Transcript R&D Lab Safety for Physicists

Lab Safety for Particle Experimentalists
SLAC Course 110
• Martin Breidenbach
• June 2006
• December 2008
• January 2009
June 2006
Lab Safety for Experimentalists
M. Breidenbach
1
Lab Safety for Particle Experimentalists
• This is aimed at physics grad students and postdocs.
• The lab is the R&D lab – and not the accelerator or big
detectors.
– No accelerator hazards e.g. Radiation or magnets
• The lab is not particularly dangerous – you are more likely to
be hurt in traffic at the entrance – but there are hazards
that many of us have learned about through close
experience. The points here are offered to jump you over
these. There is not, and cannot, be a prescription to avoid all
risk. Thinking is required…Risk can be intelligently managed!
• Many accidents are the result of a chain of errors or
misjudgments. Reasonable precautions can make a single
error inconsequential.
• I will avoid graphic illustrations of various accidents.
However, the word horrible means exactly that. You would
not want to see the pictures. When horrible is used, I mean
it.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
2
Disclaimers
• This talk is not a substitute for any required training.
However, it does substitute for SLAC Course 251
• This talk is relatively dense and assumes you know basic
physics.
• You and your line management are responsible for your
safety. This talk is meant to expose you to some experience
in a concentrated dose in the hope that you wont do again
what many of us have learned by experience.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
3
Topics
•
•
•
•
•
•
•
•
•
Electrical
Electronic
Explosions
Implosions
O2 deficiency
Lasers
Chemistry
Falls
Radioactive Sources
June 2006
Lab Safety for Experimentalists
M. Breidenbach
4
Electrical Hazards
• Three routes to trouble:
– Electroshock
– Arc Flash
– Reflex
June 2006
Lab Safety for Experimentalists
M. Breidenbach
5
Electroshock
>~10 ma 60 Hz through the body is bad.
“let-go” threshold 10-17 ma
Chest paralysis (suffocation) ~30 ma
Cardiac fibrillation 75-100 ma
• Cardiac fibrillation – need defibrillator in <5 minutes, preferably <
2minutes. This level of shock unlikely but plausible in the lab. If
there is an AED within running distance, its good to know
where…(There is an AED in Central Lab Annex, 2nd floor, center
corridor)
• Body resistance not really prdictable – dry skin, <50 V, resistance
high
• Damaged or wet skin, 600 V, significantly lower resistance
• Neural damage, internal heating – very bad. But this is not really a
non-accelerator lab hazard at SLAC. Requires gross violation of
electrical safety.
• If you or someone else gets shocked more than trivially, get
professional help or call 911.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
6
Arc Flash
• : Electrical distribution system – including 480 VAC panels in
labs – can deliver enough current to make a horrible,
continuous arc releasing substantial energy and radiation (i.e.
an explosion) until some slow breaker opens. At 480, the arc
will ionize air and Cu and just keep going. So: Stay out of
480 circuits hot or cold. The required training, experience,
and PPE to work safely is unlikely for almost all physicists.
• Example: Consider a 480 V phase to phase fault with a
current of 10,000 amps. This is 5 MW. If the upstream
protection takes 10 cycles to open, this is 0.8 MJ. This is
equivalent to 200 g TNT.
• Arc Flash hazard is divided into categories between 0 and 4,
each with appropriate PPE. The PPE is rated by the incident
energy it can take before the onset of 2nd degree burns.
• Arc Flash Hazard 1 PPE is rated at 4 Cal/cm2 (17 J/cm2).
Note that this is much less than sunlight for an hour –
because of the large UV component of the arc spectrum.
• Above Arc Flash Hazard 0 is for professionals!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
7
Reflex
• Small shocks will cause an often large, more or less
involuntary startle reaction. It can make you fall – as in off a
ladder or platform.
• This kind of shock is classic when debugging proportional
chambers, drift chambers, etc. The current and stored
energy are usually too low to be an electroshock hazard.
Think about where a startle reaction might take you!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
8
Reasonable Electrical Practice
• Stay out of 480 circuits. ( note the period. Just stay out!!!)
