Trying to get some troubleshooting help on what I believe now is a more major problem with my 4g63. For fast responses, first my basic questions, then the story details:

Is it possible that a leak down test won't work well if the MIVEC gear doesn't have any oil pressure in it? Could no oil pressure cause the MIVEC to slop around so that a leak down test shows leakage because the cam is slapping closed while a compression check shows decent compression?

history & details:

A few months ago the car starting getting down on power when on boost (daily driving). I thought I may have run into another boost leak. So I connected my boost leak tester and it held 30 psi no problem. That was unexpected, and a drag, because a boost leak is easy to fix.

Then I moved on to other troubleshooting. replaced the plugs, checked the gaps, checked the timing belt and alignment, cleaned the MAF, etc. In one of those sessions I grabbed the turbo compressor shaft and it wiggled around more than I thought was appropriate. So I thought I found the problem, turbo done, need new one. I decided to buy the HTA71. A few weeks later I got it and was about to shoot some videos of my failed stock unit, and found that the new turbo shaft was exactly the same. Crap!, I realized the stock 9 turbo was fine, and I had a different problem.

Then I decided to do a leak down test. Got my tester out and tried to get good values. But things seemed to be moving around on me. I redid everything 5 times, and essentially the tests were telling me I had bad rings, bad intake valves, and a breech between 1 and 2 cylinders. What? How did I have all of that all of a sudden, when I could drive the car fine just off boost. I decided to get another tester and confirm my gauges were good.

The second leak down tester uses way more psi. Why? I don't know. Mine only used like 15psi through the regulator, this one would need 90 psi to get the leak down gauge to setup in the SET zone. This tester didn't indicate any significant leakage. ?

Then I did a compression check, from timing belt over to the right; 150, 125, 150, 150. So I don't think the compression check indicates a head gasket failure, but the leak down shows everything is broken. Those two don't go together.

The only other clue I have is that when I tried to prime the oil for the new turbo oil feed inlet, I couldn't get any oil coming out of the head feeder passage. I cranked over (relay pulled) 10 or 15 seconds. Nothing. I decided not to start the car, and focus on the above tests. Then I also realized I was hearing the MIVEC gear slam over while doing these tests. This was indicating no oil in the gear or pressure, which matches no oil at the turbo feed passage.

I am trying to figure out what my problem might be before I go ripping the head off. trying to troubleshoot smart and get some help from the gurus here before I waste a ton of time and parts on nothing because of a bunch of false positives.

One possibility is my stubby oil shaft/pump could be bad (no balance shafts), but that doesn't mix with my cars behavior. Oil light only flickered when idle is stumbling low, never had metal in oil, or over heating.

Could a possible oil pump failure cause the MIVEC weirdness that could cause the leak down test to be weird? I suspect that the leak down with the high pressure tester, was slamming the intake cam around to close the valves. But, the low pressure tester didn't flow enough air to quickly seal those valves and so the tester showed massive leakage. Actually it was just the time for the valves to close from the pressure. This is possible because the cam is lose from the MIVEC gear/timing belt which I am adjusting with the crank to be at top dead center compression stroke for the appropriate cylinders during the test. However, the MIVEC gear doesn't have a hold of the intake cam because of the oil problem? Does that make mechanical sense?

Sorry that was way too long, but I needed to complete those tests before I had a post folks could help me with. Thanks for any ideas or input.

 

MIVEC seats the intake cam at full retard at rest (AFAIK), which shouldn't cause a mechanical issue where a leakdown test is concerned. Both valves are closed at TDC of the combustion stroke.

Leakdown test problems are often an issue of improper method, and/or incapable test equipment. After discovering that many (if not most) cheapie testers are useless garbage, I made my own FAA grade unit very inexpensively:
https://www.evolutionm.net/forums/ev...wn-tester.html

An FAA spec leakdown test is performed with a warm engine, and with an input pressure of 100 psi. The object is simply to measure the leakage through a 1 mm orifice, whereby 100 psi on the input gauge and 95 psi on the gauge opposite the 1 mm orifice = 5% leakdown through the cylinder.

Let's be sure you've got these ducks in a row before we look at anything else.

 

Well i watched several videos on the proper use of the leak down tester, but I suspected mine was not giving me good data.

Mine is the cheap HF type. I set my compressor regulatory at 90 psi, and then connect to the tester. I'd slowly open the valve on the tester until the second gauge got to zero/set. At this point however the primary gauge only shows about 15 psi. If I let in more pressure the second gauge would overshoot and pin. So the only way to get a good set is at 15psi. The videos I watched on this gauge matched this, and commented only 15psi or so was needed.

