#

3 Protecting Your Laser Diode

For application assistance or additional information on our products or

services you can contact us at:

ILX Lightwave Corporation

31950 Frontage Road, Bozeman, MT 59771-6310
Phone: 406-556-2481  800-459-9459  Fax: 406-586-9405
Email: sales@ilxlightwave.com

To obtain contact information for our international distributors and product

repair centers or for fast access to product information, technical support,
LabVIEW drivers, and our comprehensive library of technical and
application information, visit our website at:

www.ilxlightwave.com

Copyright  2003 ILX Lightwave Corporation, All Rights Reserved Rev01.060103

Protecting Your Laser Diode
By Doug Hodgson and Bill Olsen

Introduction determine which mechanism is dominant.

It’s been said that there are two types of The primary concern of this application note,
researchers - those who have blown a laser however, is catastrophic facet damage (CFD)
diode and those who will. While that may be and latent damage resulting from electrical
an exaggeration, it’s certainly true that today’s transients or other mechanisms related to
lab envorinments pose a greater threat than handling and operating conditions.
ever to laser diodes. Furthermore, laser pro-
tection is commonly misunderstood and often To understand CFD, consider that the size of
poorly implemented. Few will dispute that laser the emitting region of a typical 5 mW laser
damage is frustrating, time-consuming, and diode is about 2 µm by 4 µm. This equates
often costly. to a normal optical power density of 625 W/
mm2. Damage can occur when the optical
Fortunately, laser damage can be avoided by energy densities at the output facet exceed
taking the appropriate preventative measures. 104 W/mm2. This may seem like a sizable
This application note examines the mecha- “margin of error” — until surface absorption
nisms of laser damage and outlines precau- effects are considered.
tions and procedures that you can take to pro-
tect your laser diode. A summarized checklist In the pumped part of the active region,
of specific actions is provided at the end of this gain is greater than loss, and the material
application note. We strongly urge you to follow is transparent. At the facet surface, how-
these recommendations and use the checklist ever, non-radiative recombination causes
to minimize your risk of laser damage. the material of the active region to become

Physics of Laser Damage Surface Reduced Increased

Facet
Recombination Population Optical
Under ideal conditions, laser diodes can at Facet Inversion Absorption
Heating

demonstrate excellent reliability, with lifetimes

exceeding 100,000 hours. However, they are Local

extremely sensitive to electro-static discharge, Bandgap

Reduction

excessive current levels, and current spikes, Loop Becomes Unstable at High Power

or transients. Symptoms of damage include

Figure 1. Thermal runaway at laser facet will result in
reduced output power, shift in threshold cur- catastrophic facet damage.
rent, change in beam divergence, difficulty
focusing to previously-attained spot sizes, and absorbing. At high energy densities, tem-
ultimately failure to lase (LED-like output only). perature rises faster than normal. This tem-
perature increase causes greater absorption
Semiconductor injection lasers can be dam- at the surface resulting in even greater tem-
aged by a variety of mechanisms. Many peratures. This thermal runaway melts the
factors, including wafer growth procedures, mirror facet and/or facet coating, resulting in
device fabrication, and operating conditions gross crystal faults in the interior of the laser,
and pits at the surface. (Figure 2)

1
Typical laser diodes have rise times in the 80 Most transients in today’s laboratories carry
ps range. Mirror damage is the most rapid energy in the high-frequency spectrum. Since
form of laser degradation, and the destruc- laser diodes are so responsive at these fre-
tion of the laser mirror can be virtually instan- quencies, they are easily damaged by tran-
taneous. Laser diodes are therefore very sients with relatively little energy content.
sensitive to fast overshoot events, such as
brief electrical transients and electrostatic dis- Worst of all, data first collected at British Tele-
charge (ESD), as well as DC over-current. com Research Laboratories indicated that
high-frequency transients too small to cause
catastrophic damage may result in latent
damage, leading to premature failure under
actual working conditions. Lasers damaged
in this way are obviously a serious liability in
applications requiring long lifetimes. (You’ll
notice that much of the protection strategies
discussed in this application note focus on the
prevention and attenuation of high-frequency
transient energy.)

Laser Protection Strategies

Our ongoing studies of laser diode protection
and electrical transient effects have taught us
a great deal about the nature of (and coupling
mechanisms of) electrical transients. The
results of these studies, combined with infor-
mation from laser manufacturers and experts
Figure 2. Scanning electron photomicrograph of a cata- in the fields of laser damage, ESD, and EMI,
strophically damaged laser facet showing fractures (pits) have led to the concept of “multiple-tier laser
and droplets. The arrows show the location of the p-n diode protection”. These tiers, or levels, are
junction. (Ref. 6) illustrated in Figure 3.

4-Tier Laser Protection Strategy

LEVEL Description Damage mechanism
I Instrumentation (Current source, Over-current, overheating, power
Temperature controller) line surges and spikes (transients)
II System setup (Cables and mounts, Radiated electrical transients
Proper grounding, Shielding)
III Lab environment, Power line Severe fast transients
conditioning, Other instruments
IV Laser packaging, Handling, and Electrostatic discharge (ESD),
Human contact Foreign-matter contamination

2
These levels have been created to highlight
principle areas of concern. As we’ll see, LEVEL I
these levels of protection are organized by Laser Control Instrumentation
the damage mechanisms against which they
are intended to protect. The order of these The instrumentation you use to control your
protection levels is not intended to imply rela- laser diode should be designed with appropri-
tive importance. In fact, human-contact ESD ate protection topologies and AC-line transient
(Level IV) is arguably the greatest cause of suppression. As a minimum, your drive instru-
premature laser failure. mentation must: 1) include an independen
adjustable limit of the drive current, 2) prevent
The following sections present the four levels reverse bias of the diode junction, 3) suppress
of protection in greater detail. Note that these electrical transient spikes from the AC power
levels are analogous to proverbial “links of a line, and 4) certainly not produce its own cur-
chain.” To ensure thorough laser protection, rent transients.
and minimize the risk of laser diode damage,
all levels of protection must be carefully imple- When used, thermoelectric temperature con-
mented. A single “weak link” in protection can trollers should at least prevent TE module
negate the time and investment made at other damage by providing a current limit. Firmware
levels. At the very least, familiarize yourself temperature limits and appropriate control
with the summaries presented at the end of algorithms (which prevent thermal oscillations)
each section. For your convenience, a check- are also recommended.
list of specific action items is presented at the
end of this document.

Figure 3. Multiple levels of laser diode protection. All levels must be fully implemented to ensure adequate defense
against damage.

3
 directly controlled with a feedback loop, the
loop response must be controlled to avoid
overdrive of the output in response to an input
step. To provide an adequate level of laser
protection, full control of the current-source
loop is required at all times. The circuits
described below provide these controls. Make
sure your laser driver has been tested to pro-
vide the necessary topologies.

