How to: Make Your Own Muffler Heat Shields (for just $4 each)
as published in British V8 Magazine, Volume XVI Issue 1, May 2008
by: Curtis Jacobson
Notice: It will take more time to read this article then to do the project herein described.
Call it "entertainment."
Overview
Muffler heat shields are often used to keep the car's interior comfortable. In some
cases, heat shields are used to protect parts of the fuel, electrical, brake, or
suspension systems. In some cases they also help to reduce noise transmission.
Three basic physical properties determine how well heat shields function: reflectivity,
emissivity, and conductivity. Heat shields work primarily by reflecting heat downward
and away from (or back into) the exhaust system. Since heat shields aren't perfect reflectors,
and since air conducts some heat (i.e. we don't live in a vacuum,) they inevitably get
hot with the rest of the exhaust system. Emissivity and conductivity properties determine
how much heat is transferred through an air gap or through physical contact respectively.
Ideally, there should be an air gap between the muffler and the heat shield, and also a
second air gap between the heat shield and whatever it's protecting - and these air gaps
should be as large as feasible. Usually it's also preferable to mount heat shields to
the exhaust system because this reduces or eliminates metal-to-metal conduction of heat
and noise into the body of the car. When retrofitting heat shields onto older cars, we
usually have to make compromises due to space and cost constraints. This article is
intended to help you select a heat shield, or to help you construct your own.
These new muffler heat shields are mounted right under my seats.
The Test Vehicle
This spring, I decided to fit muffler heat shields on my MGB GT V8. However, I couldn't find a suitable "store bought" solution. The options all seemed extravagantly priced, and most of them wouldn't work well with my small mufflers or restricted packaging. Okay, honestly, I just thought it might be entertaining to design and construct heat shields from scratch. I budgeted ten dollars and three hours to the job - and I came in under budget on both counts. It was a satisfying experience, so I thought I'd pass along a report about it.
My little beast was converted to V8 power almost twenty years ago. Back then, I wasn't
remotely concerned about heat shields. When we started the conversion, my
design objectives were more directly performance related: light weight, low ride height,
low polar moment of inertia (i.e. heavy things like mufflers should be as close to centered
as feasible), etc. I didn't worry at the time that driver comfort might
negatively affect driver performance or fatigue.
The design we came up with involved pocketing the mufflers up under the seats. As part of
the conversion, we mounted the seats further back than the stock rails would go, and
strengthened their mountings. The seats were mounted on ladder frames constructed of
1" x 1" x 0.06" steel tubing with the ladder ends welded to the "heel-board" and the
under-floor crossmember respectively. The ladders were skinned with 0.030" steel.
Reinforced cut-outs in the under-floor crossmember facilitated getting the exhaust system
right up under the floorboards. The exhaust and tail pipes were cut short for a
significant weight savings. (This made more room for a Panhard rod, etc.) Since cost
and weight were important considerations, I picked the smallest, lightest, cheapest
performance mufflers I could find: "Thrush Turbo" mufflers at about $19 each. I actually
like the way they sound; it makes me feel 16 again.
My only complaint has been that the seats get uncomfortably hot on long trips.
How Do Heat Shields Work?
A very thin piece of aluminum can make a pretty good heat shield because (a)
it reflects thermal radiation very well and also because (b) it emits (or
"gives off") heat very inefficiently.
Remember how your momma covered the Thanksgiving turkey with aluminum foil?
Foil reflected the oven's thermal radiation, and the turkey cooked evenly from
top to bottom. As a radiant barrier, aluminum foil performs so well that it
actually needs to be removed, briefly, for the turkey's skin to brown nicely.
Used as a radiant barrier (i.e. a "reflector"), a cut or tear isn't a big deal
so long as it doesn't turn into a gaping hole. After letting the turkey's skin
brown, your mom probably removed the bird from the oven and replaced the loose
foil covering. Since aluminum doesn't emit much heat, the turkey stayed hot until
her praline sweet potatoes and delicious butter rolls were also ready.
As a reflector, it shouldn't surprise you that a shiny surface outperforms an
unpolished surface and that a clean surface vastly outperforms a dirty one.
Anodized aluminum generally outperforms painted aluminum.
The concept of high reflectivity is more intuitive than the concept of low emissivity.
To understand emissivity, think of a light bulb. When turned "on", a light bulb emits
light. If you paint the bulb black, there's no light emission. You can put your hand
relatively near a black-painted light bulb and you'll feel relatively little heat on
your skin. Just don't touch the painted light bulb, or you'll get a hard lesson in
thermal conductivity! That's a different physical property.
Like most metals (and light bulb glass) aluminum is a very good thermal conductor.
