Epoxy's outstanding adhesion to most materials is key to JOINING things into very secure groupings. Over 2000 pounds of adhesion per square inch are possible.
No. It is yellowish - not very attractive - in its natural state. I generally add color to epoxy. Adding a little color gives a tint, adding a lot gives an opaque color. Dyes of many colors are available.
About $30 per quart, so you wouldn't, say, build a wall out of epoxy bricks.
C A S E S T U D Y - C A B L E A N C H O R A G E
The need was for a compact, position adjustable, self aligning cable anchorage. Off the shelf approaches, included a turnbuckle and lots of other hardware, and were deemed too bulky and visually cluttered.
Thanks to the adhesive power of epoxy, the resulting devices (left: cross section, Right: photo) are simple, strong and functional.
In this design, a 1/2-13 threaded stud is drilled out (big one end, small the other) and in it the cable's splayed-out wires get immersed in epoxy.
The spherical "open acorn nut" forms the bearing surface, and allows the unit to self-align several degrees by pivoting.
The feasibility of the concept hinged on one key question: how well will the wires be gripped in the epoxy? Published data were confusing, not quite applicable, and figures of over 2000psi shear strength seemed hard to believe. So I did some actual tests.
The cable to be used was Type 302 stainless 3/32" 1 x 7 (7 strands of .032" diameter, McMaster Carr #3458T16). The individual wires were tested to pull-out or failure as were finished assemblies.
Here are the individual wire test specimens being made (left) and a finished test specimen.
My simple tensile test rig (left) allows more and more pulling force to be applied to a test specimen until something fails. The rig has 10:1 mechanical advantage, so 100 pounds of weights creates 1000 pounds of pull.
6 single wire tests (right) showed that the wire would slip down at 105 to 142 pounds. Specimens poured earlier (epoxy at its most fluid) with squarest wire positioning fared best. Slanted wire led to early chipping which initiated failure. Strangely, bending the wire end 90 degrees did not help.
The best performing specimen withstood an equivalent of 2840 pounds per square inch shear force - pretty amazing, and good news for the design concept.
At left a prototype anchor withstands 1234 pounds. At 1284 pounds the cable itself (rated 1200) failed.
This design has about 1 square inch of "wetted area" (total wire surface area submerged in epoxy). The cable itself breaks at 1200 pounds, at which point the wires would be experiencicing 1200psi shear force.
1200psi is well under observed adhesive strength. So the cable would break well before the wires could pull free of the epoxy mass. An even more conservative design could be achieved by making the unit longer or using 1 x 19 or 7 x 7 cable.
(left) Anchors getting ready for epoxy pour, wires get cleaned with acetone. (right) Anchors have been sealed with 5 minute epoxy (to prevent leak out) and then backfilled with Tap 314/143 epoxy.
C O N C E P T : F I N E R G R I P S B E T T E R
In the above cable anchors, 1 x 7 cable was used. How would the performance vary if a different type of cable was used? What if, say, 7 x 7 cable of equal strength was used?
In fact, the cable type matters a lot....
At left we see cross sections of 4 cable types, each with the same total area, thus (roughly) the same strength. As the number of wires increases, the total surface area of a given length of cable increases. This "wetted area" is the area of wire surface embedded in epoxy. And the GRIP is proportional to this area.
The Finer the strands of the cable, the better the adhesion grip.
So, for example, 7 x 7 cable will grip 2.64 times better than 1 x 7 cable. It turns out the grip ratio is the square root of the [# of wires ratio].
A corollary to this principle is that the "required anchorage length" gets smaller as the fibers get finer. And this applies not just to anchorages but to internal reinforcement as well: the strength of an epoxy block can dramatically improved by the addition of short glass fibers, fibers so short and fine they appear to the naked eye as a powder.
Type: SOLID, # Wires = 1
Area total = .78
Area per Wire = .78
Diameter wire = 1.00
Circumference Wire = 3.14
Total Circumference = 3.14
Wetted Area Ratio = 1.00
Type: 1 x 7, # Wires = 7
A total = .78
Area per Wire = .11
Diameter Wire = .38
Circumference Wire = 1.19
Total Circumference = 8.31
Wetted Area Ratio = 2.64
Type: 1 x 19, # Wires = 19
A total = .78
Area per Wire = .041
Diameter Wire = .229
Circumference Wire = .718
Total Circumference = 13.64
Wetted Area Ratio = 4.34
Type: 7 x 7, # Wires = 49
A total = .78
Area per Wire = .016
Diameter Wire = .142
Circumference Wire = .447
Total Circumference = 21.91
Wetted Area Ratio = 6.98
C A S E S T U D Y : F I N G E R B E N D E R
C A S E S T U D Y : B I R D B O N E
Relatively thin walled tubes are often used as beamlike structural members., especially when weight is a prime concern. This makes good sense since stress is mainly handled "towards the outside"; a solid rod is wasteful in that the center does no work. Thin walled tubes are vulnerable, however, to getting crushed or creased. In bird bones, nature has optimized thin wall tubes with the addition of criss crossing trusses called trabecula. These, in effect, give stability to the thin walls at a small weight cost.
A rough simulation of birdbone (cross section, left) can be made of thin wall aluminum or steel tube which is filled with epoxy / microspheres. The blue represents the metal, the red is epoxy, and the gray circles (depicted much larger for clarity) are glass "microspheres", actually glass balloons. Tap 314/143 epoxy with enough Tap Microspheres mixed in can have a density as low as .5g/cc - like cork but very stiff.