Shorting Output
Making a small investment in time and effort
5 5
to choose the right instrument for your needs

Forward Voltage (V)

Light Output (mW)
can easily save time, money, and frustration 4
Optical Power
4

in the long run. The following sections outline 3

Forward Voltage
3

some of the features you should look for while 2 2

selecting laser control instrumentation. 1 1

0 0
Current Sources 0 10 20 30 40 50 60 70 80 90 100
Laser Current (mA)
Since a semiconductor laser is inherently a
current device, a true current source is rec- Figure 4. Optical output power and voltage vs. drive cur-
ommended for driving laser diodes. Typical rent of a typical laser diode. A laser diode is inherently
voltage sources (bench power supplies) care- a current device, and voltage supplies offer little control
at best.
fully ramp voltage at turn-on, but current is not
controlled. In fact, a capacitor is frequently The laser driver you choose should maintain
placed across the output of voltage sources the output leads at the same potential while
to hold the voltage “stiff.” Unfortunately, the laser is not being operated. ILX current
this allows the current to change quickly for sources have a shorting relay across the
changing loads, putting the laser at risk. output that maintains the laser leads at the
same potential. Until the output is turned on,
In short, a voltage source simply cannot pro- the shorting relay remains closed. The relay
vide the same level of control as a true current provides shorting protection even if the power
source. Direct control of current allows for to the instrument is off. This “shorting output”
precise linear control of laser diodes over a provides safety against ESD damage while
wide range of impedances. Unlike a voltage the laser is connected to the instrument.
source, the current will not increase if the cir-
cuit reaches the compliance voltage limit. Be aware that some laser instrumentation
utilizes FETs for output shorting. These tran-
Unfortunately, current control is not without sistors are active only when the instrument is
its own design challenges. When current is powered-up. In such cases, you should

4
turn the instrument on (and leave the current Current Limit
output turned off) during laser connection or Since laser diodes are sensitive to excessive
installation. drive current, the need for an independent
current limit feature is critical. Accidental
Slow-Start excess drive current will result in instanta-
To protect against turn-on transients, your neous overheating in the laser, and melt the
laser diode driver should feature a slow-start mirror facet and/or facet coating.
circuit that guarantees an overdamped turn-on
response. All ILX Lightwave precision current Unfortunately, current limit is often poorly
sources gradually open their shorting circuits implemented by manufacturers of laser driv-
during turn-on, allowing current to be slowly ers. Unlike some products, ILX Lightwave
sourced to the laser. Current is held from the current sources provide an exclusive clamping
output until the control circuits are fully active limit. Not only does this provide for protection
and all circuit transients have “died out”. against over adjustment of the setpoint cur-
rent, it also limits drive current during analog
To prevent instrument turn-on transients, modulation and constant-power operation.
start-up times longer than 100 msec are suf- (Figure 6)
ficient. However, the US Federal Government
requires laser systems to include a minimum The implementation of a current limit will
2 second turn-on delay for personal safety determine it’s effectiveness in non-DC modes
(CDRH US21 CFR 1040.10). This should not of operation. In most limit topologies, the
be confused with a slow-turn-on protection control loop actually switches from linear to
strategy. Your laser driver should include both nonlinear control as it moves in and out of
features. limit. Depending on the time response of the
limit circuit, some transient can be expected.
Proprietary circuit topologies developed at ILX
Over Lightwave have eliminated this transient in
Voltage
Detect constant-current operation.
Current/Power Slow Clamping
Setpoint Start Current Limit
Shorting Set your current limit just above the operating
Relay
current. ILX Lightwave current limits may be
Current adjusted even while the laser is in operation, if
Sense
Linear the need for higher drive current arises. Keep
Power Supply
the limit below the specified maximum-current
Line Filter
rating provided by the laser manufacturer.

Figure 5. Functional block diagram of a typical ILX Over Voltage Protection

Lightwave precision current source topology. Note laser If a simple current source is forced to drive a
protection features. high impedance load, its current control loop
will saturate, and it will be unable to drive to

5
 Analog
1 1 1
Input

Current Limit
@125mA

Current
Output
2 2 2

Ch1 2V Ch2 50mV M 2.5µs Ch1 ι -120mV Ch1 2V Ch2 50mV M 2.5µs Ch1 ι -120mV Ch1 2V Ch2 50mV M 2.5µs Ch1 ι -120mV

Manufacturer "A": Limit Fails. Manufacturer "B": Limit only ILX Lightwave LDX-3525.
Drive current exceeds limit. partially effective. Drive current Current clamped at 125mA limit.
exceeds limit.

Figure 6. Your laser diode driver should utilize a clamping limit topology to prevent overdrive. (Signal 1: Analog
modulation input. Signal 2: Instrument output current. From ILX Lightwave Technical Note #TN-3525-3)

the required (set) current. Furthermore, if the Power Line Transient Suppression
impedance is suddenly reduced, the control AC line transients have become practically
loop will drive the output current to the short- ubiquitous in today’s laboratories. These tran-
circuit limit. The laser will be overdriven until sients arise from a wide variety of sources.
the loop feedback can adjust the control circuit High-voltage surges (caused by lighting or
and reduce the drive voltage. Although brief, poor power conditioning) have always been a
this overdrive is generally sufficient to damage problem for users of laser diodes. In addition,
the laser. Your laser driver should protect electronic fast transients (EFT’s) are becom-
against this condition by turning off the output ing increasingly prevalent with the widespread
stage if the control loop is saturated. use of computers and other electronic equip-
ment that employ switching power supplies.
We realize that laser diode impedance usually (See “LEVEL IV, Lab Environment”)
changes slowly, and it’s useful for you to be
aware of this condition prior to the protective No instrumentation can prevent transients that
shut-off. Consequently, ILX Lightwave preci- are radiated directly onto the laser or drive
sion current sources have been designed with cables. Nevertheless, the driver must sup-
a two-stage overvoltage indicator. The first press transients that reach its power input
stage indicates when the upper control range stage, and prevent direct coupling through the
is approached. If the control loop becomes instrument. Usually, a combination of input
saturated, the second stage shuts off the filters is used to achieve appropriate suppres-
output to protect the laser from overcurrent. sion.
This two-stage trigger protects the laser, while
providing a useful indication if the compliance An extreme isolation transformer provides
voltage of the loop is reached. the best transient suppression, but cost and
size make it prohibitive for a standard cur-
rent source. Typically, current sources use a
simple capacitive filter on the rectifier circuit.

6
 diode packages contain an internal photo-
diode that is internally connected to either the
laser anode or cathode. Figure 7 illustrates
the recommended connections and shield-
ing for various configurations of laser diodes
and photodiodes with ILX Lightwave Current
Sources. If reverse biasing is used (photo-
conductive mode), the bias supply should
have a voltage which will not cause the
common-mode voltage of the feedback inputs
to exceed its specified maximum. Avoid using
a reverse photodiode bias on a laser pack-
age with common photodiode and laser pins.
(Most ILX Lightwave controllers provide an
adjustable photodiode bias.)

ILX Lightwave instrument designs are subjected to rigor- At low operating temperatures, some laser
ous transient tests. Tests include suppression of 1000V diodes are capable of producing excessive
fast-transients and power line surges. (See ILX Lightwave output powers, even if the rated drive cur-
Technical Standard #LDC-00196.)
rent is not exceeded. It’s a good idea to set
Additionally, an input line conditioning filter a power limit on your laser driver if one is
may be used. At ILX Lightwave we have provided. Don’t rely too heavily on a power
tested various combinations of these filters. limit, however, since its accuracy is dependent
We have found that the capacitive filter works upon the detector coupling efficiency and lin-
best when used in conjunction with a spe- earity.
cially-designed, doubly-shielded power supply
transformer. In tests of our current sources Choosing a Laser Diode Driver
with a line transient generator, this combina- When selecting a laser driver, check to make
tion of filters significantly reduced the output sure it includes the features noted above.
transient at the laser diode output. (See ILX These features generally apply to all users. In
Lightwave Technical Standard #LDC-00196, addition, select a drive instrument with maxi-
“Measuring Transient Suppression of Laser mum current levels commensurate with your
Diode Drivers”.) laser’s operating current. (In other words, a
high-power current source should not be used
Constant Power Operation to drive a low-power laser. Certain param-
Some current sources offer a constant opti- eters, including some types of transient sup-
cal power mode. When photodiode feedback pression, scale with the maximum current of
is used to control the output to the laser, care the instrument.)
must be taken to ensure the proper connec-
tions of the monitor photodiode. Many laser Additionally, if you plan to run your laser in DC
(continuous-wave) mode, choose a current

7
 Figure 7. Possible configurations used to connect laser diodes and photodiodes to an ILX Lightwave
precision current source or controller.