(breaker operation exception described next slide)
• Stay out of 208/120 3φ panels-(breaker operation exception
next slide)
• Wiring a (disconnected!) chassis:
– Know the standard electrical color code:
•
•
•
•
Black (or possibly red or blue) is “hot”
White is “neutral”
Green is ground.
If, for some bizarre reason, you are forced to use cable that does
not conform, use properly colored shrink tube or colored tape to
identify the wires.
• Insulate the chassis connections so you cannot touch them when
you debug your work. RTV glob is the minimal acceptable insulation.
Fiberglass covers are better.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
9
Breaker and Disconnect Switch Operation
•
•
Occasionally it may be necessary to open or close circuit breakers
or switched disconnects.
It is reasonable for you to operate these devices at SLAC if all of
the following conditions are satisfied:
–
–
–
–
•
The NFPA70E Arc Flash Hazard Rating is 0 or -1.
The voltage is 480 V or less.
You are wearing appropriate PPE.
If you are closing a breaker, you reasonably understand why it was
open.
– No other jurisdiction forbids it (e.g. you are not at SLAC).
More details:
– NFPA70E is the counterpart document to the National Electrical Code
that deals with operations as opposed to construction standards. It
specifies hazards associated with various equipment configurations.
– In this context, operation of a circuit breaker or switch with covers on
is permissible if Arc Flash Hazard 0 (and in probably rare cases of <10
KA short circuit current, Arc Flash Hazard -1).
– Panels at SLAC should be labeled with their Arc Flash rating for covers
on and off.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
10
•
•
•
•
•
Most 115-208 V panels are Arc Flash Hazard 0. In rare cases, they
may be Class 1.
Personal Protective Equipment:
– For Arc Flash Hazard 0:
• You wear safety glasses.
• You wear cotton (or non-synthetic) long sleeved shirt.
• You wear cotton (or non-synthetic) long pants.
Technique: when operating a breaker or switch, stand to the side
close to the wall and look away. If there should be an arc, you wont
get it in the face.
Reasonableness: If a 15 or 20 Ampere 115 or 208 V breaker tripped
because of an overload that is understood and corrected, resetting
is reasonable. If a 480 V breaker tripped mysteriously, leave it to
an electrician.
Some equipment, such as welders, connect with plugs and sockets
that are mechanically interlocked so that the plug can not be
inserted or removed with the switch on. Ensure that the equipment
is off before operating the switch. (Some older sockets do not
have this interlock, again ensure that the equipment is off before
inserting or removing the plug.)
June 2006
Lab Safety for Experimentalists
M. Breidenbach
11
Clean Room Issues
• There are many clean rooms in use – e.g. EXO, Si Lab, and
GLAST. Clean room clothing may be a problem for some
activities: Electrical switching and TIG welding are the two
prime examples.
• PPA has tested the Tyvek clean room suits, and they are
reasonably ok. (The nylon zipper will burn, but the Tyvek
does not)
• Nitrile gloves burn sufficiently to be a clear hazard.
• We have not found an elegant solution for a clean,
reasonably non-flammable glove.
• The working solution is a deerskin welding glove under nitrile
gloves.
• The search continues for something better.
• Remember that welding requires a fire watch person with a
CO2 extinguisher.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
12
Sidebar – SLAC Electrical Power Distribution
• The overhead transmission lines coming down from Skyline
are 230 KV 3 φ with a capacity of ~100 MW.
• The Master Sub has transformers taking the 230 KV to 12.6
KV for distribution around the site.
– 60 KV lines come in from campus and are used when the 230 KV
line is unavailable.
• Substations usually reduce the distribution voltage to 480 V
to panels in buildings.
• Smaller transformers take the 480 V to 208 V.
In a 3 φ system, the stated voltage refers to the
line to line voltage. The line to neutral voltage is
down 1/√3. So the standard 120 V is line to
neutral of a 208 system. The system is often
referred to as 208/120 Volts.
• Breaker panels are used to distribute these voltages.
• Again: Stay out of the lab power distribution system!!!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
13
Electrical Issues
• It requires a lot of paperwork to work hot on a chassis with
exposed 120 VAC and it is an unnecessary risk, so insulate
those connections to the power supply.
