Checking for vacuum leaks

[manicmechanix]

OK say you want to check your engine for vacuum leaks (unmetered/false air) with a vacuum gauge (as far as I know a vacuum gauge would be the easiest way). Now, even a slight vacuum leak should show lower vacuum reading right? And might make the idle a little rougher, and the idle speed slightly higher, maybe not noticably higher, but the higher idle is where my my question lies. On a fuel injected car with IAC wouldn't the computer readjust the idle within it the IAC's limits to normal rpm by closing in the IAC and therefore the vacuum and idle speed would seem normal, yet there would still be unmetered air getting into the intake (the main problem with a vacuum leak in the first place)?

I think if you hook up a scanner or go into diagnostic mode you get a constant IAC, so I assume then if you don't have approx 20 in a vacuum, there might be a leak, but what if you don't have it in diagnostic mode? Can anyone shed some light on using a vacuum gauge to check for vacuum leaks?

[geldartb]

easiest and most direct way ive found is to use one of those small propane torches that you use for soldering plumbing.

i did that when i was searching for a vac leak that i couldnt find and with the torch found it with 1 minute with the propane.

if you have a leak while your sweeping over the engine with the propane it will cause the engine to rev higher where the leak is.

i think this is a little more direct than using a vac gauge cause ive got a vac gauge in my car and while it was reading low it doesnt tell you where the leak is, and using DHP my iac counts were also way off indicating a vac leak/

[PCGUY112887]

If your vac leak is TINY then your IAC may compensate enough for you to not notice, but if it's of any decent size to (in my experience) bring idle 300+ RPM's from normal then eventually (on OBDII) you will set a High Idle code.

[manicmechanix]

Good replies so far. What I was getting at more than finding the source of a vacuum leak is just how to simply and accurately detect small vacuum leak on FI engines.

In practice it seems even a small vacuum leak does affect at least the idle quality and maybe the starting on FI cars. Now in theory, a carbureted engine should always show a vacuum leak on a vacuum gauge (lower vacuum reading, maybe slight faster idle) and would be more affected by a vacuum leak because not only is excess air getting in at idle it is also reducing the signal to the idle port and actually the signal all the way through the driving range (on a carb all air must go through the carb to be metered).

But on FI the IAC could compensate to a degree and maintain the proper idle speed and vacuum. So a small leak wouldn't show up on a gauge or tach. I assume with a scanner you could read if the IAC steps were a little high but not high enough to set of a code? Also, besides the fact that the IAC would maintain vacuum level there wouldn't be false/unmetered air due to the MAP sensor measuring lower vacuum anyway, or would the MAF sensor overide it and there would be unmetered air (not to mention the O2 sensor in closed loop which I know overrides the MAP and th MAF)? I also assume that with a scan tool you could measure the vaccuum reading from the MAP and the output of the MAF.

I guess what my question is isn't a vacuum gauge reading almost worthless on FI for small leaks? And also I was wondering about how or what a scan tool would tell you about a vacuum leak. Or maybe I'm making too much of it and small vacuum leaks don't even affect FI engines .

[bszopi]

By using the propane torch, as suggested above, the vacuum leak will suck in the propane and cause the engine to surge, since it just recieved more fuel to burn.