source that provides a low-bandwidth mode. shielded transformers, power line filters,
When low-bandwidth mode is selected, ILX clamping current limits, slow-start-up topolo-
Lightwave current sources engage an output gies, and other protection topologies. In
stage filter and use a slower control-loop addition, the ability of our current sources to
bandwidth. This further reduces the magni- suppress power line transients is measured
tude of noise and transients on the output. against IEC 801-4 and IEC 801-5 line tran-
sient waveform standards.
It is important to note that a precision current
source is only one part of a laser protection Our own comparative testing and discussions
plan. No instrument can protect against all with researchers have indicated that some
conditions. The instrument itself cannot pro- products marketed for driving laser diodes are
tect against ESD-related damage that occurs clearly not suitable for their intended purpose,
to the laser prior to connection, or transients even when laser protection claims are made.
that are radiated directly onto the laser or We strongly recommend that you thoroughly
through the cabling (see “System Setup,” evaluate any laser driver for its ability to pro-
below). However, ILX Lightwave precision tect your laser diode.
current sources and controllers offer the most
advanced laser protection against the damage Temperature Controllers
mechanisms mentioned above. ILX Lightwave When considering laser protection, most
current sources features include double- researchers tend not to think about tempera-
ture control. Of course, many critical

8
laser diode parameters, including wave- be used. For more information, refer to ILX
length, threshold current, and efficiency, are Lightwave’s Temperature Control Application
highly dependent on junction temperature. Notes. (See References.)
So, for many applications, highly stable tem-
perature control is desirable. Operating tem- If you use a temperature controller, be sure to
perature also greatly affects the lifetime of a set its current limit below the maximum rating
laser diode. As a rough estimate, lifetime is of the TE module. To avoid oscillations, set
reduced by an order of magnitude for every the gain of the instrument feedback loop to
30°C rise in case temperature (Figure 8). It’s match the thermal load. If you are not sure
usually desirable to operate the laser at as low where to set the gain, start with a low gain,
a temperature as possible, depending on your and gradually increase the gain until you
application and laser type. achieve a quick settling time without signifi-
cant overshoot. (ILX Lightwave temperature
Simple passive heat sinks can be used to controllers with a “hybrid smart integrator”
draw heat away from low-power laser diodes. algorithm minimize this condition.)
High-power laser diode arrays usually require
water cooling. If the application demands a We have recently evaluated several commer-
high degree of stability, temperature control cially-available laser mounts with TE modules
instruments that drive Peltier, or thermoelec- that are unable to hold the laser at their speci-
tric, modules to actively remove heat should fied low temperature rating for more than a
short time. If heat sinking is not adequate to
1000
dissipate generated heat, the operating effi-
ciency of thermoelectric modules is reduced,
500
400
and large temperature drifts can result. This
300 is not only frustrating, but puts the laser at risk
200
as well. We suggest that you test your mount
100 without a laser load (or with a “dummy” load)
Lifetime Relative

prior to actual use. ILX Lightwave heat sinks

are tested to hold the specified low tempera-
ture indefinitely under normal conditions.
10
Chose a temperature controller with a high
temperature limit (or thermistor resistance
limit), and set the limit just above operating
conditions. This will help protect against the
1
0 10 20 30 40 50 60 70
thermal runaway condition mentioned above.
Temperature (°C)

Figure 8. Laser lifetimes decrease dramatically with

higher operating temperatures. Relative lifetimes are ref-
erenced to 70°C. (Ref. 22)

9
LEVEL II
System Setup
(Mounting, Cables, and Grounding)

Choosing the right driver is imperative in function generator via bias-T network, ther-
ensuring a protected laser. But the instrument moelectric coolers, or monitoring equipment
itself cannot protect against transients that such as oscilloscopes). When constructing a
are radiated directly onto the laser or through system, remember that the weakest link will
system cabling. To protect against these determine the relative strength of the system.
threats, careful consideration of interconnec- This is especially true concerning radiated
tion must be made during system setup. noise susceptibility.

Here, we define “system setup” to encompass

the diode mount and cabling in addition to
the driver or controller. It also includes other
test equipment connected to the system (e.g.,

Overview of Recommendations — Level I

 Do not use a voltage source to drive your laser. At a minimum, select a
current source with a shorting output, slow-start, independent clamping
current limit, over-voltage protection, and proven transient suppression.
 Select a drive instrument with maximum current levels commensurate
with the laser’s operating current. (Certain parameters scale with the
maximum current of the instrument, including the level of power-line tran-
sient suppression.)
 Set current limit to appropriate levels, following the laser manufacturer’s
recommendations (or to just above expected operating current, if the
manufacturer does not specify maximum current). Use a power limit as
well, if your instrumentation provides one.
 Operate your laser at the lowest temperature possible, depending on your
application and laser type. Make sure your laser mount is capable of dis-
sipating the heat generated by your laser without thermal runaway.

10
 that are most common.

Grounding
Although system grounding will be discussed
along with each element of the system, a
thorough understanding of grounding will aug-
ment the following discussions on mounts,
cables, and other test equipment.

Unfortunately, the “ideal” system connections Grounding Terminology

will change for different test setups and user A discussion of “grounding” cannot be com-
needs, and system setup frequently requires plete without a review of terminology. Be
some changing or “tweaking”. This section will careful how you use the word “ground,” or
aid in determining what is ideal for your situ- more specifically, the “ground node”. Web-
ation by exploring the issues most commonly ster’s dictionary defines ground as “surface of
encountered in laser setups. the earth; or a conduction line between electri-
cal equipment and the ground.” Interestingly,
To isolate and eliminate electrical transients, another definition for ground is: “to establish;
we must first understand how these transients to instruct in elementary principles — a basis
arise. It is best to start with a simple model. or foundation (groundwork).” These definitions
For our purposes, radiated transients can be epitomize the confusion that’s usually associ-
considered as noise signals. The elements ated with discussions of “ground” in electronic
of our model are simply: 1) the source of systems.
noise, 2) coupling mechanism, and 3) receiver
(Figure 9). The first definition of ground can be clearly
described by “earth ground,” and any connec-
Typical noise sources include: transformers, tion to this node or potential is identified with a
fluorescent lights, FM radio stations, switching specific symbol. This “earth ground” is always
power supplies, high speed data communica- at earth potential. The second definition of
tions channels, lightning, motorized position- ground infers the zero-voltage plane (which is
ing equipment, and other peripheral equip- the reference for the electronic circuitry). This
ment. The receiver in our case is the laser reference plane does not need to be (and in
diode. The coupling mechanism is usually some instances should not be) connected to
the most difficult element to recognize. If all the earth ground.
three elements are identified, a fix will usually
become obvious. In many circumstances, the
cost associated with identifying all elements of
transient coupling in your laboratory is imprac-
tical. But precautionary steps can be taken to
prevent and eliminate the types of transients

11
 for two reasons — 1) to determine the volt-
Noise
Source
Coupling
Channel
Receiver
(Laser) age potential at which the laser operates (with
respect to earth ground and other electronic
Figure 9. Typical noise path. All three elements are equipment), and 2) to provide a return path for
necessary to produce a problem. the device current. Keep these two functions
To complicate matters further, there can be in mind when connecting your system ground.
more than one reference plane, depending on
the requirements of the electronics within the The first purpose of a ground system is to ref-
system or setup. For such reference planes, erence the diode with respect to earth ground.
we recommend specific descriptions along This is sometimes defined for you. Often, the
with the word “ground”. Descriptors such as heat sink requirements will dictate where the
“analog ground” and “digital ground” identify earth potential connection is made. (Mounts
the functionality clearly. Finally, avoid using are usually grounded.) Another example is a
the word “ground” by itself, or where the word modulated laser with the anode connected to
“return” more accurately conveys the function the case. In this configuration, your modulat-
of the electrical node. ing source (i.e., microwave generator) con-
nects the anode to earth ground. Since the
The word “node” refers to a single point, and anode is at earth potential, no other node in
connections to this single point should be your electronic system should be connected
observed. The system ground node in your to earth. (Otherwise, you may short your
schematic is undoubtedly the most impor- diode, or worse, short a current-sensing ele-
tant (although not the most exciting) node in ment.)
the system. In practice, the system ground
can rarely be implemented as a single point. Tests conducted at ILX Lightwave have shown
Where greater demand for accuracy is that earth-ground location can profoundly
needed, the amount of attention required by affect the magnitude of transients coupled
the ground node increases. onto the system. When earth grounding your
system, avoid ground loops. (See sidebar,
(Note that bus wire and PC board traces pg.12) A single point ground node (with only
have a small, yet significant resistance that one physical connection point) is the best
results in voltage changes along the ground solution. Unfortunately, this is usually not
trace. While this usually doesn’t pose a seri- practical. If multiple grounding cannot be
ous threat to the laser, it can cause unwanted avoided, attention must be paid to the ground
errors and drift in your system.) current flowing in the loop.

