New valves and seats leaking air

Messages
215
Location
utah,usa
Need some technical help with leaky valves and seats.
I have an aluminum head vtwin engine that I sent both heads to a machine shop for new valves and fresh cut of the seats. They finished them and said that they held 22in of vacuum and sealed perfectly.
I got them home and decided to check them anyway. Filled the combustion chambers full of water and blew air into all the ports and not one bubble came out of any of the 8 valves. Thought to myself that was confirmation that the machine shop did the job right.
Bolted them on the engine, and before I mounted the carbs or even turned the crank, I decided to do a leakdown test. Got 10% leakage on both cylinders, but the real bad news was that the leaking was mostly coming from the intake valves on both cylinders!
Called the shop and they said they guarantee that the valves and seats were perfect, and that maybe the springs were causing the leaks because the valves now sit further in the head, and the springs were more relaxed. But I don't buy that because the pistons were at top dead center for the install of the heads, and with the proper valve lash and both intakes and exhausts closed, the rocker arms never touched anything. In other words, the heads were in the same state on the engine as they were sitting on my bench 30 minutes earlier. It held water on the bench, but won't hold air on the engine.
So my question is which one of the following would you do?

1. Shim up or replace the springs to get these to seal.

2. Take heads to a different shop and double check the work.

3. Just run it a few hundred miles and see if the valves "hammer" in with rpms and heat cycles.

Thank you.
 
Last edited:
Messages
5,015
Location
Ohio
Option # 2. I doubt the condition will get better with time. Get it right now before bringing it to life. I wonder if the valves were lapped in.
 
Messages
25,448
Location
MA, Mittelfranken.de
Need some technical help with leaky valves and seats.
I have an aluminum head vtwin engine that I sent both heads to a machine shop for new valves and fresh cut of the seats. They finished them and said that they held 22in of vacuum and sealed perfectly.
I got them home and decided to check them anyway. Filled the combustion chambers full of water and blew air into all the ports and not one bubble came out of any of the 8 valves. Thought to myself that was confirmation that the machine shop did the job right.
Bolted them on the engine, and before I mounted the carbs or even turned the crank, I decided to do a leakdown test. Got 10% leakage on both cylinders, but the real bad news was that the leaking was mostly coming from the intake valves on both cylinders!
Called the shop and they said they guarantee that the valves and seats were perfect, and that maybe the springs were causing the leaks because the valves now sit further in the head, and the springs were more relaxed. But I don't buy that because the pistons were at top dead center for the install of the heads, and with the proper valve lash and both intakes and exhausts closed, the rocker arms never touched anything. In other words, the heads were in the same state on the engine as they were sitting on my bench 30 minutes earlier. It held water on the bench, but won't hold air on the engine.
So my question is which one of the following would you do?

1. Shim up or replace the springs to get these to seal.

2. Take heads to a different shop and double check the work.

3. Just run it a few hundred miles and see if the valves "hammer" in with rpms and heat cycles.

Thank you.

Remove the rocker arms completely and check again. The actual valve sealing is easy enough to check, remove the spring, clean the seat and valve face and make a carpenters pencil mark every 5mm or so around the valve seat then drop the valve straight in and turn it 1/4 turn with the stem the marks should be gone.
10% is too much and I cant see it getting better by shimming up the springs its not like they ground 1/16" off the seat. Properly seated valves do not require much pressure to seal

Check them and if necessary lap them by hand until they do seal. fill the back of the intake port with solvent once all traces of lapping paste are gone and the springs installed and it should hold it for a while before you see any wetting around the valve.
Check the valve guides.
 
Messages
5,936
Location
Suburban Washington DC

kawie_guy said:

I decided to do a leakdown test. Got 10% leakage on both cylinders,

10%? That seems excellent. On my tester, up to 40% is in the green.

large.jpg


Why not do a cranking compression test?
 
Last edited:

kawie_guy

Thread starter
Messages
215
Location
utah,usa
kawie_guy said:
I decided to do a leakdown test. Got 10% leakage on both cylinders,

10%? That seems excellent. On my tester, up to 40% is in the green.

large.jpg


Why not do a cranking compression test?

