From Thermionic Studios
Jump to: navigation, search

Electrical Safety




Safety Tools

  • Rubber-soled Shoes
  • Isolation Transformer
  • Variac
  • Lightbulb Limiter
  • Capacitor Drain Jumper


The proper grounding scheme of a vacuum tube amplifier has been the subject of many debates, blogs, forums, and web pages. There are actually two important aspects related to proper grounding, hum reduction, and safety. The safety aspect is often overlooked, or ignored in the quest to eliminate hum. With careful design it is possible to satisfy both criteria simultaneously.

The important safety aspect that I must reinforce is the fact that EVERY CONDUCTIVE PART THAT THE USER CAN TOUCH MUST BE GROUNDED. This is usually fairly easy to do, and if done right will help reduce hum.

One of the highly debated aspects of the above statement is whether or not to ground the secondary of the output transformer. There are people who strongly believe that grounding the cold (black lead) side of the output transformer causes a degradation in sound quality. I find no technical reason for this belief. There are strong safety reasons why the secondary and the metal housing of the transformer should be grounded. It is possible for a short to develop between the primary winding and the secondary or the metal core of the transformer. A short of this type may not adversely affect the operation of the amplifier, however it could cause a high voltage (several hundred volts) to appear on the speaker leads. If a user touched the speaker wiring and a grounded object at the same time, he could be fried. This is very possible on a guitar amplifier. In my 40 years of working with tube amplifiers I have seen 3 shorted transformers, 2 were in guitar amps. The guitar amps blew the fuse or the rectifier tube because the transformers were properly grounded. The custom built high end stereo amp was working normally until I found the transformer problem. There was over 200 volts measured between the speaker jacks, and the input connectors on this amp. This is more common in vintage transformers because they used paper insulation, and paper absorbs moisture.


Safety — Electrical Safety

Please read this page even if you are an experienced technician! Read it twice if you have never worked on vacuum tube electronics before. If you don’t understand EVERYTHING presented here have someone who does understand it perform all of the testing on your vacuum tube project. PLEASE READ the RESPECT FOR ELECTRICITY sections, then examine your work space carefully. Correct any deficiency (or at least understand the risks) BEFORE working with vacuum tube electronics. The importance of electrical safety when working with potentially lethal voltage.

It may seem that I have devoted considerable space to electrical safety. I must assume that whoever reads this is actually considering working with potentially lethal levels of electricity, possibly for the first time. Therefore the topic of safety is of extreme importance. There are countless web sites and books devoted to the topic of vacuum tube (and other) electronics. Very few of these have any information on electrical safety. I personally attended a three year electronics program at a vocational high school during the vacuum tube years, a 2 year technician program in a community college, a 4 year engineering program at a university and a masters degree program at a different university, and electrical safety was NEVER mentioned. That leaves the new technician or engineer to learn on his own. This could possibly have disastrous results. The technician or engineer graduating today may spend his entire career working on computer or cell phone circuits powered by 3 to 5 volts where the threat of serious electrical shock is almost nonexistent. This is definitely NOT THE CASE WITH VACUUM TUBE ELECTRONICS. Almost any vacuum tube circuit, even those powered by batteries, have the potential to cause death if proper precautions are not taken. What happens when a human (or any living animal) gets shocked?

At one time or another most of us have experienced some form of electric shock, where electricity causes us to have an unpleasant experience. If we are fortunate, the extent of that experience is limited to tingles or jolts of pain. Many have noticed that a more severe shock, even from static electricity will cause involuntary movement, which can cause secondary accidents. When we are working around the voltage levels present in vacuum tube circuits, electric shock becomes a much more serious issue.

As electric current is conducted through ANY material heat is generated. The amount of heat is determined by the voltage present and the resistance presented by that material. Since different tissues in the body have different resistance levels (determined by their moisture and mineral content) it is possible for current passing through the body to cause heat damage to internal organs or tissue. This heating effect is one of the major source of permanent physical damage in cases of severe electrical trauma such as lightning or high voltage power line contact, if the victim survives.

