Topic: Leak down test questions

[denward17]

CB750K 1978

I purchased a cheap Amazon leak down test kit, and the gauge that showed percentage would not go to zero.  Stayed up to around 40% before connecting the hose to the cylinder.  I didn't hear any air leaks so I assume the gauge is faulty.

Needless to say I do not put any trust into this gauge.

The engine is out and cold. I hardly heard any air when connected to #2 or #3.  Just a little on #1, but the air noise increased when connected to #4.

I took the valve cover off and checked the tappet clearance to make sure they were not to tight on any cylinder, all clearances were in the ball park of EX .08mm and IN .05mm.

Is it normal to hear air noise when doing a leak down test?

Can I use the regulator from my air compressor to perform the test and get readings?

Or do I just bite the bullet and buy a better leak down test kit?

Any ideas appreciated.

[bryanj]

A proper leak down test is a fairy complicated thing to do, with connector to cyl open gauge should read 100%, with connector to cylinder blocked off should be zero%.
With cylinder at tdc of compression and leak down kit connected listen for air at inlet port, exhaust port and engine breather, or with cover off, valve gear.
Repeat this for each individual cylinder and compare % of leak on each cylinder, you are more looking for a variation than a specific number.

[BenelliSEI]

The % gauge should not be at 40% unless the end fitting is leaking. Does it zero when not in use? I assume it’s not leaking when connected? What does the leak down gauge read when connected? Even higher?

You could always pop the face of the gauge and often the pointers are just a push on fit. Set to zero with a drop of glue. See how they work now?

[denward17]

So I tried again this morning, and discovered the pressure from the compressor had to be set to 100psi before the % gauge would go to zero.

With this setup I discovered some bad things, #1, #3 and #4 all had readings near 40% leakage.

Number 4 especially had a bad exhaust valve leak, it was near 80/90%, and I could hear / feel air out the exhaust port.

Arghhh, didn't want to pull the head, looks like I need to now.

[Don R]

 If your valves are closed and the rockers were at zero lash it should be a lot better than that. My 836 read 4% on a good gauge but still smokes like a 2 stroke. I like to do a warm engine leakdown test as well as a cold one but that's just me.   

[Flyin900]

If your valves are closed and the rockers were at zero lash it should be a lot better than that. My 836 read 4% on a good gauge but still smokes like a 2 stroke. I like to do a warm engine leakdown test as well as a cold one but that's just me.

Just to confirm. You are setting and testing each cylinder at TDC individually and confirming that there is valve lash play on both valves on that test cylinder before the test? Any other setup will result in incorrect readings.

[bryanj]

Leakdown will not tell you if stem seals are shot or guides worn, only that valve seats so you can get a good reading but still smoke

[denward17]

Yes, testing at TDC on each cylinder and making sure there is play in valve lash.

Now I have the cam out and just retorqued the head to 20 ft lbs.  There is no mention of the 5 6mm bolts under the pucks and on the front of the head near the chain tunnel, do these get torqued as well?

There was movement on about 3 bolts near where the oil leak was coming from.

Going to perform the leak down test again after torqueing.

[BenelliSEI]

Yes, testing at TDC on each cylinder and making sure there is play in valve lash.

Now I have the cam out and just retorqued the head to 20 ft lbs.  There is no mention of the 5 6mm bolts under the pucks and on the front of the head near the chain tunnel, do these get torqued as well?

There was movement on about 3 bolts near where the oil leak was coming from.

Going to perform the leak down test again after torqueing.

Good plan!

[denward17]

Another test completed:

Cam out, head retorqued (except the bolts under the pucks)  Is there a torque value for them?


Number 3 was the best at around 20%, Number 1 and 2 were about 35% and number 4 was just over 50%

Numbers did improve just a little after re-torque.


What to do now, leave it, or take head off for valve job?

[denward17]

more pics; pucks were puckered to the inside for some reason.

