| Engine Sealing Techniques
Submitted by http://www.autosite.com
Ever since Noah used pitch to seal up the hull of the Ark,
mankind has been fighting leaks. But it wasn't until the advent
of the automobile over a century ago that this challenge became
a precision undertaking and was faced by so many people so
often. Engineers and mechanics used all kinds of materials
and substances to keep engine oil, coolant, vacuum, and compression
properly separated soft metals, leather, paper, even, believe
it or not, beef tallow.
It was a tough battle, however, and leaks of every kind were
some of the most common mechanical failures. And the same
can be said even now with high technology in full swing. So,
here's a course in the science of engine sealing.
Big four
In any seam, four factors are of supreme importance:
- the condition of the surfaces in terms of straightness,
flatness, smoothness, and cleanliness
- the quality of the gasket design and material
- the care taken during installation
- the clamping pressure
- Keep these in mind when you approach any assembly job.
Headache
The engine gasket with the most difficult job is the one
between the head and the block, and things are worse today
than ever before because of higher temperatures, increased
pressures, and fewer and thinner head bolts. Then there are
bimetal engines (aluminum head/iron block), wherein differing
expansion rates scrub away at the gasket and cause head warpage.
If the head is going to a machine shop, cleaning and surface
checking should be done there. Otherwise, these critical details
are your responsibility. Make sure you remove all of the old
gasket shellac, carbon, and corrosion. Especially on aluminum,
be careful not to make any deep gouges or take off any metal
in the process. I've heard from gasket engineers that savage
cleaning procedures, such as using abrasive disks on a drill,
can cause unsealable unevenness almost instantly.
Well and truly flat
Check for warpage using a straightedge and feeler gauges
lengthwise and crosswise. The traditional rule of thumb says
if you can slip a .006 in. gauge between the head and the
straightedge anywhere over its entire length, or a .002 in.
gauge in any six-inch span, the casting is too wavy to guarantee
a decent seal.
But that may not be good enough anymore. For example, at
least one of the major gasket makers has tightened its recommendations.
For a V6 or three-cylinder head, it's now .003 in. over the
whole length, for a V8 or four-cylinder it's .004, and for
an in-line six .006 (come to think of it, that's easy to remember
because the thousandths are the same as the number of cylinders).
Max deviation from flatness across the width remains .002,
and there should be no "sudden irregularities" exceeding
.001 in any three-inch diameter. By the way, this is the sum
of the values for both head and deck.
But flatness isn't the only sealing surface consideration.
The level of smoothness or "tooth" is also important.
If it's too rough, the gasket won't be able to fill up the
scratches, but go too smooth and there won't be enough bite
for a good grip, so there'll be unwanted movement in the sandwich.
The typical range recommended today is 30110 RA (that's "Roughness
Average," which you multiply by 1.1 to get RMS) for an
iron head with 60100 preferred, and 3060 RA for aluminum,
5060 preferred. But there's a caveat: These figures apply
to all head gaskets EXCEPT the new multi-layer steel type
(also known as "RCE" for "Rubber Coated Embossed").
This variety will only tolerate a maximum of 30 RA. Actually,
the smoother the better here.
This is another situation where you'll have to rely on your
machinist, but if you want the insurance of an extra quality-control
measure, get yourself a surface comparitor gauge.
I'd better address the question of supplementary sealants.
If the gasket has faces of graphite or a soft synthetic material
such as Teflon, or has elastomeric beads printed on around
potential trouble spots, don't apply any sticky stuff as it'll
soften that carefully-manufactured surface. Metal faces should
receive an even coat of a suitable chemical sealer.
Nail it down
Achieving the correct clamping force on the gasket is critical.
To insure a proper translation of bolt torque to pressure,
clean the head bolts with a wire brush and coat them with
whatever the car-maker recommends. Generally, that's motor
oil (put some on the underside of the bolt head, too, because
that contact area represents a significant percentage of total
drag). Where sealer is required it'll also do well enough
as a lubricant. On blind holes, be careful not to overdo either
or you could end up with a hydraulic lock that stops the fastener.