• Grounding limits the potential of a (conductive) device with a
fault to (assumed grounded) you. A ground is effective only
if it can carry enough current to trip a breaker and/or
reduce the potential to non-hazardous levels.
– It in a worst case fault, the impedance of the ground
connection should keep anything you might be touching below
50V to ground!) Assume you can get 100 amps out of a wall panel
before a small breaker opens. Then you need < 0.5 Ώ to a solid
ground.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
14
Sidebar - GFI
•
A Ground Fault Interrupter
(GFI) compares the current on
the hot and neutral by running
both through a toroid
transformer and amplifying the
difference. An imbalance of 4
to 6 ma triggers the spring
mechanism of the breaker (or
outlet). A GFI breaker has no
ground connection! (A GFI
outlet has a ground connection
for the 3 wire cord.)
•
Note that 100V x 10 amp x 5
mS = 5J, which will hurt.
However, 10 amp through you
is unlikely (at 115 V), unless you
are in salt water.)
June 2006
Lab Safety for Experimentalists
M. Breidenbach
15
Typical SLAC configuration:
Grounding
480 V Building distribution system
feeds 480 V to 208/115 V
transformer to breaker panel.
(Typical residential configuration:
Primary:
Distribution
Voltage – 412 KV
480 V, 3φ,
Delta
Residential Transformer
Secondary 220 V, Center
tapped neutral
208/115 V, 3φ,
Wye
230 volt single phase center
tapped pole transformer feeds a
breaker panel)
Breaker Panel
The center tap connects to the
neutral bus, and is grounded at the
panel. There is usually a separate
ground bus.
Neutral Bar
Ground Bar
Outlet
Note that the motor frame, or in
general, any accessible conducting
parts of a device are connected to
the ground wire.
Double insulated devices are
considered safe enough not to
have a ground connection.
June 2006
Lab Safety for Experimentalists
Normal Motor
M. Breidenbach
16
Grounding – Current flow during a short
Note the short from the hot side
of the motor winding to the frame.
480 V, 3φ,
Delta
208/115 V, 3φ,
Wye
If the short impedance is low
enough, enough current (dashed
lines) will flow as shown to trip the
breaker.
Breaker Panel
In any case, the impedance of the
ground wire should be low enough
to keep the potential difference
between the frame and building
ground below dangerous levels
(50V).
Neutral Bar
Ground Bar
Outlet
The same strategy applies to most
laboratory equipment, where the
ground conductor system must be
adequate to keep the frame
potential below 50 V.
June 2006
Lab Safety for Experimentalists
Short to motor
frame
Abnormal Motor
Person Touching Frame
M. Breidenbach
17
Wire Sizes
Wire must be sized to prevent excessive heating and voltage drop. Reasonable
practice with insulated wire is in table. Voltage drop ~5% usually considered
ok – but note 20 amperes in #12 wire is ~30 meters (out and back).
Use proper extensions! The rule of thumb is #12 for 20 amperes, #14 for 15
amperes, and #16 for 10 amperes. However long extensions may need heavier
wire. Calculate for a 5% voltage drop or less.
Do not daisy chain extensions.
AWG Wire Size
(solid)
Diameter
(inches)
Resistance
(Ohms/Km)
Nominal Current
Capacity
(Amperes),
Insulated Cable
18
0.0403
20.9
5
16
0.0508
13.2
10
14
0.0640
8.28
15
12
0.0808
5.21
20
10
0.1019
3.28
30
June 2006
Lab Safety for Experimentalists
M. Breidenbach
18
Non Contact Voltage Detector
• These inexpensive devices ($10-20) capacitively sense AC
voltages. All do 115V, 60 Hz AC, and some do 24 VAC. Very
useful for homework!
• The sensors do not detect DC, and may not be used at SLAC
as verification of the zero voltage state.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
19
Capacitors store Energy
• “Doorknob” capacitors
• Ceramic dielectrics such as Strontium Titanate – give 2.7 nF
@30 KV or 1 J – (Nasty) in a device 2 3/8” x 7/8”
Pulse Capacitors
100 μF @ 2KV or 200 J – Totally deadly
Device size ~3 ¾ x 4 ½ x 7 3/4
Large Capacitors should be shorted when not in use
Bleeder resistors should not be trusted to discharge a capacitor.
