Controlling your Discharges
Ryne C. Allen
ESD Systems
Desco Industries, Inc.
19 Brigham St., Unit#9
Marlboro, MA 01752-3170
Email: Ryne C. Allen" <ryne@esdsystems.com>
I INTRODUCTION
The control of electrostatic
discharge is an important aspect in the manufacturing, assembling
and repairing of devices that employ electronics. Electrostatic
discharges can damage an electronic component at any stage
of its production or application if not controlled. The
primary method of control is to ground (or bring
to the same potential) all conductors that come in contact
or near proximity to the electronic device(s). These conductors
include humans, tools, ESD mats, other electronic devices,
boards, connectors, packaging, etc.
There are other components
to a good ESD Control program including, removal of unnecessary
insulators, shielding, ionization, environmental controls,
training, education and top down compliance. This paper
will talk about controlling discharges to a grounded ESD
mat on a workstation.
Of specific interest in controlling
an electrostatic discharge is the time rate of the discharge.
A discharge will occur much quicker in/on a conductor with
a surface resistance of 102 Ohms than in a conductor
with a surface resistance of 109 Ohms. How fast
or slow should the controlled discharge be? Understanding
the importance of discharge times will help you choose the
right ESD control materials in building, maintaining, or
auditing your own ESD Safe workbench(es).
The upper and lower boundaries
of an ESD safe discharge rate are determined by the application
and materials used. To limit the discussion, the potential
energy sourced from the Human Body Model (HBM), [refer to
ANSI EOS/ESD S5.1-1993], is applied into an ElectroStatic
Discharge Sensitive (ESDS) work area or ESD mat.
II BODY & MOVEMENT
You should be familiar with
the timing of the human body’s movements relative to handling
or working near ESDS devices to have a handle on the upper
limit of the controlled discharge. To reduce the likelihood
of an operator discharging onto an ESDS device, they should
drain any charges before bringing an ESDS device in contact
with themselves or another conductor, whether floating or
grounded.
Table I
Movement times (averaged) from
typical operations.
|
|
Reaching
|
Grabbing
|
Lifting
|
Relocation
|
Landing
|
|
Time (ms)
|
455
|
153
|
231
|
924
|
247
|
|
Std. Dev. (ms)
|
48
|
11
|
61
|
137
|
73
|
Table I depicts averaged times,
in milliseconds for the handing of tools or devices at a
work bench with a corresponding standard deviation in millseconds.
The shortest time of 153 ms, or worst case, is the time
that we will design our ESDS workbench table top with. You
want to be sure that your device is fully discharged well
before the 153 ms landing time. A good rule of thumb would
be to engineer a x2 safety factor. Therefore your device
should be fully discharged before reaching 76.5 ms (76.5
ms x 2 = 153 ms). The time constraint of 76.5 ms for body
movement defines the upper boundary of the controlled discharge
rate (not including the standard deviation of 11 ms).
III ENERGY CONSIDERATIONS
Table II
Typical Discharge times [t=R*C*ln(V/V0)]
for an RC circuit
where C=200 pF and V0=249
Volts
|
R
|
102
W
|
2.2x103
W
|
106
W
|
107
W
|
8.3x107
W
|
108
W
|
109
W
|
1011
W
|
|
Time
|
92 ns
|
2 m s
|
920 m s
|
9.2 ms
|
76.5 ms
|
92 ms
|
920 ms
|
92 s
|
Table II shows calculated discharge
rates for the human body model onto an ESD grounded mat
with surface-to-ground (RTG) resistances from 102
to 1011 Ohms. The more conductive the ESD mat
on the workbench is, the faster the discharge, but there
is another consideration too.
How fast is too fast? When
does the discharge energy at any given time reach a critical
level that can damage a semiconductor? The answer depends
on several variables relative to the semiconductor’s construction
such as line spacing, composition, density, packaging, etcetra,
all leading to an ESD component classification [refer to
table I in the ANSI EOS/ESD S5.1-1993 and the manufactures’
device specifications].
For simplicity’s sake assume
the worst case, class 0, which has a 0 to 249 Volt tolerance.
Applying the HBM, a conservative worst case capacitance
would be 200 pF, twice that of the HBM and resistance of
10KW . Therefore the maximum power (P) level based on Ohm’s
Law is P=V2/R (J/s) and the worst case HBM is
((249)2/10K)=6.2 Watts or Joules per second (Js-1).
The maximum energy (E) stored
in a worst case HBM capacitance (C) of 200 pF and at a maximum
voltage (V) of 249 Volts, (using E=1/2 CV2),
yields 6.2 m J. The next concern is to relate energy to
time. The time constant (t ) is the measure of the length
in time, in a natural response system, for the discharge
current to die down to a negligible value (assume 1% of
the original signal). For an RC circuit, the time constant
(t ) is equivalent to the multiple of the equivalent resistance
and capacitance. In this case, the time constant (t ) of
our RC circuit is (10KW )(200pF) or t = 2 m s. Discharging
this energy upon touching a conductor at zero volts yields
a current, (using I=P/V), of (6.2Js-1)/(249V)
or 24.8 mA. To avoid damaging a class 0 ESDS device, the
discharge current must be below 24.8 mA. Engineering in
a "2x" safety factor, the maximum discharge current would
be 12.4 mA. To maintain a discharge current below 12.4 mA,
we need to look at our grounding equipment on the ESDS workbench.