If the bolts screw into aluminum threads and you opt to use
anti-seize compound, you should be aware that this stuff is
a much better extreme-pressure lube than oil, so it'll increase
clamping pressure at the same torque, perhaps squashing the
gasket beyond its design limit. You might want to use the
lower end of the specified ft. lb. range.
In cases where bolt tension has drawn the metal up around
the holes in the deck, chamfer or countersink them. Chase
the threads with a tap, then clean them with a rifle brush
and compressed air. If it looks like the bolt heads were digging
in around the holes in the casting, add hardened flat washers.
Torque the head bolts in the proper sequence so the gasket
can spread evenly without wrinkling. Use at least three increments
to reach the spec.
Super twist
These days, we've got another complication to deal with:
angle torque, also called "TTY" (Torque-To-Yield)
or "torque and turn." Whoopee. Nobody who's used
to handling wrenches feels comfortable stressing a bolt that
much. But if you don't heed the instructions, you'll risk
a blown head gasket.
The idea is to achieve the consistency of clamping force
modern engines need by compensating for variations in thread
friction. It's based on the fact that a bolt has a certain
amount of elasticity. It'll stretch up to a point, then retract
to its original length when the force is relieved. Traditional
torque specs were calculated to keep fasteners in their elastic
range. With TTY, however, the bolt is tightened enough to
stretch it beyond elasticity to what's called the "yield
point." It stays stretched and won't come back to its
original length when loosened. Once this point is reached,
further torquing won't increase clamping force very much.
The bolt replacement question is important. Ford, for one,
says new ones must be installed any time they're loosened
(some aftermarket companies supply them in the head gasket
package or as a separate product), but others tell us to reuse
the fasteners. Follow the recommendations for the engine at
hand, or you could twist one off. Better yet, since you don't
know how much or how often they've been tightened in the past,
replace them in every TTY case.
Modern high-quality aftermarket head gaskets typically require
no retorquing because their compressibility is limited and
uniform, and they're so labeled. Any gasket that doesn't specifically
state "No Retorquing" on its package or instruction
sheet, however, must receive this extra step. Run the engine
up to normal operating temperature, let it cool down, then
tighten the bolts to specs again. It's a good idea to do this
a third time after 500 miles or so.
Suction side
The head casting-to-intake manifold joint is critical because
a failure of the seal here will cause a vacuum leak, resulting
in a poor-running condition that may be difficult to diagnose.
Check both the head and manifold sealing surfaces with a straightedge
and have them machined flat if you discover warpage. As with
the head/block seam, premium gaskets of graphite or with synthetic
faces and perhaps a printed-on bead are commonly available,
and usually no retorquing is needed.
Another relatively new type, first used in Japanese V6's
but now spreading to domestics, has a metal or plastic carrier
with rubber inserts or deposited rubber beads. It's pretty
hard to make a mistake while installing this kind of gasket
because the carrier acts as a positive stop so the rubber
is automatically compressed to the proper degree. Never use
sealant on these. On "V"-type engines, be very careful
of the little endstrip gaskets, if present. They're easy to
dislodge while you're lowering that big casting into place.
Glue them down, but don't overdo the adhesive.
Intake manifold bolts should be tightened in a pattern just
like those that hold the head down. If you can't find the
info, work from the center outward. Be careful not to over-torque.
Not only will this save you from the disaster of a snapped
bolt, but on the 5L/302 Ford V8 it'll help avoid a blown head
gasket. It seems that the wedging action of the manifold coming
down can actually push the heads outward enough to reduce
clamping force and ruin the integrity of the seam (a bulletin
tells us to increase the torque on the upper row of head bolts
to 80 ft. lbs. as a preventative measure).
Umbrellas and squeegees
Failed valve stem seals are by far the most common cause
of oil burning, so you need a good understanding of these
little parts. First off, in cases where the guides are in
bad shape, new seals will be no more than a temporary measure
a band-aid on a bullet wound.
