Find Intake Leaks!!

[jholder]

Originally Posted by Enthalpy

Sure.
If you directly connect the leak tester to the turbo inlet and not to any pre-turbo plumbing, you won't have to worry about the breather being pressurized.

Ahh, I understand. The example usage I saw for this test was to remove the air filter plumbing, and connect the tool directly to the intake of the turbo.
As I was out I picked up a small tube of silicon caulk/sealant to help seal off the lines, then I remembered that one of the reasons an o2 sensor could fail if was exposed to silicon. I think I'm going to stick with hot glue, epoxy, or plumbers dope/teflon tape to seal the gaps.

[jholder]

So I got everything set up, and started pushing into the air at 10-15psi or so. Immediately heard a hissing, then realized I forgot to plug the IAC hose. Got that blocked off and then listened again. What I expected to hear was a hissing noise if there were any leaks.

I didn't hear a hissing noise, but it wasn't completely quiet either. The best way to descibe it was the noise that a tire makes when you're pumping air into it. A muted hissing I guess is the best way to put it. I held it for a while expecting that filling noise to stop, but it did not. It sounded like it was coming from the area around the intercooler or intercooler-to-throttlebody piping

So my question is, should I be listening for complete silence, or should there be some amount of noise as air goes through the system.

[Enthalpy]

You need to rotate the engine to find the spot of least valve overlap. If you have high duration cams, this will be tougher. Use 5th gear and physically move the car.

The hissing won't necessarily go completely away. If you have a stethoscope, then you can pull off the end and "sniff around" for the leak. It will practically deafen you when you come across it. If you don't have a stethoscope, you can put your hand around all connections you can think of that shouldn't leak and listen for a change in the pitch of the hissing.

Don't forget to check down around the injector holes.

[jholder]

Found a nice little hole right in one of the lines.. how'd that get there. :/

[PVpower]

i tried using one of these a few weeks ago. What am i susposed to do about the egr? i am pretty sure that the air is going through those lines? maybe i am wrong though. I am not real confident that i did it right.

[Enthalpy]

The EGR shaft seal is probably leaking. Mine did the same thing.

[jholder]

So, still trying to track down stalling issue (more here-> Engine stalls once warm...). I've tested most of the system, and everything is within spec. What several people have said, and I am leaning towards now is a mysterious vacuum leak somewhere.

Video to follow, here is what I've done.

1.)Hooked my DIY intake pressure tester up directly to the intake of the turbo (airbox and accessories removed).
2.)Plugged the PCV hose (valve cover-to-intake hose)
3.)Plugged the idle air intake controller hose.

I cranked up the air compressor and set the regulator to somewhere between 15-20psi.

I also borrowed a squirt bottle and some "Super Magic Bubbles" from my daughter.

I felt around and listened, found a couple of small semi-leaks around the intercooler piping, pretty much all went away after tightening the clamps down again.

I could still hear hissing, so I kept touch hoses and listening. Finally I pinched the valve-cover-to-throttle-body hose, and the hissing dropped by a good 80%. I also sprayed some bubbly goodness on the top of the TB, and noticed very tine (looks like hydrogen peroxide on a cut) pools of bubbles forming very slowly along the TB/intake manifold seam.

Watch the video, around 17 seconds or so, you'll hear the difference in sound as I clamp the hose. you'll see the bubbles I mentions around 40 seconds or so. Keep in mind those took a while to show up.

So, am I doing it right, does the system sound normal? Is the TB pressurizing the valve cover, and the gasket is shot and that is where my leak is coming from? Argghh....

(Sorry for the video quality)

YouTube - VIDEO 004

[baktasht]

that one you're clamping goes to the valve cover. Your vave cover gasket might be leaking or your valve cover bolts might just be loose.

[jholder]

Originally Posted by baktasht

that one you're clamping goes to the valve cover. Your vave cover gasket might be leaking or your valve cover bolts might just be loose.

Valve cover bolts are pretty secure, but having a leaking gasket could be a possibility. Would that type of leak affect the idle. I'm not sure of all the locations that an intake leak could occur that would cause an issue.

Thanks for replying btw.

[PVpower]

Originally Posted by Enthalpy

The EGR shaft seal is probably leaking. Mine did the same thing.

Could this cause the engine to idle really bad? also i feel like i have a really bad vacuume leak could this be causing it too?

[Enthalpy]

Yes, without a doubt, a bad vacuum leak will cause very crappy idling.

That valve cover would have to be LOOSE for it to cause idling issues. You can figure out if it is or not easily enough by disconnecting both PCV lines, leaving them open to the atmosphere and plugging the throttle body port and the pre-turbo intake port.

[Bowenaero]

Hey what if I block off this instead of clamping off the vacuum line right at the ISC valve:



It looks like the same line. Correct me if I'm wrong.
It's obviously already blocked off in the picture.

Also, you said to move the car in 5th gear to the point of least valve overlap. How do we know that point without looking at the cam gears?