One of the more significant effects of electric current on the body is disruption of the body's central nervous system. The central nervous system is the network of cells in the body responsible for control of most body functions. The brain, spinal cord, muscles, and most of the organs in the body function like a complete electrical circuit to allow it function. Nerve cells communicate with each other by creating tiny electrical signals (very low voltages) in response to neurotransmitters, and releasing neurotransmitters when stimulated by tiny electrical signals. Any electric current passing through the body can override the tiny electrical signals normally generated by the neurons, prevent the system from operating normally. Electrical current passing through a muscle, or the nerve cells controlling that muscle, will cause it to involuntarily contract. Current of sufficient magnitude will override the body's attempt to relax that muscle.

This problem is serious if the victim grabs an energized conductor with their hands. The muscles responsible for closing the hand tend to be stronger than the muscles responsible for opening the hand, and so if both muscles contract because of an external electric current, the closing muscles will win, causing the hand to be clenched into a fist. This clenching action will force the hand to grasp the conductor firmly. The victim will usually not be able to let go of the conductor. This situation can only be interrupted by stopping the current through the victim.

Even when the current is stopped, the victim may not regain control over their muscles. This effect is usually temporary and lasts for a few minutes. This principle is used in stun guns which shock the victim with high voltage pulses similar to that delivered by an automotive ignition coil. Recent incidents publicized by the news media reveal that even the controlled shock delivered by a stun gun can cause permanent disabilities and death in some people, particularly those with heart problems.

Electric current is able to affect any muscle in the body. Of particular concern are the diaphragm muscle controlling the lungs, and the heart muscle itself, These too can be paralyzed by electric current. Even currents too low to cause paralysis are often sufficient to disrupt nerve cell signals so that the heart cannot beat properly. This can result in a condition known as fibrillation. A fibrillating heart flutters rather than beats, and is incapable of pumping blood through the body.

Electric current that passes through the chest area is capable of causing death due to asphyxiation or cardiac arrest. With this in mind you must not put yourself in a situation where current could pass from hand to hand or hand to foot. The path of left hand to right foot is the worst case. This is why many texts will tell you to work with one hand in your pocket.

I personally can't keep one hand in my pocket, so I will power up a new circuit for the first time with my meter (and other test equipment) already connected to it, from a safe distance away. In the case of the 845SE amp which operates at 1200 volts, I was 6 feet away behind a thick piece of Lexan in case anything exploded. When probing a circuit with a meter I always have the black meter lead connected to ground with a clip lead, and I use one hand to touch the red lead to the circuit under test. The other hand is on my waist or behind my back. My feet (and the rest of my body) are well insulated from ground. Electric Current Path Required to Deliver a Shock

Electricity requires a complete circuit to continuously flow. This is why the shock received from static electricity is only a momentary jolt. The flow of electrons only occurs until the static charges are equalized between two objects.

Without two contact points on the body for current to enter and exit, there is no hazard of electric shock. This is why birds can safely rest on high-voltage power lines without getting shocked. There is only one point of contact since the bird is standing on the wire and not touching any other object, including the ground. If the bird (or any live creature) were to touch the wire and the ground at the same time it would be fried. Even though the ground (dirt, or asphalt) is a very poor conductor of electricity, power lines can carry several thousand volts. Even a poor conductor is capable of passing a lethal amount of current with enough voltage. It should be noted that concrete can also conduct enough current to provide a lethal ground path with voltages as low as 110 volts.

The same situation is possible at the voltage levels present in a vacuum tube circuit. If you are stand on the ground even wearing shoes you can still be grounded well enough to receive a lethal shock if you only come in contact with one energized conductor. There are actually two points of contact here since your shoes are not a perfect insulator. The path of current will be from your hand through your body to your feet.

It is very important to remember that there are very few good insulators in the world and the surroundings tend to degrade the good ones. A good pair of shoes (especially with rubber or plastic soles) should be a good insulator, but sweat is a very good conductor. Your feet never sweat? How many times have these shoes been worn in the rain? The floor in your work area is also important. Damp concrete (like in a basement) is especially dangerous. The humidity in your work area can make you and other things like carpet more conductive. The point here is that you must be aware of any possible points of contact with electric current. The circuit you are working on is one, if you are probing around in it with a test lead (or scope probe) be aware of where your other hand and both feet are. Don't reach for the knob on the test instrument with the other hand, especially if it is an older metal one. If the insulation on the test lead were to fail, you could be fried. Put down the scope probe to adjust the scope. The idea is to only touch one thing at a time. Here again if one hand IS in your pocket, you are forced to follow this rule. So, how much electricity does it take to fry me?