Although plenty of people still believe these oil barriers
are only important on intakes, it's been proved that a considerable
amount of liquid lube can be drawn past exhaust valve stems
by the vacuum pulses present between the "slugs"
of waste gases. One study showed that consumption jumped from
2,100 to 450 miles per quart when the exhaust valve seals
were removed.
There are two basic designs, the deflector or umbrella type,
and the positive variety. The former fits tightly on the valve
stem and rides up and down with it, while the latter is secured
to the guide boss, and the stem slides through its hole. It's
important with either type to use one of those little plastic
installation sleeves so you don't tear up the seal as you
slide it over the keeper grooves.
Seeping and weeping
Leaks at what's variously called the valve, rocker, or cam
cover and the oil pan not only increase oil consumption, they
also make the engine very untidy.
Get the flange as clean as possible, then straighten any
bends or distorted bolt holes by hammering them against a
piece of steel bar stock. Glue the gasket in position with
a little adhesive. Tightening the bolts too much will tend
to split cork/rubber and over-stretch plain rubber. The amount
of force achievable with a screwdriver handle is usually sufficient.
The thick molded rubber valve cover gasket is a nice improvement.
It won't fight you during installation since it stays flat
in its groove, and its resiliency makes leaks unlikely. Then
there's the kind of cover that comes with a rubber gasket
deposited directly on it, as first seen on Fords and Jeeps.
It's great on the assembly line, but what are you supposed
to do in the field? Well, if the rubber isn't damaged, you
can just reuse it. Even if it's got a few nicks or small gaps
in it, you can fix it with RTV. But if the rubber is really
trashed from heat or brutal removal methods, the car companies
tell you to buy a whole new cover assembly. That seems almost
immorally wasteful, so the aftermarket is developing alternatives.
Trends in oil pan sealing are toward metal grommets in the
bolt holes or shouldered fasteners to guard against over-torquing,
and one-piece rubber gaskets.
Typically, paper-like gaskets are used for timing chain covers,
thermostat housings, etc. These require the application of
a good sealant and careful torquing.
The news in crank seals is that a combination of harder or
more abrasive modern materials and a tighter fit seems to
be causing increased shaft grooving. So, it's fortunate that
the aftermarket offers those thin repair sleeves.
Goo
Chemical sealers (sometimes referred to as "formed-in-place
gaskets") deserve some space. RTV (Room Temperature Vulcanizing)
silicone rubber is by far the most common. While it's been
used on the assembly line and can work well for a repair,
it's the opinion of most seasoned engine builders that a regular
gasket is the best choice during service (providing one is
available). Chemicals require extra cleanliness, and the bead
you apply is apt to be wiped off the flange during part positioning.
On the other hand, there are cases where the substitution
of RTV can cure a problem, such as sealing up the ends of
the intake manifold on a small-block Chevy.
Odd ends
I'll conclude with some miscellaneous tips:
- Don't let steam from the tailpipe on a test drive fool
you into thinking that head gasket you just installed didn't
work. It can take a lot of driving to boil the coolant out
of an exhaust system.
- Localized overheating and air pockets are big head gasket
enemies, so you may squander all the care you took putting
that engine together if you neglect the cooling system.
- Abnormal combustion, such as detonation, preignition,
and lean running, can burn through a head gasket's armor
and cause it to fail. Make sure you correct such conditions
before staking your reputation on that head sealing job.
- On joints that retain oil, even excellent gaskets and
procedures can be defeated by excessive crankcase pressure.
Check PCV function.
- Spacer shims are available in the aftermarket to make
up for heavy machining of a head's sealing surface. They'll
bring the compression ratio back to stock and help correct
head/manifold alignment on "V" engines.
- Don't use RTV or other slippery sealants on those rubber
semi circular plugs you'll find at the ends of some valve
covers because it may allow them to slide and pop out of
place. Ditto for most other rubber sealing components. Apply
just a dab at the corners.
|