You need more training to work with large pulsers.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
20
Electrical Limits
• There is a variety of advice on what is dangerous:
– SLAC EHS Manual Chapter 8
– NFPA 70E – National Fire Protection Agency
– DOE Electrical Safety Orders
• Below 50 V is ok.
– There is serious burn hazard with high current supplies at even
a few volts. Car batteries can deliver 400 Amps without
blinking. So that neat ring can dissipate 4 KW and amputate a
finger painfully and quickly. Experienced people remove jewelry
around car batteries and VME (or Fastbus) power supplies.
– Some sources claim stored (capacitive) energy > 10 J (below 50
V) is a hazard. Most of us are more excited about the burn
possibilities of supplies that can deliver more than ~10 Amp…
June 2006
Lab Safety for Experimentalists
M. Breidenbach
21
Electrical Limits. continued
• Above 50 V
– Below 5 ma power supply capability – you can’t fry yourself. But
you can jump.
– The 10 J limit is becoming very real!
• Below ~ 250 V it’s hard to electrically puncture the skin, so the
body impedance makes it hard to deliver 10 J. Note that this is for
a dry, intact body! By 500 V, its only what the circuit can deliver.
Believe!
– At high voltage, you will probably survive 10 J, but you will
remember it until Alzheimer’s takes over. Be real careful when
there’s more than 1 J.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
22
Sidebar – Control of Hazardous Energy
• Formerly known as LOTO – Lock Out – Tag Out
• Work on de-energized equipment connected to wall (no plug
to pull) by COHE procedures. This is hardly ever for
physicists. But its good to understand the issues:
– The idea is that upstream disconnects (breakers, knife
switches, fuses) must be open. And locked open. And tagged
with locker’s name etc. There’s a course on this.
– But how do you know you got the right breaker? Sometimes its
obvious, but often the panel is far away, and there is no visible
conduit from the load to the breaker. So you measure the
voltage. That’s hot work. And that requires PPE for anything
that might go wrong – like using a bad meter that presents a low
impedance to the bus (and a horrible arc). And you need a Hot
Work Permit. Which PPA has never granted – at least in the last
several years. So Fugeddahaboutit.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
23
•
Electronics
Most modern signal processing electronics – from charge amplifiers through
computing – are harmless. You can diagnose the circuit being reasonably
confident that you are far more likely to blow up the FET through
Electrostatic Discharge (ESD) than it is to tickle you.
–
•
•
•
When you change boards in your computer, turn it off but leave the AC connected.
If you are not using an ESD wrist strap, hold the board in one hand and touch the
cabinet with the other before inserting board. Etc.
Silicon detectors may have bias supplies up to ~500 V (for high radiation
damage environments). Its rare that the supply can deliver 5 mA. Its almost
weird that the stored energy could approach 10 J. Special precautions are
needed, including paperwork, if the supply can deliver more than 5 mA.
Photomultipliers operate at 2-3KV. There are many older “bulk” supplies
designed to power many PMT’s that can deliver 20 mA or more. These are
serious supplies. Almost always, the HV is delivered in co-ax (RG-59), and
the co-ax is terminated with modern HV connectors (MHV, Reynolds) that
make it exceedingly difficult to accidentally contact the HV. However: If
you need to debug a PMT base, use a NIM bin Power Supply that has a max
current of 1 mA or less.
Avoid adapter cables that change HV connectors into non-HV connectors,
and especially forget HV cables that have alligator clips on the ends. They
are called suicide cords for a reason!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
24
Electronics Technique, basic!
• Remember to turn off the HV before sticking your hands
inside. (Remove plug or lock off the equipment if not cordconnected*).
• Remember to discharge capacitors
• Use proper grounding. The grounding of a PMT base or LST
or most other detectors than Si are usually through the HV
co-ax.
• Be particularly careful if you have to work in the dark – e.g.
searching for light leaks for PMT or APD based counters.
