far as valve sealing is concerned, this is normally done by filling
the combustion chambers with solvent and watching for leaks out the
intake or exhaust ports, or by filling the ports with solvent and
watching for leaks into the combustion chambers.

(1) Is anybody still doing modest overbores to KZ-1000's and GS-1100's
when they can just go out and buy a 600cc sportbike that has over 100
horsepower??

 

<<Snip>>

So what happens when the air pressure pushes the piston down because it
was somewhere between TDC and BDC?


Later,

Charlie

 

But this is a motorcycle board, and I'd assumed he was talking about
motorcycles without propellers.

 

It's good that readers are examining Nomen Nescio's suggestions with a
very critical eye. Do NOT trust that guy.

In another group, we once saw a helpful hint for cleaning motorcycles
by spraying them with acetone out of a bug sprayer. The same guy later
faked his own accidental death, and the readership of that NG was
actually sympathetic until they figured out that it was the same guy
trolling them.

Nomen Nescio is a troll. In regard to Nomen Nescio's latest "helpful
hint":

If you applied 20 PSI air pressure to one cylinder of the typical I-4
with a cam chain, without regard to piston position, the cam chain
might be ripped free of whatever safety wire you had suspending it
above the cylinder head as the crankshaft turned and you just might
have to take the cylinder head back off to fish the lost cam chain out
of the its tunnel.

If the cam chain should happen to kink itself in the bottom of the
crankcase, you might wind up removing the engine from the frame and
splitting the cases to retrieve it.

 

I'll have to sternly disagree here. 20psi is nothing for most
engines, and although I've never run compressions tests
on the models you describe, I suspect they easily reach
at least 100-120 psi.

Gary

 

The amount of pressure is not the issue. The problem that might occur
is that the crankshaft might rotate and the cam chain might fall into
the tunnel and kink, causing a lot of extra work.

Even professional written manuals are sometimes wrong.

I ran into two days of extra work doing a valve job on my pickup truck
when the Haynes manual was unclear as to what was actual being done
when I installed the home made wooden block that they suggested would
support the cam chain from falling into the
tunnel.

As it turned out, there was a spring loaded cam chain tensioner behind
the water pump cover which pushed all the slack out of the cam chain.
When I tried to reinstall the cam chain sprocket to the end of the cam,
there was no way to get any slack. So I had to remove the radiator and
water pump and the timing cover. But the timing cover could only be
removed by dropping the oil pan. I had to remove the water pump pulley
and the crankshaft damper and change the front oil seal and install all
new gaskets.

And all because the manual writers had *never* done the job before
writing the manual.
They didn't understand what was actual involved in working on that
engine any more than Nomen Nescio knows about *your* specific engine
when he offers his generic tips.

 

On the other hand, there's not much volume in some cylinders, and it
might be pretty easy to accidentally get 100 psi with a compresser
while trying for 20. I've seen people blow fork seaks while trying to
quick "squirt" 10psi in at a gas station.

Thanks for the explanation, Oskar. I'm obviously playing devil's
advocate here and learning something in the process.

What I've learned is:
Don't try this on a plane
If I do try it on my motorcycle, put 20psi in a portable tank and fill
the cylinder from there.

 

Well, I guess I'm misunderstanding.... You've apparently
been there, and I have not. But, my comment(s) were
spoken from the perspective of an operational engine,
not one in the midst of a reassembly process. Perhaps,
that's the discrepancy.

Gary

 

Here's a better time-saving tip: when you see a Nomen Nescio post, delete it
without reading it.

 

While that is one issue, if the valve timing is such that the valves are
open you will never be able to build up any pressure.

Be sure the piston is at the bottom of the stroke before applying air.

20 PSI isn't much. Compression pressures reach at least 75 PSI.

pierce