[bryanj]

50% is too much, pull it down and inspect it now whilst its out

[denward17]

50% is too much, pull it down and inspect it now whilst its out

That's what I am thinking as well, as much as I hate to admit it.

[Tracksnblades1]

Well done  dw17..

Make sure your readings repeat consistently..

Then follow the other advise…

Now you Know….!

[BenelliSEI]

Do you hear where the leaks are? Should be loud from the crankcase (rings) or the intake or exhaust port (valves). If that engine has not run for ages, the valves that were open will have corrosion on the seats..... pulling the head and giving them a little relapping will probably do it.

[denward17]

Most of the air leaks I heard sounded like they were from the valves.  I could not distinguish if air was coming from crankcase.
I guess I could hook my hoses again and try to listen more carefully.

Would I be stupid if I don't do rings?

[bryanj]

No, but you would be if you didnt pull cylinders and look

[PeWe]

The rubber pucks behave like that, dome downwards probably expand by heat. They cannot go the other way.

I usually flip them next time. If rubber is still soft.
+10 years old might need a change though.

[BenelliSEI]

Most of the air leaks I heard sounded like they were from the valves.  I could not distinguish if air was coming from crankcase.
I guess I could hook my hoses again and try to listen more carefully.

Would I be stupid if I don't do rings?

What’s the mileage?

[denward17]

Most of the air leaks I heard sounded like they were from the valves.  I could not distinguish if air was coming from crankcase.
I guess I could hook my hoses again and try to listen more carefully.

Would I be stupid if I don't do rings?

What’s the mileage?

Mileage is 22k.  Another test this am with a hose to my ear.  Heard quite a bit of air noise at 3 exhaust valves, and all had air noise down the cam chain tunnel and the oil tank breather port on the back of the block.
I guess that points to rings as well.

BTW, with the cam out, I'm not rotating the pistons to TDC, but that should not matter, correct?

[PeWe]

No cam, all valves closed

[TwoTired]

A worn cylinder bore no longer has straight sides.  Part of the test is for this type of wear.  And for that, the piston must be at TDC. 

[BenelliSEI]

Most of the air leaks I heard sounded like they were from the valves.  I could not distinguish if air was coming from crankcase.
I guess I could hook my hoses again and try to listen more carefully.

Would I be stupid if I don't do rings?

What’s the mileage?

Mileage is 22k.  Another test this am with a hose to my ear.  Heard quite a bit of air noise at 3 exhaust valves, and all had air noise down the cam chain tunnel and the oil tank breather port on the back of the block.
I guess that points to rings as well.

BTW, with the cam out, I'm not rotating the pistons to TDC, but that should not matter, correct?

You can “tap” the exhaust valves open and closed a few time to try and seat them a bit better. Make sure the pistons are down in the cylinder you are tapping the valves.

[Don R]

 My experience has been if I pull the head, I usually disturb the cylinder base gasket. That leads to pulling the cylinder which leads to new rings. I learned to put rags around the rods because I'll usually drop a piece of something through the crankcase holes.
 I never understood the process by which rings become unusable when they are separated from their cylinder but that is standard practice.
 
 An engine that has sat may have carbon or even a bit of light rust that can wear away when running. Pitting however will not heal.

[denward17]

Most of the air leaks I heard sounded like they were from the valves.  I could not distinguish if air was coming from crankcase.
I guess I could hook my hoses again and try to listen more carefully.

Would I be stupid if I don't do rings?

What’s the mileage?

Mileage is 22k.  Another test this am with a hose to my ear.  Heard quite a bit of air noise at 3 exhaust valves, and all had air noise down the cam chain tunnel and the oil tank breather port on the back of the block.
I guess that points to rings as well.

BTW, with the cam out, I'm not rotating the pistons to TDC, but that should not matter, correct?

You can “tap” the exhaust valves open and closed a few time to try and seat them a bit better. Make sure the pistons are down in the cylinder you are tapping the valves.

I tried tapping the top of the valve with a rubber mallet lightly, it did help with the readings, but they are still bad readings.