Compression test and leak down test

FYI: Your readings can be inaccurate if the car isn't regularly driven.

 

Don't forget consensus seems to be that your readings will not be absolutely comparable to anyone else's as they only represent the tester's procedure at the atmospheric conditions on that day - with luck however the spread will give hints as to overall well-being and then repeating the test on another day in the future with similar atmospherics by the same person using the same procedure may give a comparable set

 

For Compression Check: Should be done with hot engine.
Put car in Neutral
Remove all spark plugs
Disconnect Ignition blocks, and carefully put the spark plugs wires out of the way.
Remove Fuel Pump fuse at the foot panel of the passenger foot well.
Connect a battery charger to the 12v-Grd ports
Connect the compression gauge to cylinder 1
Depress the throttle all the way, crank engine over 8-10 cycles, stop.
Read compression readings.

 

WSM says to remove the ECU relays before doing compression test. Would also apply to leak down.

 

Move on to Leak Down test, should also be done warm or hot for most relevant results.
Find TDC compression stroke. Use a Remote Start device that connects the 12V to Starter terminal. Use a screw driver or a rod to confirm that the piston is at the top.

Connect cylinder to leak down gauge.
Apply 100 psi pressure to one side of the gauge. Sometimes this will cause the cylinder to slip.
Read the resulting pressure on the other side of the gauge.
The difference is the leak down percentage.

 

I've read that 100 PSI is too much for it can cause the valves to be forced closed and not provide an accurate reading...

 

Regarding leakdown tests for these motors, I have found with 355 motors it makes almost zero difference if the engine is hot. I could only test one cylinder when it was hot for this particular 355, but the difference was not even 1% compared to cold.

I'm reasonably certain that the piston design and piston materials has a lot to do with this. The 355 stock piston does not really have much taper if any compared to say a JE forged piston made of 2618 which has a visually smaller crown area compared to the skirt (but expands half a thou or so more than 4032 when hot). I can only assume the alloy used for stock pistons runs a tighter static clearance as it doesn't expand as much therefore yielding negligible difference in hot vs cold leak tests.

 

Nonsense. IMO, the higher the pressure the better as the idea is to test real world cylinder combustion conditions which are of course far greater than just 100psi. Take each tester individually though, they can be different.

Low psi leakdown testers are more sensitive and their results shouldn't be compared the same as a 100psi tester, and even then different testers can be off from one another. The industry standard for a 100psi tester is dictated by the size of the metering hole between the two guages, a 100psi tester uses a small hole (.040-.045 is standard), while the lower psi tester uses a larger hole (haven't measured any except the cheapo HF unit which is .080) which makes it more sensitive. A great example was the previous 308 rally engine I built with P6 cams (which was awesome btw, but a total 'on off' switch), the high psi Mac tools tester showed 2.5% leak while the HF low pressure tester showed almost 10%.

 

Im sure this is a dumb question.... but someone please educate me a little.

Will a car that is not driven that often show higher leakdown #s than one that is driven weekly?
If so, will driving the car, etc, improve the leakdown #s over time?

I'm just curious as I've heard both yes and no answers from multiple sources...

Example.. A friend of mine just had his major, all his compression #s are the same across all cylinders... roughly 195.
but he had a slightly higher leakdown on cylinder 6 than the others. He generally got 2 different pieces of advice...
but Im not sure which is more legit??

Tech A) - if only 1 or 2 cyl show a higher leakdown than the others, it could be a build up of carbon deposits on the valves that can be cleaned out with a good drive or carbon cleaner, etc... and that leak down numbers of of 10% to 20% are not really that big of a deal considering the age of the car (some are almost 20 years old)

Tech B) - if only 1 or 2 cyl show a higher leakdown than the others, it could very well mean a valve job in the future, but the engine is still strong and very usable. Keep an eye on the compression #s and if they start to slide on those specific cylinders, then make a move to get the top end done....


any education you guys might offer is greatly appreciated.

Thanks gentlemen

 

I look forward to hearing their chatter. I've heard the various arguments before, but there are only so many ways an extremely hot combustion under very high Psi can be tested, and low psi is not my personal favorite, but it works. For instance, a low psi tester can completely lose track of a head gasket failure that a high psi tester would catch and show bubbling through the expansion tank with.

If a tech really wants to get suspicious of a cylinder and they find themselves scratching around with various figures yielded from various leak testers but they know where the air is coming from (valves, or rings etc), then they should insert a vacuum tester and see how many inches of Hg the cylinder pulls. I use a vacuum tester I normally test valve sealing with after rebuilding a cylinder head that has a leakdown hose only attachment which secures to a vacuum pump and gauge.