Sorry, I forgot to mention I did do a hot compression test after the leakdown test.
Service manual says 43psi to 75 psi is needed (cams have an auto decomp system for starting). Mine tested at front 45 and 50 rear. Before I had the new valves installed and seats cut, the front was at 68 and rear at 75.
I could definitely live with 10% leakdown if the compression numbers were nearer the top of the spec window, and if I couldn't feel air coming out of the intake on brand new valves. I just don't think brand new valves and fresh cut seats should be leaking any air. If they sealed up, my leakdown would probably in the 4% or 5% range.
 

kawie_guy

Thread starter
Messages
215
Location
utah,usa
Option # 2. I doubt the condition will get better with time. Get it right now before bringing it to life. I wonder if the valves were lapped in.
They said they checked and lapped them in after cutting. But some machinists I've talked to say that you shouldn't need to lap in valves on freshly cut seats done with a 3 angle cutter.🤷‍♂️
 
Messages
25,448
Location
MA, Mittelfranken.de
kawie_guy said:
I decided to do a leakdown test. Got 10% leakage on both cylinders,

10%? That seems excellent. On my tester, up to 40% is in the green.

large.jpg


Why not do a cranking compression test?
20% is okay in most cases but not 40% this a brand new valve job and he can hear it coming out of the intake port, that alone is telling something is not 100%. With good rings I would be looking for 3-5% or a little more but not much more. No engine is going to have 0% loss but again he can hear the loss at the intake.
It would certainly run okay with 10%
 
Messages
25,448
Location
MA, Mittelfranken.de
Here's a video of how much air is coming out the intakes. Sorry about the air compressor turning on at the end.
They said they checked and lapped them in after cutting. But some machinists I've talked to say that you shouldn't need to lap in valves on freshly cut seats done with a 3 angle cutter.🤷‍♂️
Multi angle seats tend to seal better as they wear (to a point) but I cant think of any reason you wouldn't want to lap the valves to finish them except it saves labor not to. The better the finish on the seats and valve the better it will seal and the longer it will maintain an optimal seal, they are correct you dont have to but think of it as a optional and beneficial step.

Seeing as it held a vacuum I would look at the guides carefully (I think I mentioned that before), wear in the guides can cause the valve not to close square in the seat.
 

kawie_guy

Thread starter
Messages
215
Location
utah,usa
Multi angle seats tend to seal better as they wear (to a point) but I cant think of any reason you wouldn't want to lap the valves to finish them except it saves labor not to. The better the finish on the seats and valve the better it will seal and the longer it will maintain an optimal seal, they are correct you dont have to but think of it as a optional and beneficial step.

Seeing as it held a vacuum I would look at the guides carefully (I think I mentioned that before), wear in the guides can cause the valve not to close square in the seat.
I asked the shop in the beginning if I needed new guides when I ordered the valves and they said no, the guides were good. The new valve stems fit well.

So you think running it might seal them up eventually?
I should mention that these valves and seats from the factory are a little narrow, so a lot of guys who work on these engines recommend increasing the width to .080" for more contact. I opted to do that. Some say that helps with valve longevity but may need some run time to seal off.
I don't know.🤷‍♂️
 
Last edited:

kawie_guy

Thread starter
Messages
215
Location
utah,usa
Sounds like the head is warping as you torque it down.
Could be, but there is one thing that refutes that; the front cylinder tested at 68 psi on a hot compression test just before I took it apart and had the new valves installed. Afterwards, it tested at 45. It was torqued down with the same torque wrench at the same number. So whatever caused the loss happened while the engine was sitting in my garage, and the heads were at the shop.
 
Messages
25,448
Location
MA, Mittelfranken.de
I asked the shop in the beginning if I needed new guides when I ordered the valves and they said no, the guides were good. The new valve stems fit well.

So you think running it might seal them up eventually?
I should mention that these valves and seats from the factory are a little narrow, so a lot of guys who work on these engines recommend increasing the width to .080" for more contact. I opted to do that. Some say that helps with valve longevity but may need some run time to seal off.
I don't know.🤷‍♂️
It may I really cant say but its probably worth giving it a shot, the worst case is it doesnt and you have to take the heads off again.
 

kawie_guy

Thread starter
Messages
215
Location
utah,usa
Well after a 15 minute ride around the block, I ran another leakdown test today. Nearly all the air coming out of the front intake valves was gone, and the leak down was less than 2%.
The rear intakes are still leaking but less than half what they were, and it's now at 3% leakdown. Check out the difference between the the video above which where the engine hadn't been run yet and compare it to this one from today:
 

kawie_guy

Thread starter
Messages
215
Location
utah,usa
So now the question is how can you have such mediocre compression with such an awesome leak down? Is my compression tester giving bad reads?
 
Top