Vacuum/boost leaks and pressure testing intake (long)

I accidently deleted the private message that asked for me to write this, so I'll just post it publicly (I'm afraid I forgot if it was here or TF too.)

The original idea for this came from a post copied by Ted B from somewhere years ago and reproduced in total below.

Vacuum/Boost leaks (and you'll see why I seperate the two in i little while) seem to be a very important part of tuning. Many people spend a lot of money on smoke tests, or buy vacuum pumps to check for vacuum just like the good old days when n/a was all they really cared about. I personally feel this is very important, but also very frequently overlooked, or simply ignored due to expense.

With simple plumping parts that cost no more than a few dollars, and a source of compressed air, you can check for vac/boost leaks anytime you wish.

Close up of various adapters required with sizes shown: http://troutlake.net/gallery/album125

Here are more pics of the process the first time I tried it.: http://pics.l7.net/index.py/Photos/Misc/Pressure%20Test (I found the rubber would compress enough I didn't need the duct tape to make it seal)

You can click on any thumbnail above to get a 640x image, then click on it to get a full sized one.

First step is to unhook your air filter and use that as the attachement point if possible. If not, drop back and hook it to the front of the turbo. You want to pressurize as much as the system as possible.

Instead of the plumbing parts shown above, you could use just a rubber adapter and pvc pipe endcap of the correct size. With that you could drill and thread a male quick connect for your air hose or a schrader valve into the pvc endcap. Be sure to teflon tape those connections!

Matt S wrote:
Basically all the shops said there aren't any replacement parts, and all 10 or 15 of them I called said they have never seen one NOT leak a little smoke when doing the smoke vacuum leak test. so I put it back on, still get 18"-20" vacuum depending on the day.

I was also getting 18" of vacuum, however, I found to my suprise, quite a number of vacuum/boost leaks, the main one (the throttlebody shaft seal) alone bumped my vac to 19" after I had the original seal milled out and a bronze bushing installed in it's place

It's amazing the number of small leaks you can find this way. BTW, I never pressurized it over 20psi (that's when the dipstick blew out) but I plan on testing to ~25 next time. Although I heard no leaks at 20psi that I couldn't hear at at 5psi, so higher pressures may not be neccisary, except to see what hoses you're going to blow off if you actually run that much boost pressure.

Another problem with vacuum testing (and the reason I made the distinction above) is that some hoses and stuff will only leak under boost -- not vacuum. If you think about it, vacuum is pulling the cracks together, the boost is opening them up. The turbo to throttle body hose is a good example on my car of this. You can stand under the hood and rev the engine and see the cracks pulling together and tightening up.

I imagine this could be the cause of some wierd rich/lean conditions and difficult A/F tuning some people have - but that's just my opinion

Another thread that touches on the subject: http://www.turboford.org/cgi-bin/ultimat...356#000000

Post by Ted B, and, i believe, the post that was my original source for the idea:

Ted B:
Here is a nice method for finding boost leaks that someone - posted in another forum recently. I felt it clever and useful enough to re-post it here:

After reading a few posts about finding vacuum leaks, I did my own check.
I went to Lowes and bought a 2" PVC peice and an end cap.
I drilled out the cap with a 1/2" bit. I made my own tap by cutting groves into a spare 3/8" NPT fitting. I cut the threads into the end cap with it, then used a new one with lots of Teflon tape to seal it in. I used PVC cement to join the two pieces...instant pressure checker.
Next I removed my intake (HKS RS). I secured the pressure checker to the turbo inlet, connected my air hose and dialed in 10psi of pressure from my air compressor.
I immediately heard hissing.
1st leak was the pressure line going from the compressor housing to my HKS EVC boost sloenoid. It was leaking BADLY at the housing. I replaced the entire line with a new, smaller line (4mm I think), and double zip tied both ends.
The next leak was my Turbo XS H34 (recirc) BOV. It was seeping. I checked it by spraying water with a little dish soap in it. I removed it and cleaned the seat. After reinstalling it still leaked a bit, but not as bad. I think if I use some valve lapping compound it will completely seal.
the last one has me.
It's leaking at the throttle body, at the butteryfly shaft. If I giggle the shaft, I can hear it leak more and less. I'm not sure what to do? Can it be rebuilt with new bushings? It's not too bad, but I think it is actually a vacuum leak during non-boost times. My idle sucks, I do have 272s but I have cam gears set at -4/-1.

