Vacuum Leak Checking The Weber Induction System

PostPost by: type26owner » Mon Aug 09, 2004 3:21 pm

Thought I might share some results from testing I've done recently on leak vacuum checking the twincam induction system. Looking for an air leak by observing an rpm change by flooding the outside areas with WD-40, water, gasoline, propane, etc. is a total waste of time I've found. The rpm is changed but by such a small amount it's in the noise. The noise is the normal fluctuation of rpms from a engine in good tune and this value is about +/-25 rpms. What does change the rpms considerably is the varying the amount of air that is available for combustion but then you can't do that test. Changing the mixture strenght will not cause much of an rpm change though. Try this for yourself by introducing these products into the throat of the carbie while it's running and you'll see exactly why I'm posting this info. All the common advice is worthless so don't waste your time by believing these myths.

What does work though is to be able to do a binary type test for an air leak. You can tell if a cylinder is firing or not reliably so it's prudent to use that as the basis for a test. This kind of test can be done by closing the idle mixture screw on a single cylinder and forcing that cylinder to misfire from lack of fuel. Now introduce the combustible substance into the throat of the carbie to guage the amount needed to sustain cumbustion. It's surprising just how little it actually takes. Now you're ready to flood the suspect areas like the orings or the ends of the throttle spindles for a leak. If the cylinder starts to fire again then you've found a leak. Suggest you only use propane or butane to do this because a liquid can actually seal up a leak and fool you. Unfortunately this test only works when there is a separate carbie per cylinder. A smoke machine is the only real alternative for an engine with a plenum. Going to have to buy one now for testing my V8s. :(
-Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1704
Joined: 18 Sep 2003

PostPost by: rgh0 » Tue Aug 10, 2004 12:00 am

Keith

I have always just used a rubber tube as a listening device to detect leaks. Any significant leakage will generate sufficient hiss to hear with a tube held close to it. I run a tube around the O rings and if there is any leakage I can hear it.

May not be as sensitive as your method especially around the throttle spindles where there is noise from the throttle plates being conducted through the shafts.

Rohan
In God I trust.... All others please bring data
User avatar
rgh0
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 8814
Joined: 22 Sep 2003

PostPost by: type26owner » Tue Aug 10, 2004 4:10 am

Rohan,
That use to work well for me too along time ago. Now that my hearing is in decline I need to find other means to get the job done though. The trick is to be clever at it and find an even higher precision diagonistic technique which is quick and really easy to do. By experimenting with my AFM I've found the twincam will idle without a misfire at an air/fuel mixture of 16:1 or less. Roughly figuring that means this diagnostic technique should be able to detect an air leak at about the 7% level of the total air volume inhaled. Not bad!
Regards,
-Keith
type26owner
Coveted Fifth Gear
Coveted Fifth Gear
 
Posts: 1704
Joined: 18 Sep 2003

Total Online:

Users browsing this forum: No registered users and 12 guests