Grounding Considerations If you choose to locate the ground potential

Grounding is a topic that is simple in concept, at an unusual place, give yourself a reminder,
but becomes complicated (and misunder- such as a Post-it note. Otherwise, you may
stood) in practice. Most confusion arises from inadvertently pose a shock hazard to both
the fact that system elements are grounded yourself and your laser.

12
 What is a “Ground loop”?
...That’s probably one of the least asked but most frequently considered questions in system design.
The term “ground loop” is often used when trying to explain poorly understood setup problems,
which simply adds to the confusion.

In short, a ground loop is a closed path, or loop, of electrical conductor. Electrical fields that inter-
sect a loop will induce a current in the loop. (The fact that the loop’s electrical potential is that of the
system reference plane makes little difference to the electric field.) More specifically, a ground loop
is a conductor loop created by the specific connection of the various ground points in an electrical
system. Any circuit with many connections on a single node will create a loop. The bigger the loop,
the more noise energy that can be coupled into the loop. Furthermore, once the noise couples onto
the ground, it can easily couple everywhere in the system.

The signal ground node in most circuits normally has the most connections. This increases the
chances of large conductor loops. The earth ground node may also have common connections to
other instruments that are not intended to be part of your system. Inadvertent connection to these
conductors provides additional chances of loops and induced noise.

Not surprisingly, ground loops are a common occurrence in electronic systems. The best solution to
ground loop problems is to simply break the loop. Unfortunately, the hardest part is identifying the
ground loops. In those cases when the loop is unavoidable, then reduction of loop area is advis-
able.

The second function of a common ground is will help define voltage reference planes in the
to provide a current return path for system system and identify the current return paths.
current flow. This is usually where problems Ideally, each wire should be identified with
associated with “poor ground” arise. Avoid some associated series resistance and induc-
using the laser return path as a return for tance. The schematic should also include
other circuits as well. wire lengths and other details of the physical
setup that may affect transient coupling.
Even if you are using ILX Lightwave mounts
and cables in a standard configuration, we A schematic of a test setup with poor ground-
strongly recommended that you sketch a ing is shown in Figure 10a. Review of this
simple schematic of your setup, with a focus schematic points to some system connec-
on ground return paths. This schematic is a tion problems. Changing the connections to
tool that will help identify critical ground con- that shown in Figure 10b will result in greater
nections and potential noise problems prior to immunity to radiated transients and avoid
connection, and may also payoff while trouble- problems that can be caused by ground loops.
shooting system interference. The schematic

13
The actual implementation of ground connec- Laser
Mount
Controller
tions you make will be a compromise between Monitor
Laser
conflicting requirements. On one hand, Scope
TEC
ground loops (and alternate current return Scope
Ground
paths) should be avoided. On the other hand, AC
Soldering
low inductance, high-frequency grounds are Iron

needed to drain fast transients*. Line

Filter

Your specific application and laboratory envi- Building

Power
Building
Power
ronment will dictate choices of system config- Main

uration. If you are mainly concerned with high Figure 10a. Schematic of a poor system setup. Note the
frequency noise, then low-inductance multiple large ground loop formed by the laser instrument and
ground connections are needed. If low-fre- oscilloscope.
quency noise (50/60 Hz) is coupling into your
system, a single-point ground connection and Scope
isolated system elements may be preferable.
In most systems, both noise sources are pres- Mount Controller

ent. In these cases, careful control of ground Laser

connections will reduce return paths while AC

maintaining good high-frequency ground con-

Line
nections. TEC Filter

Through our own evaluations, we have

learned that grounding strategies vary among Soldering Building
Iron Power
instrumentation manufacturers in both imple-
mentation and effectiveness. It is wise to Figure 10b. Ideal system setup with isolated laser. Ground
become familiar with the manufacturers’ rec- loops have been eliminated.
ommended procedures.
sible, it’s best to leave the laser diode floating
with respect to earth ground. (Figure 10b)
ILX Lightwave current sources employ a
Nonetheless, some testing situations require
floating output to accommodate a variety
the laser diode be earth-grounded. In these
of laser configurations and systems. This
cases, at least try to reduce the ground loop
also reduces the possibility of line transient
area and impedance as much as possible.
damage and helps avoid ground loops. If pos-
Keeping the controller in close proximity to the
* A quick way to test your high-frequency ground diode will help. Providing a good (low induc-
is to place your hand on the earth ground at vari- tance) connection to earth ground from both
ous points along the ground path. If there is little the mount and the controller will also help.
or no change in the system operation, you can
assume that the high-frequency ground in suffi-
cient. If touching the ground causes a change (i.e., Laser Mounts
increased noise), your high-frequency grounding Laser diode mounts perform the seemingly
system could be improved. innocuous task of providing a fixture (and usu-

14
ally heat sinking) for a laser diode. But certain
considerations must be made to ensure a safe
environment for your laser.

Obviously, when selecting or fabricating a

laser diode mount, one must first consider the
diode package. There are two major packag-
ing styles (in addition to many special variet-
ies). Window-can style laser packages are
often used in free space (non-fiber-coupled)
applications (Figure 11). The package (can)
of the laser is usually connected to either the
laser anode or cathode. The mount configura-
tion determines whether the case (and there-
fore diode) is referenced to earth ground.
Figure 11. Fiber-pigtailed dual-in-line (left) and window-
If possible, the package should be left isolated can laser packages are commonly used.
from earth ground. This allows more freedom
and the package must be earth grounded. In
of connection and will usually reduce the likeli-
these situations, follow the guidelines above
hood of ground loops and interference. ILX
for grounded lasers. In addition, the ground-
Lightwave mounts for window-can lasers allow
ing of other equipment (such as function gen-
the option of isolating or earth grounding the
erators and oscilloscopes) should be made
laser package. If your mount does not allow
as close to the mount ground as possible to
for isolation, and the case is earth-grounded,
reduce ground-loop area. Avoid using any
then the mount-to-ground connection should
unnecessary equipment on the same power
be a very good RF ground (very low induc-
ground as the diode mount.
tance). Connection directly to power line
ground (the round prong on the AC plug) is
When working with an unpackaged laser die,
the minimum required. Ideally, you should
or package styles other than the types men-
connect the mount directly to earth ground
tioned above, it is important to understand
(through a metal rod), use wide braid wire to
any special mounting constraints. Grounding
reduce inductance, and/or keep cable lengths
guidelines identified above generally apply,
as short as possible.
however.
Fiber-pigtailed telecom lasers, usually
It is essential that your laser mount also sup-
encased in DIL (dual-in-line) or Butterfly
presses high-frequency noise. If the mount
packages are also quite common. In these
ground is not sufficient, or the radiation envi-
packages, the case is usually connected to
ronment is extremely harsh, the mount itself
the laser (typically the anode). Often, these
can act as an antenna and pick up radiated
lasers are modulated at high frequencies
signals, which will induce currents in the