I also sparyed the soapy water over the entire boost circuit, from turbo, IC pipes, IC, BOV and all vac lines. Also the fuel injector seats.

I didn't even think I had a leak and found 3. I suggest this to all. Cheap insurance. Cost less than $5 for the parts, as long as you have an air source.

I like his idea about the soapy water -- I'm going to use that next time myself to locate those leaks that are too small to hear.

I hope this helps, and feel free to ask questions if I've not been clear enough.

------------------
-asi

1986 TC, 132k pretty well taken care of miles. k&n, gillis@15, timing @ 12 (spout out), A/C delete. 3" Downpipe (no other exhaust... Whoever described that sound as a "pissed off tractor" hit the nail on the head)

[email pics to asimov at forced-induction org for sales/id/showoff purposes (free hosting)]

[This message has been edited by Asimov (edited 04-16-2005).]

Asimov. Great article. We will massage it and put it in the tech article if it's OK with you.

Are there any places you blocked off, like the PCV line or the fuel pressure regulator line??

Here are some of the picts originally shown in Asimov's links. The device can be connected to the turbo or the throttle body.
http://natomessageboard.com/uploads/0000...Needed.JPG http://natomessageboard.com/uploads/0000...talled.JPG http://natomessageboard.com/uploads/0000...essure.JPG
http://natomessageboard.com/uploads/0000...3Large.jpg


[This message has been edited by Pete D (edited 04-16-2005).]

Originally posted by Pete D:
Asimov. Great article. We will massage it and put it in the tech article if it's OK with you.

Are there any places you blocked off, like the PCV line or the fuel pressure regulator line??
]

I've done it a number of times, and never had a need to block anything off -- It seems anything that sees pressure with this method would see pressure under boost too... so... Should I have? My EGR is deleted, mainly because of the vac leak it had when I first tested.. if that makes a difference.

You're welcome to edit however you please before adding as a tech article (You're far more familiar with these cars than I am)

thanks!


[This message has been edited by Asimov (edited 04-16-2005).]

I did a similar method to test my FMIC for leaks. blocked off both ends then used a valve from an old bike tire hooked up to a vac line to test it.
The best way to test would be to bring it to a certain PSI, and watch the pressure drop. you'll want a VAC/Boost gauge to do it right. I couldn't find all the leaks but i stopped after i got it to holding steady and dropping like 1 psi every 20 minuites or so. i'm sure if i used better hoses then it would last a lot longer. I have seen places where it wont leak at 5 psi but will at 10. I'd suggest running it a couple PSI higher then what you will actually run the car at.

I didn't think the intake would hold pressure like that. i'll give that a try soon.
What would be better is to find a way to smoke it also, but most methods i can think of would leave a greasy film and you dont want that.

How about putting dry ice inside one of the hoses the cold wouldn't hurt?

Then connect everything and pressurize it. Should fog out of the leaks, right?

Should have put this in the last post, sorry.

If you think about the volume of air moving through your intake system, I personally wouldn't worry about even 1 psi a minute, much less in 20 -- But you can be damn sure it's not leaking at that rate [Image: smile.gif]

I used a combination of the gauge on my air compressor and my boost gauge (my boost gauge seems to be more accurate by far)

Your advice about running a few psi higher than you plan to run the car at makes sense. Just because I didn't have any more leaks at 20psi than I did at 5 doesn't mean others would have the same experience.

Just had a thought (ouch!)

If you have a cam with a bit of intake/exhaust valve overlap (where both are open at the same time) wouldn't that mess this test up if the cam was in just the right position?

If memory serves, stock cams don't have overlap, and it seems like I remember reading that forced induction engines don't usually like that anyway -- So it may not be an issue at all, but I did want to bring it up.

Well, here is 4 posts in a row!

You can test your PCV valve when you're done testing/fixing vacuum leaks by pressurizing the system and taking off the oil filler cap in the valve cover (or taking the breather off) and then comparing the pressure loss. If it's enough to tell the difference, you should probably think about replacing or de-junking the pcv valve.

[This message has been edited by Asimov (edited 04-21-2005).]