•
*This comment sounds simple, but it isn’t. EXO refrigerators and
compressors are connected with very expensive plugs and sockets
to avoid locking issues!!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
25
Sidebar – RG59
• Note that the capacitance of most random co-ax is ~
100pF/foot. You can easily destroy electronics with a
disconnected cable that charges up to a few KV. For higher
HV, a charged cable is dangerous. (100 ft @100 KV ~ 50 J)
• The dielectric in most co-ax is polyethylene. Poly is a good
dielectric, but it’s chemically close to napalm. A few cables in
the lab are no problem, but a rack full is a serious issue.
SLAC has had two(!!) serious fires that started from minor
arcing in co-ax. The SPEAR 1 SLAC-LBL Magnetic Detector
(aka Mark I) had spark chambers. In the 70’s, the pulsers
ignited a very exciting fire. Aluminum racks melted. Months
of work to rebuild. In the 90’s, ion pump HV cables started a
cable fire in the SLC e- damping ring. Again, a major mess
and months of expensive recovery. Bromated polyethylene or
teflon dielectric is a little more expensive, but much safer…
June 2006
Lab Safety for Experimentalists
M. Breidenbach
26
Electronics, continued
• Laser supplies are serious. The flashlamp supplies often
break the 10 J limit. If you open laser enclosures, you need
laser safety training, but remember the HV basics:
– The power supply should be disconnected from the wall
– The energy storage capacitor should be grounded with a ground
hook.
• A SLAC laser will (most likely??) have stored energy far below 1 KJ.
• A simple ground hook without a series resistor is acceptable.
• Two ground hooks are needed to discharge “floating” capacitors.
• Supplies for Pockel’s Cells can be hefty.
• There are occasionally high voltage low impedance
operational amplifiers (Trek) that are lethal. No hot work on
these guys, and make sure the load is enclosed.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
27
Electronics Techniques
• On occasion, it is necessary to debug a circuit that can hurt.
If it has > 50 V and (5 mA or 10 J), there are hot work
(energized circuit )requirements – permits and non-routine
JHAM’s. But there are seatbelts for this car:
– Make sure you are floating at high impedance to ground. A
dielectric mat on the floor, or dry wood for a ~few 100V is
good. ESD wrist straps are relatively high impedance, so
delicate components are protected but you are not grounded.
– Use one hand. If you touch something, make sure it will be
finger to wrist or less. Pull your hand out of the chassis when
adjusting the scope. Eliminate the potential of a hand to hand
or hand to foot shock. And what is the chair made of?
– Think about what will happen if you are (very) startled by a
shock.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
28
High Voltage Connectors
Reynolds 10 KV
Note non-recessed pin. Use
for low level signals only!
SHV 5 KV
June 2006
BNC 500 V
Lab Safety for Experimentalists
M. Breidenbach
29
Miniature Connectors
Note recessed pin
Lemo HV 1500 V
Standard Lemo Signal only
June 2006
Lab Safety for Experimentalists
M. Breidenbach
30
More Electronics Advice
• If its >50 V, make sure somebody else is around. This is
particularly true if you are debugging a drift chamber in the
detector!
• If you get zapped, get checked out by medical. Too bad that
you are embarrassed, don’t make it worse.
• Might be a good idea to take that CPR course, and know
where the AED is.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
31
Explosions
•
Explosions are all about stored energy… In the lab the prime suspect
is the gas bottle. The standard K bottle is 200 SCF at a pressure of
2200 PSI. Or V=42 liters and P= 147 Bar =15x106 Pascal
•
•
U~PV/(γ-1) (for expansion to 1 bar) where γ=Cp/Cv
U~0.9 MJ for a monatomic gas (e.g. helium or argon), U greater for
nitrogen, oxygen, CO2.
Scale: 1 gram of TNT ~ 4. 2 KJ
So gas bottle is ~ ¼ Kg TNT!!!!
•
•
•
– (For reference, a jelly donut is ~200 Calories (note those food calories
are Kcal = 0.8 MJ, but at least jelly donuts don’t explode rapidly)
Breaking the valve stem of a gas bottle is a big deal. It’s a deadly
rocket. Handle bottles carefully!
– They must have their valve cover on when not in use.
– They must be strapped to a solid support at two heights to prevent
tipping.
– They must have a proper regulator to control the output pressure.
– Wear safety glasses if there is any possibility of a gas jet to your face!!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
32
Gas Bottles, continued.