15
case node of the laser diode. Attention to Loose Connections
ground configuration will normally improve A cable connection that is intermittent, or has
these problems. If the radiation source can be a contact that’s bouncing during connection,
identified, a grounded shield (sheet or screen poses a threat to the laser. Laser drivers see
of any conducting metal) can be placed this break as a higher impedance load, and
between the source and the mount to reduce try to drive harder. Ensure that the driver
the field strength. output is turned off prior to connecting your
laser diode. Fasten all cables, and make
Unfortunately, thermoelectric cooler modules, sure cable connections and solder joints are
common in laser mounts, can act as a capaci- robust, prior to turning the instrument output
tor between the case connection (heat sink) on. Avoid draping cables across your work
and the laser. Noise on the temperature con- area. Secure current source cables to your
trol instrument can be coupled through this bench or optical table to avoid bumping them
capacitance. Choose a low-noise, filtered TE (but don’t “bundle” current source cables with
controller to avoid this problem. other instrument cables in your lab). Don’t
attempt to “multiplex” your current source to
Also note that if low frequency (50/60 Hz) multiple loads with an external switch or relay.
noise is a problem, one cause may be mag-
netic coupling. If a transformer is near by, Twisted Pair Cable
then any cables running through the magnetic This type of wire is fairly common in electronic
field of the transformer will carry an induced equipment, and is preferred in long cable runs.
current. Magnetic coupling can be difficult to Twisted pair helps reduce low-frequency noise
reduce since normal shielding does not affect that is otherwise inductively coupled into the
the magnetic field. To attenuate magnetic signal conductors. (This is because twisting
fields, high permeability metals must be used. the wire effectively reduces the loop area and
These can be expensive. A more cost-effec- therefore the inductive term in the coupling
tive alternative is to shield the transformer or equation.) Unfortunately for twisted pair to
source of the magnetic field. have any effect on capacitively coupled noise,
a balanced-line receiver is needed. Since
Cabling & Shielding this is not the case for laser diode applica-
Since most laser diodes have a very low toler- tions, twisted pair cables offers no reduction
ance for reverse current, most users are care- for noise that is capactively coupled into the
ful to observe the proper polarity when con- signal wires. A shield is also needed.
necting their laser diode to a current source.
(If there’s any question about polarity, an LED Cable Shielding
makes an inexpensive polarity tester that will Our studies of laboratory transients have
light if the polarity is correct.) shown that, depending on the characteristics
of the transient, a surprisingly large amount of
Other, less obvious dangers of cabling can be energy can be coupled directly onto the
just as disastrous, but are easily overlooked.

16
current-carrying cable, even when the cable Ideally, the shield itself should be a braid with
includes a shield. In many cases a shield as much coverage as possible (within the
is present, but its effectiveness is limited by practical constraints of cable flexibility and
improper shield termination. Cable shielding diameter). A metal foil shield can be used,
is an area of concern that is often misunder- but the seam of the foil tends to leak high-fre-
stood, yet is imperative in minimizing risk of quency energy, and is not as effective against
laser damage. transients.

A shield on a cable serves to drain electrical ILX Lightwave Cables

fields through a low-impedance path to earth The proper cable will depend on both the
ground, restricting radiation of energy from system setup and the noise environment of
the internal wires. More importantly, it also the system. Of course, these factors will vary
prevents other electrical fields in the lab from greatly among applications. Based on find-
coupling onto the wires. ings from laboratory testing at ILX Lightwave,
we have created a unique twisted-pair cable
To illustrate how an unshielded cable can with braided outer shield. This cable has been
act as a transient coupling mechanism, the designed to provide the best possible rejec-
barbell model is useful (Figure 12). Here, it tion of most transient noise signals.
is assumed that the diode mount and laser
driver are adequately shielded. A good cable As of February 1996, this type of cable is
shield then completes the “barbell,” which available from ILX Lightwave as our standard
encloses the drive electronics, laser diode, current source cable. Although every situation
and laser connection wires in a Faraday Cage. is different, we feel that this cable provides the
Any breaks in the shield will cause “leaks,”
which will allow RF energy into the system.
Proper termination of the cable shield is there-
fore critical.
Controller Laser

Do not use the shield as a current-carrying

conductor. Rather, terminate the shield at
both ends to a low-inductance earth ground.
If this sets up a ground loop and you observe RF Field
significant low frequency (50/60 Hz) noise,
terminate one end (preferably the mount) to
earth ground through a few-hundred picofarad
capacitor. This will break the loop for low Controller Laser

frequency noise, while preserving the high RF Leak

frequency connection to earth ground to drain

RF Field
radiated transients.
Figure 12. System shielding represented with the “bar-
bell” model. Breaks in the shield may allow transients to
“leak.” (Refs. 10, 11)

17
best protection over the widest range of labo- is highly recommended. We realize that it’s
ratory conditions. Contact an ILX Lightwave impractical to eliminate all sources of tran-
representative for more information. sients. However, there are important implica-
tions for lab layout and organization that are
LEVEL III actually quite simple to implement. When in
Lab Environment doubt, it is always best to take the most con-
servative approach.
Whether a laser is damaged by a power-line
transient depends on several factors, including
the magnitude of the transient, the temporal
qualities of the transient, and the sensitivity of
the laser. These factors vary with the specific
laser and the particulars of your laboratory
environment.

Planning a lab environment to minimize the

quantity and magnitude of electrical transients

Overview of Recommendations — Level II

 Grounding is a complex issue, and is application dependent. Familiarize
yourself with the terminology and function of ground nodes. Avoid using
the diode wires as return paths for other currents. Avoid ground loops.
When using ILX Lightwave instruments, it’s best to leave the laser diode
floating with respect to earth ground.
 Use appropriately shielded cabling to reduce coupling of radiated tran-
sients. (As of February 1996, ILX Lightwave’s standard current source
cables are a unique braid-shielded twisted pair. Contact an ILX Lightwave
representative for more details.) Use proper cable termination.
 Make sure that all cables to the laser diode are securely fastened. Do not
“bundle” current source cables with other cables in your laboratory.

18
 designed with the maximum amount of tran-
sient suppression practical, and the use of
proper grounding and appropriately shielded
cabling can help suppress radiated transients.
Still, high frequency transients have a way of
finding their way past even the best defenses.
We strongly urge you to isolate sources of
EFT spikes as much as possible.

Radiated Transients
Laser diodes are sensitive to all sources and
types of electromagnetic interference (EMI),
including EFTs. As one laser manufacturer
points out, “Laser chips can be damaged or
destroyed by induction voltages that can occur
Electrical Fast Transients near equipment that emits high-frequency
Transient suppression techniques have tradi-
EMI. Avoid using laser chips near fluorescent
tionally focused on transients caused by light-
lamps or other external sources of EMI emis-
ning strikes in close proximity to the labora-
sions.” (Ref. 4)
tory (or near power lines entering the lab). A
In some environments, where power supplies
second class of line transient that has become
or industrial loads are switched on the line,
pervasive is the Electrical Fast Transient
high levels of radiated EMI transients may be
(EFT). EFTs are very fast transients (typically
present. Note that high voltage supplies for
tens of nanoseconds in duration) that tend to
gas lasers and excimer lasers are notorious
radiate and couple into other equipment due
for generating large amounts of EMI. This
to their high frequency content. (See sidebar,
type of interference can be a problem even
pg.18)
when the source is located in another labora-
tory.
In most labs, EFTs are caused by equipment
that requires large power line surge currents
Ideally, you should plan the layout of your
(e.g., soldering irons and most motor and
laboratory to maximize physical separation
compressor equipment). Equipment designed
between sources of EMI and your laser diode.
with switching power supplies (virtually all
Again, cables should be properly shielded,
computers) also generate EFTs. Again, EFTs
and securely connected to the drive instru-
are typically much faster than lightning or
mentation and the laser. In addition, it is a
power surges and, compared to surge tran-
good idea to keep laser cables as far as pos-
sients, have a greater tendency to radiate
sible from other instrument cables. Avoid
throughout the lab.
“bundling” instrument power cords and cables
As mentioned, instrumentation should be
together with the laser drive cable. Otherwise,
noise and transients from other instruments

19
Radiated Transients:
How are They Coupled?
Coupling mechanisms are the paths by which energy is transferred from a noise source into
receiving circuits. The most obvious example of a coupling mechanism is your TV antenna.
Unfortunately, coupling is not always desired. An example of this is the pickup of a neighbor’s
blender or vacuum cleaner during your favorite show.

