The Creep of Solder

In order to permanently change the shape of a material, it is necessary to apply a stress which
exceeds a critical value called the yield stress. Permanent deformation is called plastic
deformation. If the applied stress is less than the yield stress then the material only deforms
elastically, so that any strain is recovered on the removal of the stress; e.g. when a rubber
band is stretched and released.

However, there are circumstances where the diffusion of atoms can lead to plastic
deformation even when the applied stress is less than the yield stress. Such deformation is
called creep, where the plastic strain is dependent not only on the stress but also on time.

Creep in solids occurs at "high" temperatures where atoms are mobile. However, a
temperature which is high for one material may not be for another. For example, 550 K is a
temperature at which solid-state diffusion can occur very rapidly in lead (which melts at
600 K), whereas at the same temperature, atoms are hardly mobile in iron which melts at
1810 K. To avoid this difficulty, we define a homologous temperature, which is the actual
temperature divided by the melting temperature (in Kelvin). Metals and ceramics tend to
creep rapidly when the homologous temperature is greater than about 0.4.

Creep tests are usually carried out by loading a sample and observing the development of
strain as a function of time, for a given temperature and stress. The measured strain (e) is then
plotted against time (t) as illustrated in Fig.1. The curve consists of three portions: the
primary and tertiary portions are relatively small transients which will be neglected in this
practical. The central steady-state part is prolonged and the most important in practice,
because it determines how long the material can be used safely. In the steady-state regime,
the rate at which strain occurs (i.e. the slope of the e – t curve) is given by
de
= Cs n (1)
dt

where s is the stress, C is a proportionality constant and n is known as the stress-exponent. In

Tertiary
Steady-state Creep
creep
Strain

Primary
Creep

Time

Fig.1: Creep deformation curve illustrating the steady-state creep regime

The Nature of Lead - Tin Solder

Solder is a metallic alloy which is used extensively in the electronics industry to make
electrically conducting connections between wires and components such as integrated
ET00
Dem. 38 –2– GD2

circuits, resistors, capacitors etc. Components like these can be fatally damaged by excessive
heat, so that any material used in making soldered connections must have a low melting
temperature.

Both lead and tin have low melting temperatures (Fig.2). However, a Pb–Sn mixture
containing 61.5 wt.% tin has a eutectic composition and, therefore, the minimum melting
temperature of all combinations of lead and tin. This "60 – 40" mixture melts at 183˚C. The
mixture also has the advantage that it solidifies at the eutectic temperature, rather than over a
range of temperatures. This means that a mechanically rigid joint is achieved very quickly.
Both of these features are a great advantage in making electrical connections to sensitive
components.

For the purposes of the present experiment where we are interested in creep, room
temperature (about 300 K) corresponds to a homologous temperature of 0.66. This means that
with a suitable design of experimental apparatus, it is possible to observe creep deformation
in a time scale which is realistic for a Part IA practical!

Fig.2: The lead - tin phase diagram

SINCE YOU WILL BE HANDLING LEAD,

Creep tests are normally carried out using tensile specimens of the kind illustrated in Fig.3a.
The sample is loaded whilst in a furnace at a controlled temperature; the deformation (hence
strain) and time are recorded continuously. It is necessary to do many tests like this in order
to obtain the strain rate as a function of stress.

An alternative method is to use one sample but with a gradient of stress. Fig.3b illustrates this
method in which a coil made out of the material of interest is exposed to the right
temperature. The stress arises from the weight of the sample itself. This means that there is
no stress on the lowest part of the coil, a maximum stress at the highest part, and a
proportional stress in all intermediate positions.

w
s
D

(a) (b)

Fig.3: (a) Standard tensile specimen. (b) Coiled specimen.

If the turnings in each coil are numbered N beginning with the lowest turning and ending at
the top, then the shear stress in each turning varies from zero at its centre to a maximum
value t at its surface where
ND2
t µ w . (2)

where D is the diameter of the coil and w the diameter of the wire which makes the coil
(Fig.3b).

