Smoke Testing Headlight Vacuum Leak

[agildawg]

Recently had my carb rebuilt and when setting it back up, my mechanic said there was a vacuum leak in the headlight system. He hooked his "smoke machine" up to the line and smoke came out the front of 1 actuator, and out the filter of both headlight relay valves. I'm assuming I probably need to either repair the seal in the actuator or replace it, but not sure about the relay valves. Does the smoke coming out of the filters mean that both valves need to be replaced? I haven't had a chance to put a vacuum pump on the various parts to see if they are/aren't holding a vacuum but wondered if you guys would have some insight. FWIW the headlight system seems to work fine, with the headlights going up/down as they should when it is hooked up. Currently I've got that line blocked off so the car runs better and no risk of burning valves. Any help is greatly appreciated.

PS - Excellent technical article on troubleshooting the headlight system on corvette-101.com

Thanks,
Dawg

 

[KOPBET]

As you noted, the front seal is apparently bad on one actuator, but I would replace the seals on both actuators while you're at it. They're cheap and fairly easy to replace.

Not knowing how much pressure you put the smoke under, I would not be surprised at least some smoke blow-by thru the vacuum relay filters even if the relays are new and especially if they are old because of the way they're designed. The relays are easy to remove and check the operation of if you have a hand vacuum pump. However they are not cheap and are going to be ~$50-60 for repro and over $110 for AC Delco.

I would also check out your vacuum check valve.

 

[gmjunkie]

Smoke very well could have came out of the filters of the relays,they bleed off Vacuum as well as switch Vacuum!!

 

[Vettehead Mikey]

I'm curious to know how he got the smoke into the system. If he pressurised it, that's how the seals got damaged. The system sucks, not blows.

By the way, hook it back up. It CANNOT cause 'burned valves'

 

[gmjunkie]

I'm curious to know how he got the smoke into the system. If he pressurised it, that's how the seals got damaged. The system sucks, not blows.

A smoke machine doesn't produce allot of pressure!!They've been using them for years to find Water Leaks,Air leaks,Vacuum Leaks,and even leaks around Door Weatherstrips and Deck Lid Seals!! Old,Old Technology!!

 

[Vettehead Mikey]

A smoke machine doesn't produce allot of pressure!!They've been using them for years to find Water Leaks,Air leaks,Vacuum Leaks,and even leaks around Door Weatherstrips and Deck Lid Seals!! Old,Old Technology!!

Yes, I'm aware of that. The point is that headlight vacuum system is not meant to receive ANY pressure at all and the results could be deceiving. If enough pressure is put on the front port of an actuator it could dislodge it from it's mount. Even if it doesn't come loose, the smoke will seep out around the actuator rod making it look like there's a leak when there isn't.

 

[bmotojoe]

If you’re using the headlight vacuum picture on Corvettes 101 web site as a troubleshooting guide its not correct. There is a small error where the hoses connect to the headlight switch.
JMO:
Let’s think about the components involved in the vacuum system: vacuum hose, vacuum hose fittings, Switches, relays, actuators. The internal makings of the relays are mostly constructed of plastic and rubber and the actuator would be steel and rubber. And let’s think of the engine as the regulator only allowing a negative pressure of 23hg maximum.
Now let’s change our thinking to a machine run off compressed air that the controls and main cylinders have a regulated air input of 23psig. Our controls consist of 2 manual switches, 2 relays and 2 air cylinders. The switch, relays, cylinders are constructed of the same internal working parts and materials that the vacuum system parts were made out of.
I guess what I’m getting at here is that the internal parts and the material there made of are no different if used in a vacuum system or a compressed air system so there should be no damage done to any of them if regulated at a safe pressure. Sure if you put 90psig to the top of most diaphragms you’re going to damage them. If a part or item will fail under positive pressure it will fail under a negative pressure If both pressures are equal to each other.
Brian

 

[gmjunkie]

Let’s think about the components involved in the vacuum system: vacuum hose, vacuum hose fittings, Switches, relays, actuators. The internal makings of the relays are mostly constructed of plastic and rubber and the actuator would be steel and rubber. And let’s think of the engine as the regulator only allowing a negative pressure of 23hg maximum.
Now let’s change our thinking to a machine run off compressed air that the controls and main cylinders have a regulated air input of 23psig. Our controls consist of 2 manual switches, 2 relays and 2 air cylinders. The switch, relays, cylinders are constructed of the same internal working parts and materials that the vacuum system parts were made out of.
I guess what I’m getting at here is that the internal parts and the material there made of are no different if used in a vacuum system or a compressed air system so there should be no damage done to any of them if regulated at a safe pressure. Sure if you put 90psig to the top of most diaphragms you’re going to damage them. If a part or item will fail under positive pressure it will fail under a negative pressure If both pressures are equal to each other.
Brian

Exactly!!!

 

[Vettehead Mikey]

Except for the front seals on the actuators and the override switch under the dash. They are meant to have vacuum only.