I have been asked several times " How much voltage does it take to be lethal?" The answer is - it depends! Depends on what? Mostly on the surrounding conditions. Another common phrase is: "It's not voltage that kills, it's current!" While there is some truth to this, there is a lot more involved. Have you ever seen a sign that said, DANGER -- HIGH CURRENT! I didn't think so. The reason for this is that the likelihood for a lethal current increases as the available voltage goes up.

The statement that current kills is essentially correct. It is electric current that causes burns, paralyzes muscles, and stops breathing and heartbeats. However, electric current doesn't just happen. Ohm's law says that a certain current will flow given a voltage and a resistance. raise the voltage and current will increase. Lower the resistance and current will increase. The resistance of a persons body varies greatly. It varies widely from person to person, time to time, and dependant on several other variables. Therefore the voltage required to produce bodily harm depends on the total circuit resistance, the resistance of the victim and the rest of the path. The path resistance is determined by the external (to the body) factors as well as the resistance of the electrical path through the body of the shock victim.

The resistance of the body is not a constant. The resistance is highly dependent on a persons hydration. It is also dependent on several other factors. Body resistance also varies depending on how contact is made with the skin. Smooth, soft skin is more conductive than rough, hard skin (elbows, fingertips). Blood, sweat and tears being rich in salts are an excellent conductor of electricity. Thus, contact with electricity made by a sweaty hand or open wound will offer much less resistance than contact made by clean, dry skin. Using an old analog meter (digital meters do not always give accurate body resistance readings) I have measured fingertip to fingertip (on oposite hands) readings as high as 800K ohms indoors an air conditioned building with low humidity. On the other hand after working outside all day in hot Florida weather and drinking lots of Gatorade, I have measured resistance as low as 2500 ohms from armpit to armpit. It has been suggested that a resistance as low as 1000 ohms is possible with the right conditions.

But how much current does it take to fry me? There again, it depends. The largest variable is the path the current takes through the body. It is likely that enough current to burn the skin off could be applied between two adjacent toes without causing death, since none of the current flows through any of the body's vital organs. If all of the current flows through the chest cavity how much is dangerous? I suppose that number varies widely from person to person. I did some searching for numbers. How do you really know how much current actually flowed in a deadly situation? I found numbers form 10 to 100 milliamps on the internet. I looked through my collection of electronics books (some of which are quite old) and found only two mentions of electrical safety. A WW2 vintage Army training manual says that 15 to 25 milliamps could be lethal and that any voltage above 48 volts should be considered lethal. A modern college textbook has two whole pages on safety and claims that 10 milliamps could be lethal under some circumstances. I also found references to AC current being more likely to disable the heart, while DC is more likely to cause paralysis. Other data says that AC is twice as likely to kill you.

Given the data we have so far we can see that the worst case 1000 ohms resistance and the worst case 10 milliamp lethal current, ohms law gives us a deadly voltage of 10 volts. I have never heard of someone actually being killed on 10 or 12 volts. I can tell you that a 12 volt car battery will give you a nasty shock under the right conditions. I was working on an old car (a 1949 Plymouth modified for 12 volts) I was leaning over the large rusty fender, I didn't have a shirt on so my sweaty torso was in good contact with the grounded fender. I had a large wrench in my sweaty hand (hot, humid Florida weather again). When the wrench touched the positive battery terminal, I saw stars and had a metallic taste in my mouth for an hour. I never expected a shock like this from a car battery.

Now we have the information to figure out the how much voltage question. From the example above it can be shown that under the right conditions that you could be killed by a very low voltage. With good contact points a resistance of 10000 ohms through the body is very possible. Given that resistance 12 milliamps would flow upon contact with 120 volt household current. This current could definitely be lethal. People get killed with 120 volts all the time. The voltage potential in vacuum tube equipment is usually higher than 120 volts. It is a safe bet that ANY of the voltages occurring in vacuum tube equipment are potentially lethal, and MUST BE TREATED WITH THE UTMOST RESPECT.