Regardless, if leakage is much greater than other cylinders then there's a problem.

 

Go figure.... the place that said there would be a valve job needed in the future was an independant Ferrari shop. The dealer told him those numbers were respectable and that a good "italian tune up" would improve the #s on the bad cylinder.

I know when I did my major, I had compression #s of 198 across the board...
and roughly 5% leakdown on 7 of the 8... one was closer to 7-8%. It doesnt seem like 15% leakdown on a single cyl would be that troubling considering the age of the engine, etc, but I am ignorant and uneducated when it comes to this stuff....

 

Love Mario!
Please tell Dave and Alex that I said hello the next time you see them! Great family

 

The difference in compression readings alone would be enough to do me in on it. What were the differences in the compression numbers?

Also, that tester he's using is an incredibly sensitive unit, I can't tell if it's the old Snap-on tester I'm familiar with but it sure looks like it. If you breath on them wrong they will show higher leak, not all leak testers are the same and their results should be taken individually.

I'm familiar with his argument that a high psi testing will close the potential errant valve, but also remember that high psi is precisely what the engine is dealing with when it's running. At no point does that engine ever see low pressures when running, so why would it be tested as such.

An optimal PPI for me specific to an engine consists of 3 tests: dyno bhp, even compression readings, and even leakdown. If a 355 engine makes 310bhp at the wheels (very strong) but has 15-20% leakdown using a very low pressure gauge but good leak with a high pressure gauge and has good even compression, I would not in the slightest be concerned.

 

I should add there is no universally accepted input pressure for automotive leakdown testing, but the FAA has established 80 psi input pressure as their standard for leakdown testing on piston aircraft engines and they allow up to 25% leakdown in those aircraft engines.

 

On motors with sprung valves the spring pressure is much higher than 100 PSI. so air pressure isn't going to close them any more.

On early desmodromic motors that had no springs the air pressure developed was the only thing closing the valves for the last half mm or so. They were a little hard to start and would just roll away if you didn't set the parking brake but that was the only difference an air closed valve made.

But to think 40 or 50 lbs is going to seal and otherwise bad valve is just silly. I have tested a lot of motors both ways and never seen a difference in valve leakage. I used to have a low pressure tester and a high pressure tester and used them interchangeably. Never saw a difference in valve leakage.

One area where pressure will make a difference is on slightly sticky rings. But guess what, rings are pressure activated. If they are getting slightly sticky and it takes another 50 lbs to unstick them it's not bad to know but at this point is making no difference in it's current running quality. Good oil, frequent changes and an Italian tune up will fix the rings.


Testing at 40 lbs is great for people who want to aggressively sell major engine work. Thats why I gave it up a long time ago. Lots of false negatives that made zero difference to the health of the car.


As far as acceptable leak down there is no one single acceptable number. Different engines are different and seldom is anyone going to tell you, you just need to learn it yourself from experience. From experience we learned the ring, cylinder package in the 355 was different from all previous Ferrari's.

The old 4 valve motors it was not uncommon to see 20% or more on motors that ran perfectly and no oil consumption, good compression and no symptoms.

The 355 motors routinely have 0-1% on good cylinders(never see that on a 4 valve). Their ring design allows ring gaps to be .1mm. Most motors .3-.4 on a brand new motor is more the norm. The 355 normally have very little wear but once the cylinder walls and rings start to wear they go very quickly. When they get to 10% leakage it is a danger sign. A dead cylinder is not far off.

Whoever said they test the same hot or cold was correct. Mahle must have matched cylinder expansion and ring expansion for the gaps to be that tight so it doesn't change when hot.


Rules of thumb are fine but there are a ton of exceptions to most. Judging a modern Ferrari motor by the rules we applied to a 67 Ford truck just isn't correct or smart. Lots of things have changed.

 

100 PSI air into a 3.35" bore will each a force of 881 pounds. Force = Pressure x Area.

Force = 100PSI x 8.809 SQ Inches

Force = 880.9

From the workshop manual, the valve force at the static length = 201 Newtons which = 45.18 pounds.

I'd think 881 pounds will offset 45 pounds with a net result of 836.

 

I understand.

It doesn't change ****.

and a 355 valve is a little smaller than 3.35 so the direct pressure is a little less.

Like I said, it don't change ****.

 

That is true, I did not consider the valve diameter.