The epoxy remaining between the bubbles forms a truss-like volume, kindof like that depicted to the right.
The metal shell still does the structural work, but is transformed from delicate to rugged by the filling.
C O N C E P T : T H E R M A L R U N A W A Y
Epoxy generates heat as it cures, and the hotter it gets the faster it cures - creating yet more heat. This can be the formula for trouble: thermal runaway. Thermal runaway limits what you can do with epoxy, because it limits the volume of epoxy you can pour in one go. Too much volume and you'll have too much self heating. It's a very real and dramatic phenomenon, with at least ruined castings (left), and possibly acrid smoke and even fire.
Pure epoxy castings of more than a few cubic inches start to get thermal runaway. The runaway can, in marginal cases, be kept under control with fans and ice water baths (right). Embedded materials "spread out" the epoxy and help a lot, but the design may not allow this. If a large volume of pure epoxy is required, it must be poured in multiple batches, without allowing full cure between pours.
Runaway often starts in the mixing cup, and that's not good: you don't want to pour hot partially reacted epoxy into a mold because it's not as fluid and won't adhere well. If this happens, start over using a bigger diameter cup to spread out the epoxy.
Starting out with refrigerated epoxy liquids is not the solution, because reaction needs at least room temperature to get going.
Temperature, in fact, is a big part of doing quality epoxy work. Ideally, the liquids should start at about 70F and not exceed 90F for the first hour. After that you can elevate the temp with lamps to speed up the cure. A non-contact thermometer (left) is very helpful.
The "slow" catalysts have the least serious thermal runaway issues and the "fast" ones are the most troublesome.
Small pours present no problems.
C A S E S T U D Y : R E V E C T O R B L O C K S
C A S E S T U D Y : S L I P R I N G S
G O O D H O U S E K E E P I N G
A THERMOSET PLASTIC
As opposed to thermoplastics which can be re-melted, epoxy is a member of the thermoset family. Thermosets start off as 2 liquid parts, and when mixed they react - crosslinking into permanent, giant, rigid molecules. Thermosets require and create heat during the cure process, this is important for 2 reasons: curing can be speeded up by heat (or retarded by cold), and secondly, self-generated heat can lead to "thermal runaway" discussed below.
NOT FOR HIGH TEMP USE
The achille's heel of most epoxies is a limited temperature range. Some formulations loose strength at low as 120F - 145F.
Epoxy comes in many formulations, but for this discussion I am referring to TAP Marine Epoxy. This stuff is very "thin" which means it mixes easily and flows easily. It flows into molds leaving no voids, and can even be squirted through thin tubes. The ability to flow is a big part of why epoxy can do what it does - for example fully encapsulating the strands of carbon fiber yarn. The excellent adhesive poperties of epoxy are closely related to this ability to fully "wet out" on the microscopic scale.
Fillers can be added to epoxy to tailor its density, viscosity ("thickness"), strength, color, etc. Cure times can be varied from minutes to hours by the use of different hardeners and temperatures.
To construct slip rings, a very sturdy joining of components is desired because a final lathing operation (right) is involved. Epoxy "potting" (filling in around components) fits the bill nicely.
The metal components are (left) aligned and held in place with 5 minute epoxy. This setup then gets potted with Tap Marine slow epoxy, resulting in a rock-solid assembly that can be machined without trouble.
As is often the case, the epoxy plays a supporting role.
Epoxy is a tool for creative combination.
The combination may be as simple and spatially uniform as an epoxy / crushed-stone countertop or as complex and spatially defined as the giant wind turbine blades at left.
Joining materials and components in new ways can create emergent properties far beyond those of the individual parts.
Epoxy and "composites" are fun and amazing, and one of the few advanced technology areas open to all - equipment and space requirements are minimal.
Fascinating Fun Fact: nuclear reactors depend on a positive feedback concept similar to epoxy's thermal runaway. Fission of uranium generates energy which creates more fission and so on. If allowed to run away, the reactor will melt and make a mess. If carefully regulated, teetering on the brink of runaway, the reactor produces vast amounts of heat.
If enough uranium or uncured epoxy is lumped together, the reaction gets going. If it is spread out, the reaction slows. Also, filler material in epoxy or moderator rods in a nuclear reactor slow the reaction. The amount of material beyond which things get crazy is called "critical mass".
What at first seems like an impossible problem - creating a series of curved tunnels through a teflon block - is readily solved with epoxy, teflon tubing and special molds (right).
A Revector Block (described in more detail here) is a device for carrying a series of tendons around corners. In some cases there are many tendons in close proximity. Always, the paths must be gentle curves with specific placements.
The key fact here is that teflon tubing which has had its surface roughened will be held firmly in place by the epoxy. The epoxy, because it flows so well, is able to "key into" the microscopic dents in the teflon tubes.
While cured epoxy is clean and safe to handle, in its liquid uncured state it is toxic, sticky and (for some types) stinky. You should never touch it or allow it to get on things that will later be touched. A well thought out procedure for handling epoxies makes life easier (and longer).
First, get a "secondary containment vessel" (aka big plastic tub, see left) to put all primary containers in. Line it with disposable material. Second, use pumps wherever possible (right) - these make it quick and clean to dispense epoxy.
Have a plastic bag lined trash bin and lots of paper towels at the ready. Use gloves. Cover the work area with disposable material. Use disposable mixing sticks, clean up often, and feel the power of epoxy.