In other words, heat travels easily from one surface of a bare aluminum heat shield
to the other. It's highly preferable to have an air gap on both sides of the heat
shield because dry air isn't a very good thermal conductor. To the extent that air
does heat up along the surfaces of the heat shield, we rely on airflow to carry it
away. Whenever heat travels with airflow - and particularly when hot air naturally
rises - we call the resulting dissipation "convection". If the air around the heat
shield is stagnant or if there are air leaks into the car's interior, heat will
eventually travel to where we don't want it. A fan, duct, louver or vent might be
helpful in some exceptional cases. However, there's an alternative strategy to
consider first: insulation.
By laminating a less thermally conductive material (i.e. "insulation") to a muffler heat
shield, or by sandwiching insulation between layers, it's feasible to achieve a
significant temperature drop from one outside surface of the shield to the other.
Various insulation materials are appropriate for this purpose, including especially
fiberglass and ceramic fiber mats or batting. (To be thorough, it should be pointed
out that "fiberglass" is a family of materials of various chemical compositions.
Fiberglass materials have a range of physical properties. Some fiberglass media may
decompose into powder if subjected to heat and vibration.) Regardless of the
insulation material, air pockets entrapped by the media actually contribute quite a
lot to its efficacy as insulation.
Enjoying this article? Our magazine is funded through the generous support of readers like you!
To contribute to our operating budget, please click here and follow the instructions.
(Suggested contribution is twenty bucks per year. Feel free to give more!)
Ready-Made Alternatives
When I became interested in installing muffler heat shields, I started my research by
looking through my scrap-metal bin. I found what I was looking for: a nifty "double
corrugated" aluminum heat shield that I removed years ago from a large truck's diesel
engine. The big selling point of double corrugated aluminum for heat shields is it's
easy to form into elaborate shapes. It's perfect, for example if you want to wrap your
shield neatly around a turbocharger or follow the contours of your firewall. The
corrugations also add rigidity and make the shield less likely to reverberate or
rattle. This piece might have made a good muffler heat shield, but it wasn't a big
enough scrap to divide over both my mufflers. Not finding anything more suitable,
I continued by poking around online for ready-made heat shield materials.
An Australian company called "ACL" makes similarly double corrugated aluminum material,
and you can buy it on eBay. One 27.5" by 11.4" (700mm by 290mm) rectangle of "Race Series
HR140-GTR" material costs $48.50 (Australian) plus shipping and handling from Australia.
(The seller writes: "Send me a postcode for an accurate freight cost to your area." How
quaint!) If you have a turbo to gift-wrap ACL's material might be a good deal.
Flowmaster sells simple heat shield for their "50 series" mufflers (part number 51017,
typical price $37.76 per heat shield plus shipping). The Flowmaster muffler heat
shields are stamped out of plain aluminum sheet, and they mount to the muffler with
stainless steel tie straps. Even if their price wasn't ridiculous, since they measure
17" by 10" the Flowmaster shields are too large for my little turbo mufflers.
Trimming them down isn't a viable option because of how they're grooved for tie strap
installation.
A company called "Heatshield Products" makes several interesting products for muffler
heat shielding. Their "Reflect-a-Shield" product consists of a 1/4" thick ceramic fiber
pad sandwiched between heavy duty aluminum foil and bound together with a stainless
steel mesh tape at the edges. The company claims that stainless steel mesh bound edges
provide greater flexibility for a more rugged product. They're rated for 400°F contact
and 1000°F continuous. They further explain: "As with any aluminum insulation
product, it is not designed for direct contact with exhaust systems. This heat shield
is intended for wrapping around electrical boxes, air conditioning components, or
used as a catalytic converter and muffler heat shield. For my purpose however,
Reflect-a-Shield doesn't seem appropriate because it isn't rigid and therefore can't
easily be installed with an air gap on both sides. Reflect-a-Shield comes in several sizes;
for reference, part number 100614 is a single 6" by 14" piece costing $18.52 plus shipping.
Heatshield Products makes another product called "Inferno-Shield" which uses a thicker
layer of metal on the outside so that it can be bent or formed into shape. Inferno-Shield
comes in three grades: aluminum (900°F continuous), stainless steel (1800°F
continuous) and inconel (2200°F continuous). All three types have an inner ceramic
fiber pad. For reference, part number 110614 is a single 6" by 14" aluminum-based shield
that costs $22.31 plus shipping.
If that's not exotic enough for you, a company named "D & H Heat Technology, Inc."
offers "Koolmat Heat Master Shields" which consist of "NASA insulation" sandwiched
between two textured inconel panels grommeted for easy installation with stainless
steel tie straps. Three sizes are available... but they all cost $1.08 per square inch
plus shipping. If you have NASA's budget, go for it! Instead, I went to my local home center
store to look for something (much) more economical.