The physical laws by which noise travels through space and is coupled onto a receiving conduc-
tor were described by James Clerk Maxwell (1831-1879). For our purposes, a simplified expla-
nation will provide an overview of the mechanisms described by Maxwell.

One mechanism involves the coupling of the electric field from a radiating source to a receiving
conductor. This can be modeled as a capacitance between the noise source and the receiver.
The amount of energy present in the receiver depends on the coupling capacitance and the
receiver’s capacitance to earth ground. This sets up an impedance divider circuit. For this type
of coupling, the parameters that affect the capacitance (i.e., area and separation of the conduc-
tors) will affect the coupling. (Ref.8, p.30)

At higher frequencies, electric fields are the most predominant coupling mechanism. Effective
containment of these fields can be accomplished by metal shields terminated to ground. (Note
that the ground termination must be low impedance at the noise frequency of interest.) Other
ways to reduce the capacitive coupling are separation of the conductors. This reduces the
capacitive term in the coupling model.

A second mechanism by which transients are coupled onto a receiving conductor is through
their magnetic field. This can be modeled as a mutual inductance between the noise source and
the receiving conductor. For this case the mutual inductance is dependent on the induced field
strength, magnetic properties of the medium, and area of the reciever conducting loop. (Ref.8,
p.37)

Inductive coupling is most common at lower frequencies (especially 50/60 Hz). This noise cou-
pling can only be reduced by containing the magnetic field or reducing of the loop area. Since
most metals have very little effect on the magnetic field strength, special (higher cost) materials
with high permeability must be used to effectively contain the magnectic field. Minimizing the
loop area of the receiving conductor is a more cost effective way to reduce inductive coupling.

20
may easily radiate directly onto the laser levels to model typical lab environments. We
cable. have used these standards to define a repeat-
able testing standard that is applied to all our
AC Power Line Isolation instruments.)
Most manufacturers claim that their current
sources include adequate transient suppres- The testing done at ILX Lightwave is, to our
sion. Unfortunately, their transient specifica- knowledge, the most extensive to date on
tions do not indicate the level of input tran- surge and EFT line noise with respect to laser
sients or other test conditions. diodes. (For more information, request ILX
Technical Standard LDC-00196, “Measuring
At ILX Lightwave, we test our current sources Transient Suppression of Laser Diode Driv-
for transient attenuation against IEC-stan- ers.”)
dard 1000 V transients, and strive to provide
our customers with as much information as In comparisons with other manufacturers’
possible. (The IEC standards committee has products, ILX Lightwave current sources have
defined standard surge and EFT profiles and extremely low transient feedthrough (Table 1).

Even seemingly innocent lab equipment such as bench power supplies and other lasers can generate high levels of EMI.
Take appropriate precautions.

21
In most cases, when proper grounding is tector, and keep these instruments as far as
used, ILX Lightwave current sources provide possible from the laser and current-carrying
excellent transient suppression. However, cables.
when very severe line transients are a prob-
lem, there is no substitute for multiple stages LEVEL IV
of protection, such as found when an isolation Handling & Human Contact
transformer is used with a well-designed cur-
rent source. Proper handling is perhaps the most over-
looked aspects of laser protection. Yet care-
If industrial loads are switched in or near your ful storage, transport, and mounting of laser
laboratory, use isolation transformers and/or a diodes is nothing less than critical in ensuring
surge-suppresser power strip. We have found optimum laser lifetimes.
that an Extreme Isolation Transformer (EIT)
provides the best protection against tran-
Electrostatic Discharge
sient feedthrough. Furthermore, avoid using
Laser diodes, like most semiconductor
laboratory power supplies, soldering irons, or
devices, can be easily damaged or destroyed
other electronic instruments that use switch-
by inadvertent electrostatic discharges (ESD).
ing power supplies on the same power strip
In fact, it’s been suggested that ESD is the
as your laser current source. (“Surge-sup-
single leading cause of premature laser diode
pression” power strips are often designed to
failure. As one laser manufacturer mentioned,
isolate the AC main from your equipment, but
“Most researchers will go through at least one
are not necessarily effective in isolating instru-
or two lasers before realizing they have not
ments from one another.) If you must use a
taken appropriate static precautions.”
common line with such equipment, isolate
your laser driver with a separate surge pro-
ESD in your laboratory can be difficult to pin

Overview of Recommendations — Level III

 Plan your laboratory to maximize physical separation between sources
of EMI and your laser diode. Keep laser cables as far as possible from
other instrument cables, and don’t “bundle” power cords and cables
together.
 If industrial loads are switched in or near your laboratory, use isolation
transformers and/or a surge-suppresser power strip with your laser cur-
rent source.
 Isolate your laser current driver with its own surge protector when using
a common line with laboratory power supplies, soldering irons,or other
electronic instruments. Do not use such devices on the same surge-pro-
tector power strip as your laser current source.

22
 Proper discipline in following an ESD con-
trol program cannot be understated. A
static-free environment is mandatory.

The elimination of all potential ESD risks is

1.2 KV and less

1.4

Light Output
1.5

1.7

1.9

point, since laser diodes can be damaged by 0

voltages that are too small to feel through your 0 10 20 30 40 50 60 70
Drive Current (mA)
skin. Human skin is only sensitive to 3000V
or more, yet empirical evidence has shown Figure 13. Forward L/I Characteristics of InGaAsP laser
that commercially available InGaAsP and after reverse-biased ESD stressing at various voltages.
Note the abrupt onset of failure at 1400 V. (Ref. 7)
AlGaAs laser diodes can be damaged by ESD
voltages as low as 1200V. (Figure 13) Just straightforward in theory, but can be difficult
because you don’t hear or feel a static spark, to achieve. In facilities where laser diodes (or
don’t assume there isn’t a dangerous ESD other static-sensitive devices) are handled
discharge. regularly, an ESD audit conducted by an inde-
pendent certified consultant is recommended.
As with all semiconductor devices, ESD has (Ref. 16) In general, though, there are four
the potential for latent damage in laser diodes. basic “golden rules” that must always be fol-
In other words, ESD may simply weaken the lowed:
device without any immediate symptoms. The 1. Whenever handling unprotected devices,
static discharge breaks down the P-N junction wear a protective wrist strap designed to
in an area outside the optical cavity. During drain built-up electric charges safely to
normal use, these defects propagate into the ground. Choose a secure, but comfortable
laser cavity over time. The resulting degrada- strap with a 1 MΩ series resistor. Prop-
tion in performance may appear long after the erly ground tweezers, soldering irons, and
initial damage takes place. (Ref. 5) When this other tools as well.
“latent failure” finally occurs, it may be attrib- 2. Place unprotected lasers only on certified,
uted to other causes. Ignored by the user, the static dissipative work surfaces.
ESD problem remains unchecked. 3. When transporting, storing, or not work-
ing on your laser, completely enclose it in
a conductive, shielded material. Short the
pins when not in use by inserting them

23
 into conductive foam. Spring-like shunting cal contact on the laser chip or contamina-
devices that are mounted to the laser and tion of the facet. To clean particles from the
automatically short the pins are highly rec- facet, use a gentle stream of dry nitrogen
ommended. (Ref. 17) gas. Handle the device with tweezers to avoid
4. Keep all charge-generating materials at contamination with skin oils. Do not use ther-
least 12” away from unprotected devices mal grease when mounting these types of
since the device can become charged lasers to a heatsink, as grease can creep and
inductively. eventually contaminate the laser. (Thermal
grease is acceptable with sealed packages.)
In problem areas, anti-static floor coverings If the facet becomes contaminated with an oil-
and ionized air blowers are also recom- based substance, follow the manufacturer’s
mended. Wear clothing made of materials recommendation for cleaning. Use caution if
resistant to charge creation. (Cotton is best, attempting to couple an optical fiber to the
wool is fair, and synthetics are poor.) laser, as contact of the fiber against the facet
may cause damage.
Static protection products should be examined
on a regular basis for effectiveness, as they
may degrade over time. Remember, taking
preventative measures against ESD damage
is particularly important since static damage
may not show up until after the laser diode
has been used for some time.