Table III
Discharge currents from a 6.2
m J lossless energy source
(with C=200pF & V=249V)
dependent on the discharge time.
|
Current
|
24,900 A
|
24.9 A
|
12.4 mA
|
2.49 mA
|
249 m A
|
24.9 m A
|
2.49 m A
|
249 nA
|
|
Time (s)
|
1x10-12
|
1x10-9
|
2.01x10-6
|
1x10-5
|
1x10-4
|
1x10-3
|
1x10-2
|
1x10-1
|
The rate at which 6.2 m J of
energy discharges is very important. To fast a discharge
will lead to an ESD Event, which can electromagnetically
be measured using a simple loop antenna attached to a high
impedance input of a high-speed storage scope. The faster
the discharge the greater the discharge current becomes
as well as the emf (electromotive force)
on the loop antenna from the EMI (ElectroMagnetic Interference).
Table III depicts the discharge current for 6.2 m J at varying
discharge times. We are assuming lossless conditions during
the discharge for worst case. For our example, to keep the
discharge current below 12.4 mA, the discharge rate [from
Table III] must be no quicker than 2.01 m s. This energy-based-time
constraint forms the lower boundary of the controlled discharge
rate.
IV MAT MATERIALS
The upper and lower boundary
of our controlled discharge rate are now defined and can
be used to help in choosing the correct ESD mat for an ESDS
workstation. ESD mat materials come in many variations.
In general, mats are either made from vinyl or rubber material
and can be homogeneous or multi-layered. Rubber mats, in
general, have good chemical and heat resistance but vinyl
tends to be more cost effective. The electrical properties
of an ESD mat are important to know in controlling the electrostatic
discharge.
An ESD mat will be either electrically
conductive or dissipative. Both terms mean that the mat
will conduct a charge when grounded. The difference in the
terms is defined by the materials resistance, which effects
the speed of the discharge. By definition [ESD ADV1.0-1994]
a conductive material has a surface resistivity of less
than 1x105 W /sq and a dissipative material is
greater than 1x105 W /sq but less than 1x1012
W /sq. Anything with a surface resistivity greater than
1x1012 W /sq is considered insulative and will
essentially not conduct charges.
Back to our example. If the
maximum discharge current of our ESDS device is 12.4 mA,
then the discharge time based on energy must be slower than
2.01 m s and based on body movement must be faster than
76.5 ms. Using the discharge times from Table II and assuming
that the mat has a negligible capacitance relative to the
HBM, then the mat resistance must be greater than 2.2x103
W or 2.2x104 W /sq and less than 8.3x107
W or 8.3x108 W /sq. In other words, a very conductive
mat for some applications may be to quick to discharge and
yield more dangerous ESD events whether properly grounded
or not.

Graph I
Graph I shows the natural response
of a 249 Volt discharge in an RC circuit using a capacitance
of 200 pF (HBM) into resistances (mat) of 104,
105, and 106 Ohms. The natural response
of the104 W curve is below 1% of its’ initial
voltage in less than 10 m s where the 106 W curve
takes less than 1ms to discharge to less than 1% (V<2.49
V) of its initial value (V0=249 V).
V GROUND STRAPS
Another defense, and the most
common method, to reduce the risk of creating an ESD event
is wearing a grounded wrist strap at the workstation. The
wrist strap connects the skin (a large conductor) to a common
potential (usually power ground). Properly worn, the wrist
strap should fit snugly, making proper contact with the
skin, to reduce contact resistance. Refer to the manufactures
specifications and instructions of the wrist strap.
The wrist strap, since it is
connected to ground, will quickly discharge any charge the
body either generates through tribocharging or becomes exposed
to through induction. Anytime the body directly touches
a charged conductor, a discharge will occur because the
body is at a different potential (0 Volts). Controlling
this discharge is important if the conductor is an ESDS
device and in minimizing induced charges through EMI onto
nearby ungrounded ESDS devices.
The electrical properties of
the skin of an operator can have a wide range in both resistance
and capacitance depending on several variables. An operator’s
hand touching a charged device will initiate a discharge
at the rate of the time constant of the skin before including
the RL properties of the wrist strap. To reduce the potential
of an unsafe discharge from a device to a very conductive
operator, adding resistance to the operator at the interface
from skin to device may be necessary. Some solutions are
static dissipative gloves or finger cots, which if worn
properly, may add from 1 to 10 MW to the RC circuit of the
skin. This in turn slows down the discharge rate to well
over 2 m s.
VI CONCLUSION
The upper and lower boundaries
of a safe discharge rate are determined by the application
and materials used. The movements of the operator define
our upper boundary and the max energy, as defined by the
ESDS component classification, dictates our lower boundary.
We want to design an ESDS workbench to control the discharge
rate (via the circuit’s time constant) of our grounded or
conductive materials within these limits.
For the HBM, and a class 0
device, the materials chosen for a safe ESD workbench should
have electrical properties to support discharge rates between
2 m s and 76.5 ms. These discharge rates, using worst case
assumptions, equate to an ESD mat surface with an RTG (Resistance
To Groundable point) between 2.2x103 W and 8.3x107
W . This controlled discharge rate window will vary depending
on the class of semiconductor components used (class 0 to
class 3B) and the properties of production resources used
(human vs. automated).
The numbers calculated were
based on assumptions used to simplify the explanation of
the material. Real world applications are much more complex
and require a more detailed analysis, which was beyond the
scope of this paper.