•
•
•
Gas bottles valves often have different threads to prevent
inappropriate regulator use. For example, O2 regulators and
pressure gauges and plumbing must be oil free. Be sure the
regulator is correct, and don’t force the threads! Flammable gas
bottles usually have left-handed threads.
Occasionally there are even higher pressure bottles. EXO uses
6000 PSI argon for Joule-Thompson refrigeration. These bottles
have their own special regulators.
Some gases burn or explode. There is a very strong trend to use
non-flammable gases for bulk applications – such as the Babar
LST’s. However, isobutane often is a component of these gases, and
might be used when developing a mixture. Hazardous gas detectors
are used where there is a chance of a leak. These detectors are
installed and maintained by EFD. They warn at a modest fraction of
the Lower Explosive Limit (LEL) and must be heeded. Horrible
accidents have happened to HEP experimentalists with flammable
chamber gases.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
33
Cryogenic Fluids
•
•
•
•
•
•
Cryogenic liquids expand when they warm up.
– Argon at STP is x860 liquid volume
– Xenon at STP is x550 liquid volume
– Nitrogen at STP is x710 liquid volume
If a cryogenic liquid warms up, the pressure will increase. In a properly
designed system, the volume of liquid is limited so that the warm system can
handle the pressure. In addition, there should be relief valves and/or burst
disks on any plumbing segment which can be isolated by the valves. Very few
pressure systems can handle more than 2000 PSI, most much less. Weak
links are usually windows, bellows, and feedthroughs.
Simple safety principle for cryogenically recovering a gas into a pressure
bottle – e.g. recovering xenon: Never dunk a recovery cylinder in liquid
nitrogen for more than half its length.
Wear safety glasses.
Use cryogenic rated gloves when pouring LN. Don’t spill LN into your shoes!
Cryogenic “burns” are serious.
Cryogenic fluids can cause oxygen deficiency hazards as they vaporize.
Ensure good lab ventilation when using LN for cooling. Be aware of the
potential for nanoclimates – e.g. your head under a light blocking cloth.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
34
Implosions
• PMT’s - Work in the lab may involve large photomultiplier
tubes. The tubes are made of relatively thin glass, and are
evacuated. Breaking them causes glass to be projected with
high velocity. In certain situations, the shock wave from an
imploding tube can trigger adjacent tubes. Always use
goggles or safety glasses when handling these tubes. And
these PMT’s are typically quite expensive!
• Thin Windows – Large thin windows for vacuum systems are
rare in the R&D lab but common around accelerators. If the
window was aggressively designed to limit multiple
scattering, it can be quite delicate – and the shock wave
from an implosion is serious. On most lab scale apparatus,
breaking a window will “only” take out equipment…
June 2006
Lab Safety for Experimentalists
M. Breidenbach
35
O2 Deficiency
Altitude Effects on O2
800
100
90
700
80
Pressure (mmHg)
600
70
500
60
400
50
40
300
Barometric Pressure (mm Hg)
P02 in air
Po2 in Alveoli
Arterial O2 Saturation (%)
30
200
20
100
10
0
0
10000
20000
30000
40000
50000
0
60000
Altitude (ft)
As the partial pressure of O2 drops, so does arterial O2 saturation. Judgment
may be impaired first, but loss of consciousness occurs without warning.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
36
O2 Deficiency
•
•
Normal air is ~21% O2. Most O2 deficiency monitors alarm at ~19%.
When working with cryogenic fluids, O2 deficiency can be a hazard
if there is a spill. Remember that volume change of ~700.
– Example: The SLD calorimeter had 50,000 liters of liquid argon. In a
worst case (but inconceivable) spill, the heavier than air argon expands
by x 860 and produces enough gas to fill the full CEH pit almost twice!
•
•
•
If there is a large spill – get out! A 200 liter LN dewar holds a
quite serious amount of gas.
Labs where there is a potential for a leak or a spill should be
equipped with O2 deficiency monitors. These should give early
warning and summon the Fire Department.
In a small space, it is easy to displace enough air to be dangerous.
O2 deficiency is perhaps the major hazard of Confined Spaces, and
a special permit and training is required to enter a Confined Space.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
37
Lasers
• Some of the lasers in the lab have sufficient power to
permanently damage the retina.