Other Handling Precautions Figure 14. Anti-static gloves or finger cots should be worn
The window of TO-packages is usually quite when handling laser diodes. Do not touch the facet of
open heatsink lasers. (Ref. 2)
thin (typically about ¼ mm). Do not push on
the window when inserting the laser diode into Improper soldering practices are a common
a mount socket. Wear anti-static gloves or cause of laser overheating. Most manufactur-
finger cots, and push the device into place by ers provide detailed instructions for soldering,
pressing on the base of the package. including maximum solder temperatures and
times — usually less than 10 seconds at 250
Store the laser diode in its shipping container to 300°C.
when not in use. This not only helps protect
against ESD, but also keeps dust and dirt off Do not carry or handle any laser diode by
the laser window. If the window becomes the pins. Pins in many laser packages are
dirty, clean it with a lens tissue wet with insulated from the case with a glass seal. Be
reagent-grade acetone or propanol. (Ref. 2) careful when installing your laser to avoid over
bending the pins which can break the seal.
When handling open-heat-sink laser diodes,
exercise extreme caution to avoid mechani-

24
Summary
LEVEL I Drive Instrumentation
Laser diodes are extremely sensitive to elec-  Do not use a voltage source to drive your laser. At
a minimum, select a current source with a shorting
tro-static discharge and current spikes (tran-
output, slow-start, independent clamping current
sients). Damage can result in reduced output limit, over-voltage protection, and proven transient
power, shift in threshold current, changes in suppression.
beam divergence, and ultimately failure to  Select a drive instrument with maximum current
lase (LED-like output only). levels commensurate with the laser’s operating cur-
rent. (Certain parameters scale with the maximum
current of the instrument, including the level of
This application note has outlined precautions power-line transient suppression.)
and procedures that you can take to protect  Set current limit to just above expected operating
your laser diode. A summarized checklist of current. Use a power limit as well, if your instru-
specific actions follows. We strongly urge you mentation provides one.
 Operate your laser at the lowest temperature pos-
to follow these recommendations and use
sible, depending on your application and laser type.
the checklist to minimize your risk of laser Make sure your laser mount is capable of dissipat-
damage. ing the heat generated by your laser without ther-
mal runaway.
We at ILX Lightwave encourage your safe and
successful laser diode use. If you have any LEVEL II System Setup
 Grounding is complex issue, and is application
questions, please contact your ILX Lightwave
dependent. Familiarize yourself with the terminol-
representative. ogy and function of ground nodes. Avoid using
the diode wires as return paths for other currents.
Avoid ground loops. When using ILX Lightwave

Overview of Recommendations — Level IV

 ESD is the primary cause of premature laser failure. As a minimum, use
anti-static wrist straps (grounded with 1 MΩ resistor), grounded soldering
irons, and grounded work areas. Ionized air blowers and anti-static floor
coverings are also recommended.
 Laser diode leads should be shorted whenever the laser is transported or
stored. (ILX Lightwave current source outputs are automatically shorted
when not in use.) Place unprotected lasers only on static dissipative
work surfaces, and keep all charge-generating materials away from your
laser.
 Exercise caution when handling laser diodes. Wear anti-static gloves,
and clean your laser according to the manufacturer’s recommendations.

25
Laser Protection Checklist
All of the following precautions must be taken to ensure the complete protection of your laser
diode and to minimize the risk of damage:

LEVEL I Drive Instrumentation LEVEL III Lab Environment

 Do not use a voltage source to drive your laser. At  Plan your laboratory to maximize physical
a minimum, select a current source with a shorting separation between sources of EMI and your laser
output, slow-start, independent clamping current diode. Keep laser cables as far as possible from
limit, over-voltage protection, and proven transient other instrument cables.
suppression.
 If industrial loads are switched in or near your
 Select a drive instrument with maximum current laboratory, use isolation transformers and/or a
levels commensurate with the laser’s operating surge-suppresser power strip with your laser
current. (Certain parameters scale with the current source.
maximum current of the instrument, including the
level of power-line transient suppression.)  Isolate your laser current driver with its own
surge protector when using a common line with
 Set current limit to just above expected operating laboratory power supplies, soldering irons, or other
current. Use a power mimit as well, if your electronic instruments. Do not use such devices on
instrumentation provides one. the same surge-protector power strip as your laser
current source.
 Operate your laser at the lowest temperature
possible, depending on your application and laser LEVEL IV Handling & ESD Protection
type. Make sure your laser mount is capable of  ESD is considered the primary cause of premature
dissipating the heat generated by your laser without laser failure. As a minimum, use antistatic wrist
thermal runaway. straps (grounded with 1 MΩ resistor) grounded
soldering irons, and grounded work areas. Ionized
LEVEL II System Setup air blowers and anti-static floor coverings are also
 Grounding is a complex issue, and is application recommended.
dependent. Familiarize yourself with the
terminology and funcation of ground noted. Avoid  Laser diode leads should be shorted whenever
using the diode wires as return paths for other the laser is transported or stored. (ILX Lightwave
currents. Avoid ground loops. When using ILX current source outputs are automatically shorted
Lightwave instruments, it’s best to leave the laser when not in use.) Place unprotected lasers only
diode floating with respect to earth ground. on static dissipative work surfaces, and keep all
charge-generating materials away from your laser.
 Use appropriately shielded cabling to reduce
coupling of radiated transients. (As of February  Exercise caution when handling laser diodes. Wear
1996, ILX Lightwave’s standard current source anti-static gloves, and clean your laser according to
cables are a unique braid-shielded twisted pair. the manufacturer’s recommendations.
Contact an ILX Lightwave representative for more
details.) Use proper cable termination.

 Make sure that all cables to the laser diode are

securely fastened. Do not “bundle” current source
cables with other cables in your laboratory.