The deflection of the coil (a measure of the shear strain) when loaded can be measured in
terms of the spacing s between adjacent turns (Fig.4). It is found that the local shear strain g
is given by:
sw
g µ 2. (3)
!D

Harry Bhdeshia, Mike Ashby

N=5 s5

N=4

s2
N=2

N=0

Fig.4: A deformed coil

On allowing the coil to creep, the spacing of the turns at the top (where the weight of the
underlying coils is largest) will increase most (Fig.4). This spacing, divided by the time, is a
measure of the strain rate for that particular value of stress, so that the strain rate is given by
dg sw
dt µ !tD2 (4)

It follows that a single experiment can give all the data required for a plot of the strain rate
versus the stress.

Measurement of Creep Data

The main equipment needed is the "dumbell" illustrated in Fig.5. You are provided with two
samples of 60/40 solder, with two different diameters (0.71 and 1.22 mm). Wind each of
these into a tight-fitting coil around the mandrel provided, but do not use excessive force
during this operation. Begin the experiment using the coil made from the thinner solder and
then repeat all the measurements using the other, thicker, solder wire.

1. Remove the cover of the dumbell, and place one coil onto the central section. Taking
care to keep the dumbell in a horizontal position, replace the cover and thread the end of the
wire nearest to it through the hole in the central section (Fig.5) so that the position of the end
of the coil is secured. Keep the dumbell in a horizontal position until you are ready to make
the measurements described below. The coil does not creep whilst in a horizontal position,
and the elastic component of the strain is also removed.

2. You may begin the experiment by starting the stop-watch and turning the dumbell into a
vertical position, with the loose-fitting cover at the top. The coil will steadily unwind.
The experiment can be stopped when the lowest turning of the coil reaches within about
1 cm of the base. Turn the dumbell into a horizontal position as soon as the experiment
is stopped.

3. Take a double sheet of "duplicate" paper and place it so that the pressure-sensitive sheet
is in contact with the whole length of the coil. Rub the top sheet gently (to avoid
altering the coil spacing) with a soft pencil (or the mandrel itself) to make an
impression of the coil on the pressure-sensitive sheet. The rubbing direction should be
perpendicular to the mandrel (parallel to the coils). Remove the paper and measure the
turn spacing s as a function of the turn number N, noting that N = 0 at the bottom of the
coil (Fig.4). See section "Analysis of results" before measuring since entering data
directly into a table with appropriate columns which facilitate subsequent analysis.
4. Turn the dumbell, with the distorted coil into a vertical position, but in an upside down
orientation in order to reverse the creep. Note the time taken for the coil to return into
its original configuration. Steady-state creep occurs at a constant strain rate. A constant
strain rate cannot be obtained if the material work hardens during deformation. The
times taken for forward and reverse deformation should therefore be equal.

5. Once again, turn the dumbell upside down, but this time blow warm air across the coil
using the hair-dryer provided. Do not attempt to make any measurements, simply
observe what happens. Note the large effect of temperature on the creep rate.

Dumbell in horizontal
position

Fixed
end
Gentle hole
push fit

Mandrel, 16 mm
diameter

Fig.5: The creep test equipment

Analysis of the Results

The following analysis should be carried out on the data from both coils. Consider what data
are needed for the final graphs and draw up an appropriate table with necessary columns
before commencing measurements and analysis of results.

1. Obtain a measure of the strain rate by multiplying each value of s by w/t and a measure
of the stress as a function of position in the coil by multiplying each value of N by 1/w,

2. Plot a graph of ln(sw/t) versus ln(N/w).

3. Assuming that the errors in the measurement of w and of t are small compared with
those in your measurements of N and s, deduce expressions for d s and d N. Hence
estimate the error in s and in N and add error bars to your graphs.

4. Is the observed behaviour consistent with steady-state creep? Obtain a value of the
stress exponent n.

5. Would you expect pure lead or pure tin to creep at a higher rate than 60/40 solder in an
identical environment?

Harry Bhdeshia, Mike Ashby