This is similar in some way to the seals on the brake caliper pistons- meant to have pressure in one direction only.

 

[pgtr]

Except for the front seals on the actuators and the override switch under the dash. They are meant to have vacuum only.

Curious... Is this a Vettehead Mikey opinion presented as fact? Or does Vettehead Mikey have any documented proof or empircal evidence that GM explicitly designed and engineered the system seals to 'only seal 1-way' and not both ways? (presuming we are talking about nominal positive or negative deltas either way of say no more than about 12 or 15 psi or so)

 

[Rob]

Curious... Is this a Vettehead Mikey opinion presented as fact? Or does Vettehead Mikey have any documented proof or empircal evidence that GM explicitly designed and engineered the system seals to 'only seal 1-way' and not both ways? (presuming we are talking about nominal positive or negative deltas either way of say no more than about 12 or 15 psi or so)

And is this another antagonistic jab at a member or are you attempting to raise a valid point and unsure of how to efficiently do it without ****ing other members off?

 

[pgtr]

This forum is full of crazy technical solutions that include non-intuitive approaches, creativity, tricks and tools... Sure, some are ill-advised, yet others are pure genius. In short a lot of out-of-the-box thinking. It's great to share, debate, flush out and understand such solutions. Absolutely some will always share opinions, both nay-saying and supportive. In this case, other thread contributors shared constructive personal experiences, factual data and made cases using easily understood logic. If someone seeks to shoot down a possible technical solutions by presenting a 'fact' somewhat lacking in the way of supportive information - some elaboration may be in order. Yes, I was trying to be cute as I perceived an opinion being presented as fact. Please allow me to convey my sinsere appologies to the offended member.

Getting past my own indiscretion and on to specifics: Vettehead Mikey concludes the approach of using positive pressure differential provides inaccurate results and is damaging to the system. This is based upon Vettehead Mikey's presented fact that the system, apparently by original design intent, simply cannot have positive pressure as well as an analogy between the vacuum actuation system and a hydraulic brake piston in a cylinder being "similar". I must question the fact as it has been presented by Vettehead Mikey.

 

[Vettehead Mikey]

(presuming we are talking about nominal positive or negative deltas either way of say no more than about 12 or 15 psi or so)

I'll assume that the 12-15 PSI you mentioned was a typo and that you actually meant 12-15 inches of mercury- which is representative of engine vacuum.

Perhaps GM does have a technical paper on what would happen if a vacuum system inadvertantely becomes pressurised, but I would be truly surprised. I'm sure they spent their R&D dollars on more productive efforts.

I made the comment about it potentially damaging the system as some newbie may follow bad advice and pressurise it with 12-15 PSI- or whatever a shop compressor can muster.

When restoring my car years ago, I used a laboratory grade vacuum/pressure combination pump with a relief valve set at 15 in/hg. Initially this was set to 'vacuum' and connected to the system inlet tube on the intake manifold. I was unable to find many leaks as the noise of the pump drowned out the subtle 'hiss' from the leaky components. I reset the pump to 'pressure' and repeated the test, thinking that I might 'feel the breeze' from whatever was leaking. Immediately upon starting up the pump, both headlights slammed closed- the opposite of what was commanded by the headlight switch.Why? The system was operating in reverse.

I shut off the pump and set it to 'vacuum' again, expecting both headlights to raise as commanded. Nothing. Long story short, the front seal on one actuator had been dislodged from it's mount creating a huge leak to atmosphere and was now preventing the system from building any meaningful vacuum.

Dis-assembly of the actuator to replaced the dislodged seal (commonly called a witches hat) showed a design which was obviously meant to have pressure in one direction only. Had the seal not been dislodged from it's seat- ending the experiment, continued use of pressure on the front side of the actuator would have resulted in leakage around the seal/actuator rod interface. This 'leakage' would have given a false indication of a defective seal- when in actual fact it was simply been subjected to an operational configuration it was not designed for.

The comparison with the brake piston seals (common to the the aircraft engine trade as well as automotive, and known as a 'V' seal or 'Garlock' seal) is valid for similar reasons. The source of air entering the hydraulic system at thew caliper end is most often from piston 'knock back' where there is momentary reversal of pressure across the seal.

In the aviation field, the value of .5 psi is considered the maximum allowable pressure reversal as a design limit.

I hope you can see and accept the correlation

 

[pgtr]

Dis-assembling mine long ago don't recall the same as above...

At any rate I checked at work and it looks like we have an old machine just like the one described here. I have a couple buddies locally whose cars have issues this could help troubleshoot. And it will be interesting to try on mine just to see what can be seen.

Thanks for the tip on this machine! I'll post my results at a later date.

 

[agildawg]

OK, so I finally got around to checking each individual component with a vacuum pump and guage. It confirmed the smoke-test diagnosis. The grommett/seal on 1 actuator is leaking but the diaphragm itself was good, and both relay valves were leaking. Just thought you would find it interesting that individual component testing confirmed the smoke test results.