26
References:
1. Heterostructure Lasers, Volume 1&2. H. Casey and M. Panish.
2. Laser Diode Operator’s Manual & Technical Notes. 1994, SDL Inc.
3. 1993 Product Guide. 1993, Lasertron Corporation.
4. 95-96 Laser Diode Data Book. 1995, Rohm Corporation
5. S. P. Sim, M. J. Robertson, R.G. Plumb, “Catastrophic and latent damage in GaAlAs lasers caused by electrical
transients,” Journal of Applied Physics, 55, 3950-5 (June, 1984).
6. D. A. Shaw, P.R. Thornton, “Catastrophic Degradation in GaAs Laser Diodes,” Solid State Electronics, 13, 919-24
(1970).
7. L. F. DeChiaro, B. A. Unger, “Degradation in InGaAsP Semiconductor Lasers Resulting from Human Model ESD,”
1991 EOS/ESD Symposium Proceedings.
8. Noise Reduction Techniques In Electronic Systems. H. Ott., 1976, John Wiley & Sons.
9. Controlling Radiated Emissions by Design. M. Mardiguian, 1992, Van Nostrand Reinhold.
10. E. F. Vance, F. M. Tesche, “Shielding Topology in Lightning Transient Control,” NASA Conference Publication 2128
FAA-RD-80-30.
11. D. L. Sweeney, M. Sweeney, “Mitigating Excessive Emissions,” ITEM ’89
12. D. Gerke, B. Kimmel, “Grounding: Facts and Fallacies,” EDN Supplement, 39, 2, 91-100. (January 20, 1994).
13. D. Gerke, B. Kimmel, “Cables and Connectors: How to Stop the EMI Leaks,” EDN Supplement, 39, 2, 71-77 (Jan-
uary 20, 1994).
14. W. G. Olsen, D. J. Hodgson, B. Bowen, “Measuring Transient Suppression of Laser Diode Drivers,” Technical
Standard TS-00196, 1995 ILX Lightwave Corp.
15. D. E. Frank, “ESD phenomenon and effect on electronic parts,” Douglas Paper 7072, McDonnell Douglas Corp.,
April, 1981.
16. R. Peirce, “ESD Liability Analysis,” ESD Technical Services,” August, 1995
17. D. Cronin, “CRO-BAR: A New Technique for ESD Protection,” EMC Test & Design, January, 1993.
18. J. R. Huntsman, D.M. Yenni, Jr., “Test Methods for Static Control Products,” 1982 3M Static & Electromagnetic
Control Division.
19. D. M. Yenni, Jr., J. R. Huntsman, “Quality Through Static Damage Prevention,” 1981 3M Static & Electromagnetic
Control Division.
20. D. Stanisich, B. Bowen, “Laser Diode Protection Strategies,” 1988 ILX Lightwave Corp.
21. D. J. Hodgson, “Clamping Limit of an LDX-3525 Precision Current Source,” Technical Note, 1995 ILX Lightwave
Corp.
22. Physics of Semiconductor Laser Devices., G. Thompson, John Wiley & Sons, 1980, 28-29.

For more information about temperature control, request the following ILX Lightwave Applications Notes:
ILX Application Notes #1, “Controlling Temperatures of Diode Lasers and Detectors Thermoelectrically.”
ILX Application Notes #2, “Selecting and Using Thermistors for Temperature Control.”
ILX Application Notes #4, “Thermistor Calibration and the Steinhart-Hart Equation.”

Special Thanks to Roger Peirce, President of ESD Technical Services, Langhorne, PA. (215) 364-1050. Formerly a
research scientist at AT&T Bell Laboratories, Roger is an expert in the area of ESD prevention.

© Copyright ILX Lightwave Corp., 1996.

For additional copies, please contact your ILX Lightwave representative.

27
 The following publications are available for download on at www.ilxlightwave.com.

White Papers
• A Standard for Measuring Transient Suppression of Laser Diode • Typical Output Drift of a LDX-3412 Loc-Cost Precision Current
Drivers Source
• Degree of Polarization vs. Poincaré Sphere Coverage • Typical Output Noise of a LDX-3412 Precision Current Source
• Improving Splice Loss Measurement Repeatability • Typical Output Stability of the LDC-3724B
• Typical Output Stability of a LDX-3100 Board-Level Current Source
Technical Notes • Typical Pulse Overshoot of the LDP-3840/03 Precision Pulse
Current Source
• Attenuation Accuracy in the 7900 Fiber Optic Test System • Typical Temperature Stability of a LDT-5412 Low-Cost Temperature
• Automatic Wavelength Compensation of Photodiode Power Controller
Measurements Using the OMM-6810B Optical Multimeter • Using Three-Wire RTDs with the LDT-5900 Series Temperature
• Bandwidth of OMM-6810B Optical Multimeter Analog Output Controllers
• Broadband Noise Measurements for Laser Diode Current Sources • Voltage Drop Across High Current Laser Interconnect Cable
• Clamping Limit of a LDX-3525 Precision Current Source • Voltage Drop Across High Current TEC Interconnect Cable
• Control Capability of the LDC-3916371 Fine Temperature Resolution • Voltage Limit Protection of an LDC-3916 Laser Diode Controller
Module • Wavelength Accuracy of the 79800 DFB Source Module
• Current Draw of the LDC-3926 16-Channel High Power Laser Diode
Controller Application Notes
• Determining the Polarization Dependent Response of the FPM-8210
Power Meter • App Note 1: Controlling Temperatures of Diode Lasers and
• Four-Wire TEC Voltage Measurement with the LDT-5900 Series Detectors Thermoelectrically
Temperature Controllers
• Guide to Selecting a Bias-T Laser Diode Mount • App Note 2: Selecting and Using Thermistors for Temperature
• High Power Linearity of the OMM-6810B and OMH-6780/6790/ Control
6795B Detector Heads
• Large-Signal Frequency Response of the 3916338 Current Source • App Note 3: Protecting Your Laser Diode
Module
• Laser Wavelength Measuring Using a Colored Glass Filter • App Note 4: Thermistor Calibration and the Steinhart-Hart Equation
• Long-Term Output Drift of a LDX-3620 Ultra Low-Noise Laser Diode
Current Source • App Note 5: An Overview of Laser Diode Characteristics
• Long-Term Output Stability of a LDX-3525 Precision Current Source
• Long-Term Stability of an MPS-8033/55 ASE Source • App Note 6: Choosing the Right Laser Diode Mount for Your
• LRS-9424 Heat Sink Temperature Stability When Chamber Door Application
Opens
• Measurement of 4-Wire Voltage Sense on an LDC-3916 Laser • App Note 8: Mode Hopping in Semiconductor Lasers
Diode Controller
• Measuring the Power and Wavelength of Pulsed Sources Using the • App Note 10: Optimize Testing for Threshold Calculation
OMM-6810B Optical Mutlimeter Repeatability
• Measuring the Sensitivity of the OMH-6709B Optical Measurement
Head • App Note 11: Pulsing a Laser Diode
• Measuring the Wavelength of Noisy Sources Using the OMM-6810B
Optical Multimeter • App Note 12: The Differences between Threshold Current
• Output Current Accuracy of a LDX-3525 Precision Current Source Calculation Methods
• Pin Assignment for CC-305 and CC-505 Cables
• Power and Wavelength Stability of the 79800 DFB Source Module • App Note 13: Testing Bond Quality by Measuring Thermal
• Power and Wavelength Stability of the MPS-8000 Series Fiber Optic Resistance of Laser Diodes
Sources
• Repeatability of Wavelength and Power Measurements Using the • App Note 14: Optimizing TEC Drive Current
OMM-6810B Optical Multimeter
• Stability of the OMM-6810B Optical Multimeter and OMH-6727B • App Note 17: AD590 and LM335 Sensor Calibration
InGaAs Power/Wavehead
• Switching Transient of the 79800D Optical Source Shutter • App Note 18: Basic Test Methods for Passive Fiber Optic
• Temperature Controlled Mini-DIL Mount Components
• Temperature Stability Using the LDT-5948
• Thermal Performance of an LDM-4616 Laser Diode Mount • App Note 20: PID Control Loops in Thermoelectric Temperature
• Triboelectric Effects in High Precision Temperature Measurements Controllers
• Tuning the LDP-3840 for Optimum Pulse Response
• Typical Long-Term Temperature Stability of a LDT-5412 Low-Cost • App Note 21: High Performance Temperature Control in Laser Diode
TEC Test Applications
• Typical Long-Term Temperature Stability of a LDT-5525 TEC
 #
3 Protecting Your Laser Diode

For application assistance or additional information on our products or

services you can contact us at:

ILX Lightwave Corporation

31950 Frontage Road, Bozeman, MT 59771-6310
Phone: 406-556-2481  800-459-9459  Fax: 406-586-9405
Email: sales@ilxlightwave.com

To obtain contact information for our international distributors and product

repair centers or for fast access to product information, technical support,
LabVIEW drivers, and our comprehensive library of technical and
application information, visit our website at:

www.ilxlightwave.com

Copyright  2003 ILX Lightwave Corporation, All Rights Reserved Rev01.060103