How to Make Your Own Muffler Heat Shields
My local big-box home center store sells 8" by 18" by 0.025" thick aluminum
rectangles for $17.27 each, and 12" by 18" by 0.025" rectangles for $21.86. I
shouldn't have to tell you this, but that's highway robbery. Maybe it's my Scottish
heritage showing, but I just don't think a heat shield should cost as much as a
muffler. However, I found something I think will actually perform better just
two aisles away. 10" x 14" aluminum "Beware of Dog" signs for just $1.47 each.
Now that's more like it!
Now admittedly, a 0.0125" thick Beware of Dog sign isn't as rigid as we'd like
for a muffler heat shield - but we can easily fix that by folding over the edges
and by laminating two layers together, preferably with insulation sandwiched
in between. Our laminated heat shield will be more effective than a single-layer
metal heat shield and it will be sufficiently rigid. The back sides of the signs
are anodized, which will help both top and bottom of the fabricated assembly
stay clean and bright over time. Bonus: the flanged edges and the insulation
will help with sound attenuation.
Aluminum this thin is very easy to work with. You could probably cut it with scissors
or a razor knife. I used aviation snips, and a Roper-Whitney style hand punch for
internal radii. I bought my actual Roper-Whitney punch used and its larger dies were
missing, so for this project I used an old Taiwanese-made knock-off, as shown here. The
real deal is a nicer tool, of course. I strongly prefer to punch radiused inside corners
because they're neat and far less likely to ever crack, which is always "good form" even
though stress ("fatigue") cracks aren't a real big concern with this
project.
Thin aluminum is also easy to bend. I bent the four 90-degree flanges shown at left (below)
over the straight edge of my welding table. After each bend, I flipped the sign over and
folded the flange down flat. This completed the "top" of the (first) muffler shield.
The center layer of my low-buck muffler heat shield is fiberglass. I had several kinds
to choose from in my supply cabinet. Non-woven fiberglass mat, like you'd use to repair
a boat, is perfect. Fiberglass cloth, or even home insulation would work too. (The
aluminum cover will compress bat-type insulation down.) As mentioned above, all these
"glasses" actually have different physical properties, and their fibers are treated
with different "sizing" chemicals that affect how they behave (e.g. how easily "wettable"
they are by resins), but I believe that will have minimal effect here. The aluminum
housing will support the fiberglass, keep it clean and dry, and help it resist breaking
down over time. If you have any doubt at all about your insulation material's composition,
try to ignite it. If it burns, it isn't fiberglass. If it melts easily or smokes, find a
different source of fiberglass.
Incidentally, as an afterthought, I went back out to my garage and applied a blowtorch
flame to both a spare painted-aluminum sign and also to some fiberglass insulation. A
blowtorch flame reaches about 2375°F (1300°C) - which is WAY HOTTER than
an automotive muffler. In the direct flow of the flame, paint on the sign didn't smoke
or ignite. The fiberglass mat appeared undamaged except where loose fibers protruded -
and they simply curled up as they melted.
Now we're ready to make the aluminum "bottom" of the heat shield. A word here about bend
radius: whenever you bend any sheetmetal you should make sure to include a generous internal
radius to avoid stress cracks. Honestly, I didn't worry about this when making the "top"
of the heat shield, but my design did make a simple provision for a radiused bend for
the flanges on the "bottom" layer. I simply bent the bottom layer's flanges over and around
the top layer's flanges!
If you're following along and making your own heat shields as you read this, I hope you
didn't put away your handy hole punch. Now is the time to punch mounting holes in the
heat shields. Your punch will have no problem at all cutting right through the whole
lamination neatly.
Although it's technically better to mount the heat shield to the exhaust system instead
of to the floorboard, so heat and sound aren't conducted into the cabin through the
fixings, I decided my design would be a little more serviceable with the heat shields
mounted to the flat sheetmetal of my "muffler pockets". Instead of drilling holes in the
floor for bolts though, I welded bolt heads to the bottom surface of the steel. I put
short spacer tubes above the heat shields to space them away from the floor, and secured
them in place with four quarter inch nuts.
Done! And in far less time than it's taken to type this article.
How Well Do They Work?
I'm embarassed to say that I don't have a quantitative answer yet. Actually, I might not
even have a qualitative answer until I take my next road trip - to attend the
British V8 2008 meet. Ask me when you see me! I've already turned my attention to
completing installation of new parking brake cables, revised brake lines, new rear
suspension bushings... and of course to completing this edition of British V8.
Perhaps I'll revisit this article later and add an addendum. A hand-held infrared-sensing
non-contact thermometer would do a good job of measuring floorboard temperature (above the
mufflers) and a reasonably scientific test procedure could be designed by simply temporarily
removing one heat shield to facilitate simultaneous evaluation of with-and-without
conditions.
I'd be particularly curious to do A-B comparisons to determine how much advantage insulated
heat shields provides over solid metal or hollow-core heat shields of similar geometry.
Measuring the temperature of the heat shields themselves, front and back, would be interesting
too.