I should not that when I had that engine out years ago, I used the same tester and ended up with reasonable 90% values on all cylinders, so I was assuming I still know how to use it. I can't recall if I did those tests with a timing belt or not. I think I just set the cams manually to closed, because I had the engine on a stand for months.

At TDC I would connect the tester to the hose in the cylinder and would read 40% leak on 3 of them and 60% on one of them. That seemed very high, so I suspected the gauge's accuracy. I could also hear the intake cam slapping around against the MIVEC gear.

The car has been sitting for 2 months, so doing this test warm isn't an option. I also pulled the turbo and all that other stuff I mentioned so now I am not able to start it without risking more possible damage. At least for now that was my logic.

The second leak down tester is from a mechanic friend of mine, and its a quality unit. Not HF. That gauge behaved the way TedB notes. It required 100psi in order to zero/set the second gauge. When I connect the gauge to the cylinder I didn't read any significant leakage.

If I can figure out how to mount my phone or have someone hold it, I will shoot some videos of me doing the test to rule out user error.

I was wondering if since I haven't driven the car in so long that the oil pressure in the MIVEC gear has drained out allowing the intake cam to rotate inside of the gear. When I was cranking it for the compression tests I did you could hear the MIVEC slap around every revolution. I was thinking that this could be throwing off my leak down tests with low psi. Since it takes some air pressure to close the intake valves if the cam is lose from the gear position that I set with the belt.

My compression numbers are decent except for one slightly lower. However based on what I read 125 is no reason to rebuild. The manual indicates 92 psi and the minimum.

 

Get a bore scope and look inside the engine. That will tell you if you have physical damage.

 

What HF is selling is useless. Been there, done that.


Fair enough.


Keep in mind that as soon as one changes cams, the factory spec goes out the window. What matters is a relatively consistent reading across all cylinders. A significant variation in one indicates a potential issue. I'd repeat the test, and if you get the same result, you can try squirting a little oil in the cylinder to determine if it's a ring issue. If not, you may have a valve sealing issue, or possibly a head gasket issue.

 

If it were me I'd be more interested in what oil pressure is doing than what compression is. I'd get a mechanical gauge on it. The oil light is there more for entrainment than anything else.

 

I shot several videos trying to demonstrate the use of the better leak down tool. I would get about half way through a test, and then either make a mistake in my process or the test would do something unexpected. I got frustrated with all the videos and just put everything together except for the oil feed like, had someone cranked the car with the relay in and plugs back in and it fired. Oil immediately started shooting out of the oil feed line for the turbo. Shut it down, and then put that bolt in and fired and drove the car. Just ran out of patience. It seems to run OK, but it has the same problem it had before. So I have proven the oil pump is working, and that the new HTA71 did not solve my problem. Old stock turbo was fine.

I tried doing videos again tonight, and noticed that the timing belt was lose between gears. I freaked out thinking it was a failed timing belt tensioner, so I took the surp belt and all the pulleys off and am knowing inspecting the belt tensioner.

However, I just ran into some threads that indicate the timing belt stopped in certain orientations could appropriately have slack between the two gears? Is this true? I never noticed that condition, and I thought I had found a smoking gun.

Also, I did some logs of part throttle 10 psi boost and decent load. AFR's look good. advance is normal, and knock is zero. After 10 psi power drops off. Behaves like a boost leak or a MIVEC failure.

 

It is true about the belt being slack between the cam sprockets depending on how the engine stops.

 

crap. I was hoping to find a smoking gun.

Is it possible that if the belt stretches and or the adjuster is maxed that the slack in the belt could cause the cams to be off or lose in relationship between each other that would produce the type of power issues I refer to. Essentially everything fine until 4k, 10psi, and then it sounded like its running rich and power drops off?

Any good ideas on how to troubleshoot the MIVEC system? After that I only know to pull the valve cover and have a visual inspection of the top of the head and valve gear. I suppose then pull the head and get it redone or built and hope that shotgun approach addresses whatever is going on?

is there any logging data that could help identify the issue? Is there anything any of you would want to see or know?

Thanks for all and any input. I am kind of stuck.

 

I took out the cams, did a complete lifter service and cleaned and inspected all of the MIVEC components. Put it all back together, and none of it seemed to have any issues. I am going to rule out the MIVEC for now.

Back to some kind of boost leak, maybe the 3port, no ideas.