• Further, UV and IR lasers can’t be seen and can do damage.
• The primary level of control is containment – there should be
no laser light scattering around the lab from Class IIIb or
IV lasers. Some lasers contain the beam in optical fibers
with proper light tight terminations at both ends. Never
operate these lasers with the fiber removed (at either end!)
in a non laser-safe lab.
• Advanced training is required for work with these lasers.
Laser goggles must be selected for the particular laser –
“one size does not fit all”.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
38
Sidebar - Laser Classifications (Loosely)
•
Class I
•
Class II
•
Class IIIa
•
Class IIIb
•
Class IV
–
–
–
–
–
June 2006
Cannot cause eye damage either because <0.4 μW CW visible; or completely
enclosed. Note that if the enclosure is breached, controls for the native laser
power class are required. CD players, laser printers, etc
Cannot cause eye damage during the aversion response (0.25 sec) (aka blinking).
Only visible (400-700 nm); 0.4 μW < P < 1 mW (CW). Usually He-Ne lasers, laser
pointers, range finders, etc
Cannot cause eye damage during aversion response. Injury possible with optics or
staring into beam. Visible, 1 mW < P < 5mW CW. Laser pointers, laser scanners, etc
Can cause injuries from viewing direct beam or specular reflection. 5 mW < P < 500
mW CW. Diffuse reflection will not cause injury unless light collected by optics.
Spectrometry sources, etc. Eye protection required.
Primary beam, specular and diffuse reflections can injure eyes and skin. Also can
ignite flammable material. All wavelengths with P > 500 mW. All pulsed lasers that
the eye can focus (400 nm – 1400 nm). Significant controls and eye protection
required.
Lab Safety for Experimentalists
M. Breidenbach
39
Chemistry
•
•
•
•
•
This is not a general chemical hazards review, but a few special cases that
come up often. Use appropriate precautions and PPE. Material Safety Data
Sheets (MSDS’s) should be first order check.
Cleaning:
– Ethanol and acetone are often used for cleaning UHV and other
components. Both are serious inhalation, transpiration, and fire hazards.
Ensure good ventilation. A vapor hood is required if the quantities
approach a liter.
– Make sure you know a fire extinguisher location.
– For quantities more than a squeeze bottle squirt, wear appropriate
gloves.
Epoxies:
– The unreacted components of many epoxies are quite irritating. Wear
nitrile gloves.
Scintillation Phosphors – Occasional use is made of organic scintillators in
their raw form. These chemicals may be toxic. Check the MSDS!
Some chemicals (perhaps unlikely that you will encounter them) absolutely
need special training and facilities:
– Dangerous liquids: e.g. Be solutions, HF
– Dangerous gases: e.g. Arsine, Chlorine, Bromine
– Forget about it: e.g. Methyl Mercury
June 2006
Lab Safety for Experimentalists
M. Breidenbach
40
Vacuum Systems
•
•
•
•
Laboratory UHV systems may be pumped with turbopumps or ionpumps.
Turbopumps are somewhat delicate, but present few personnel hazards.
Ion pumps are reliable, moderate cost devices but operate at substantial
voltage levels. (The controller shown here will put out 3 KV at 7 ma.)
The HV cables of modern pumps are reliable and safe, and some modern
controllers shut down when the cable is disconnected. In general, there
should be an independent ground connection between the supply and the
vacuum plumbing, and the supply should be turned off before disconnecting
the cable.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
41
Vacuum System Baking
Metal vacuum systems often need to be baked to drive off
water and other contaminants. Temperatures may be as low
as 75 C for delicate internals, and up to 400 C for a
“serious” bake.
•Heating is often done with “Heating Tapes”, glass
insulated resistance wire.
•Do not exceed the tape temperature rating.
•Make sure the controller includes a GFI.
•Ground the vacuum system. Variacs are often used to
control the voltage to the heaters. Note that Variacs
are not transformers, and do not isolate the line.
–Make sure the stainless is substantially (openly
obvious to a casual observer) grounded.
•Check for fire hazards.
•Be aware of burn hazards!