[Enthalpy]

Listen to the hiss of air through the intake plumbing for least overlap. It's not critical to find least overlap. I now just run my hands across all the intake and listen for changes in the hiss pitch as my hand moves over the leak site.

Yes, you can block off the line you have pictured without issues.

[lude1888]

Originally Posted by jholder

1.)Hooked my DIY intake pressure tester up directly to the intake of the turbo (airbox and accessories removed).
2.)Plugged the PCV hose (valve cover-to-intake hose)
3.)Plugged the idle air intake controller hose.

I did the same, but my system wouldn't hold pressure at all it would go to almost 1 psi than almost instantly drop to zero. my vacuum at idle is -8 psi and car still runs fine, but why can't I pressurize my system to find my leak?

[A-RON]

im gonna have to try that thats a really good idea i jus got @ 91 so i need to cheack that out ya know jus to make sure

[Bowenaero]

Alright, just found out some useful info:

If you put the #1 cylinder at TDC (line up the marks on the crankshaft pulley - see BGB under "setting base ignition timing"), no air will seep into the combustion chambers because all the intake valves will be closed. You can do this by the aforementioned way of pushing the car in 5th gear or you can just crank it a little bit at a time with the key. What I do is crank it until it's close to TDC then push the car a little bit to get it right on.

Another thing you can do is block off the line going from the throttle body to the valve cover - the other side of the PCV system. Disconnect the vacuum hose and plug the TB port. This way no air will be going through the crankcase at all and you should be able to pressurize the entire intake tract without anything hissing (except the leak you're looking for).

Hope this helps!

[jholder]

wierdness. So I bought vacuum gauge the other day to see what exactly my vacuum was like when it was stalling. Just to be easy, I tried to T into the line that runs from the throttle body to valve cover. I thought I had screwed things up because I was showing 0 vacuum (with the car running of course). none at all. After i unhooked it and tried the mouth test, and then reconnected it and tried pinching off the lower piece of the tubing, I realized that even though the TB was sucking air, i couldnt make a vacuum while allowing it to pull air in from beyond the valve cover's hose. I could pinch off the line going to the valve cover, and I would see it shoot up. (Though only still to 13 in hg, unless it was revved up.)

I recently replaced the valve cover gaskets (then I pulled the VC again for powdercoat) so I've had it on and off a few times (though I've had this problem long before i changed anything).

I also tried blocking off the PCV hose while i had the gauge hooked up. That didn't seem to make any difference though. (i.e. block the pcv hose, left the tb->vc hose unclamped, and saw no increase in vacuum.


*sigh*

Is the valvecover _supposed_ to have, or be able to make a vaccum? I have a feeling it's not

Anyway, I moved the T over to the line running to the stock boost gauge. I couldn't get an even reading no matter how hard I tried. At first it seemed ok as I rev'ed the engine, but as I let it sit there at 1000rpm, it just kept dropping and bouncing between 10-13 in hg. If I let go of the throttle, it would immediately dip to ~500 rpm and around 6ish in hg. before stalling out.

I'm serioously thinking the worlds biggest vacuum leak, but I'm just not finding it. Would a leak in teh pre-turbo intake track cause something like this?

On a side note: Someone smarter than me can probably figure this out. There must be some way of isolating the leaking area. For instance, if you had two clamps, you could plug hose X1 and C2, and find out if part Z was leaking. (basically, there must be an order of plugging hoses that would help you locate the source of a leak.

I love wikipedia. Let me answer my own question. the pcv system basically pulls fresh air from the pre-turbo intake tract, and pulls it into he throttle body to be recombusted. IT would make perfect since that this would never present a "real" vacuum. Well damn. The leak is somewhere else.

[jholder]

As a mild follow on to my last post. Using the aforementioned tester, tonight I found my issues.

A crack in the TB -> intercooler hump coupler that would leak or not depending on the position of the clamp base.

Major leak where the cold start injector mounts to the TB. (well lack of seal actually). I believe this was my biggest issue since I never noticed the hump hose leaking before and always attributed the noise to the IACV. I'm not sure why I thought it being open would vent to the atmosphere, but whatever.

After doing this several times, I think I would give out the following advice. in using this tool.

Clamp off the like going to the idle air control valve.
Clamp off the PCV hose.
Clamp off the TB to valve cover hose.
With the exception of some air lost out of the valve, this should give you a pretty good seal).

I would attach the leak detector in two place. First at the turbo inlet. This pretty much covers the turbo connections, BPV,Hotpipe, hotpipe to intercooler, and intercooler to throttle body connections.

Second, I would connect the tester to the throttle body funnel(?) (this eliminates any loss that occurs in the intake plumbing). To some extent you can isolate which portion of the tract to look at if there is an issue.

When you have a seal, assuming you're using an air source with the regulator turned down to a safe low pressure, you should be able to hear air moving through your hose until the pressure equalizes. If you have a seal, you should hear what appears to be a rush of air, and then you should hear it "slow down" once the motor and air source begin to equalize in pressure.

I'd say the easiest thing to do is to listen for that pressure equalization, if that doesn't occur, you know you've got a leak, and then its just a matter of tracking it down.