June 2006
Lab Safety for Experimentalists
M. Breidenbach
42
Falls
•
•
•
Strangely enough, slips, trips and falls are the most likely
accidents. Falls can be quite serious.
In the lab, there may well be A ladders, but extension ladders and
scaffolding are unlikely. (Not the case near a detector)
The classic ladder accidents:
– Going on or above the penultimate step of an A ladder.
– Using the top half of an extension ladder by itself.
•
•
•
•
•
Try to tie extension ladders so they can’t slip.
Think about the surface supporting the ladder!
Don’t over-reach. Bad things happen when the Center of Gravity is
not over the base. Get down and move the ladder instead.
Fall protection or barriers are required on elevated work surfaces.
Be particularly careful of situations where the involuntary reaction
to a (small) shock can initiate a fall.
June 2006
Lab Safety for Experimentalists
M. Breidenbach
43
Radioactive Sources
•
•
•
•
•
It is assumed that you have some knowledge of nuclear physics…and
that we will not talk here about accelerators or accelerator induced
radioactivity.
Types of sources:
– α particles have ~no range and are stopped by the skin (unless they get
inside)
– β’s ionize immediately, but usually do not have the range to do damage.
– γ’s go some distance before Compton scattering or photoelectric effect
kicks out an e- which ionizes internally.
Most lab sources are modest hazards if they are not ingested or
inhaled, usually meaning they are sealed:
– Nanocuries to microcuries should not be carried in your pocket.
– 100 microcuries is a serious source, but still can be handled in the lab.
– Millicuries and above need help from Operation Health Physics.
Occasionally an unsealed source is needed when the recoil nuclei are
of interest, or a liquid solution is needed to, for example,
electroplate a source. Special handling procedures are required,
and OHP must be brought in.
For approximate point sources, dose will go ~ 1/r2. Even smaller
sources can cause unwanted doses as r gets small…
June 2006
Lab Safety for Experimentalists
M. Breidenbach
44
Sidebar – Units1
• SI units are recommended, but not yet in common use.
• Unit of Activity: Bequerel; 1 Bq = 1 disintegration/sec
– The Curie (Ci) = 3.7x1010 Bq
• Unit of absorbed dose: Gray; 1 Gy = 1 joule/Kg
– 1 Gy = 100 rad (There are lots of survey meters around
calibrated in rad’s, and occasionally even the (obsolete)
Roentgen.
• {The Roentgen (R) measures the charge produced by γ’s showering
in air. 1 R = 2.58x10-4 coul/Kg}
• Unit of equivalent dose: Sievert; 1 Sv = 1 Gy x wR
– wR = radiation weighting factor (was Q = quality factor in
oldspeak)
• wR = 1
• wR = 1
• wR = 20
•
1
X and γ rays, all energies
electrons and muons, all energies
alphas
– The old unit is the REM; 1 Sv = 100 REM
Mainly taken from Review of Particle Physics (2004)
June 2006
Lab Safety for Experimentalists
M. Breidenbach
45
Radiation Scales1
• Recommended limits for Radiation Workers:
– CERN:
– U.S.
– SLAC
15 mSv/year
50 mSv/year
15 mSv/year
• Lethal dose: (LD50, no medical treatment) 2.5 – 3.0 Gy
• Natural background: 0.4 – 4 mSv/year
• Flux to deliver 1 Gy ~ 6.24x109/(dE/dX) charged
particles/cm2
• So it should be obvious now why a Ci is a big source.
• It is assumed that you have GERT (General Employee
Radiation Training) . It is possible but unlikely that you will
need RWT1 training. RWT2 training is for contaminated
locations – not our labs!
•
1
Mainly taken from Review of Particle Physics (2004)
June 2006
Lab Safety for Experimentalists
M. Breidenbach
46
Coda
• SLAC’s PPA Safety Officers are Frank O’Neill, Joe Kenny
and Sandy Pierson
• They may not know the answer to all your safety questions,
but they usually can provide good pointers. Talk to them!
• Think!
• If there is a problem requiring emergency help – call 911
from a SLAC phone, or 911 from a cell phone (assuming there
is a signal). You will need to describe your location - obvious,
but do you know the Building Number?
June 2006
Lab Safety for Experimentalists
M. Breidenbach
47