The quick (and very generalized) summary of the above: 10 volts to 48 volts could be lethal under some circumstances, 48 to 100 volts can be lethal if applied such that the current flows through your torso, 100 to 500 volts is definitely lethal if applied such that the current flows through your torso, and possibly lethal in other situations, 500 to 1000 volts will likely be lethal in most situations, anything above 1000 volts will likely be lethal in any situation, even incidental contact. There are many exceptions to these generalizations of course. An associate that I used to work with was working on a large gas laser (I used to maintain the same laser) came in contact with the high voltage power supply (25,000 volts at 200 milliamps) the current entered his right hand came out through his foot blowing a 1/2 inch hole in his tennis shoe, and blowing a hole in the linoleum flooring to get to the concrete. The current knocked him all the way across the room. The plant safety team administered CPR and he made a complete recovery with no lasting ill effects. DO NOT COUNT ON THE SAME LUCK!

If 48 volts is the threshold of danger, is lower voltage always safe? Absolutely not! At one time I worked at a mainframe computer company (that built the computers for NASA). One of my work associates lost a finger working on a circuit powered by 5 volts! How? His wedding ring touched the 5 volt buss which was fed by a 200 amp power supply. It was instantly welded into the circuit and then it proceeded to glow red hot. His instinctive reaction to jerk his hand out stripped most of the skin off of his finger. The finger never recovered. The moral of this story, remove ALL METAL JEWELRY before working on ANY electrical circuit. Most jewelry makes an excellent electrical contact to the skin. Gold jewelry also has an extreme affinity for solder, which can not be easily removed.

Another issue that you don't think about until it happens is unexpected circuit behavior. When working on a new or unknown circuit, the unexpected can happen, which can startle the technician, causing accidental contact with high voltage. What kind of unexpected behavior? Sparks, smoke, loud sounds from the speakers, and even exploding parts. Electrolytic capacitors are particularly prone to exploding if connected backwards or exposed to too much voltage. I had an incident back in the vocational high school i attended, where several students were gathered around the bench where I was working on an old TV chassis, which was powered up. The TV chassis was standing on one end (a common practice then) to allow access to the components under the chassis. A capacitor suddenly exploded without warning which startled the onlookers. One of them knocked the live TV chassis over which landed on my leg causing a strong shock. The resulting muscle spasm caused my leg to extend, pushing me and the chair over backwards, separating me from the TV chassis. Fortunately no one was injured, but I still remember the incident with vivid detail even though it was over 35 years ago.

The reaction to even a minor shock can cause injury. In the same class I was dismantling an old TV that had not been powered up in months. There was a significant stored charge held in the picture tube itself which found its way into my hand. The involuntary reaction of jerking my arm out of the TV caused my arm to be severely cut by the metal frame around the convergence circuit board. I have the scar to remind me of that incident.

Looking back on these incidents it is now obvious that the working environment in that classroom (built in 1962) was very unsafe. The work benches were made of metal which was grounded. Most of the test equipment (old RCA scopes and VTVM's) was made of metal. The floor was bare painted concrete. The Philco Racks that we used to perform experiments on had every component and connection exposed. They operated at voltages around 400 volts (vacuum tube equipment). There were ample opportunities for accidental contact with electricity.

The consumer electronics equipment from the vacuum tube era was often designed without regard to safety. It was common practice to connect one wire from the power line cord DIRECTLY to the metal chassis. This became common in the designs that did not use a power transformer (mostly radios but some TV's). If one of the knobs were missing the exposed metal shaft was connected DIRECTLY to the power lines. This can lead to a potentially lethal situation very easily. Fortunately this is no longer common practice but you must realize that the situation exists if you repair vintage electronic equipment. Working on a 1950's radio on the grounded metal bench of the 1960's leaves two exposed metal objects which could be connected directly across the power lines. It is important to visualize these unsafe conditions and eliminate them from your working environment. The use of an isolation transformer is recommended if you repair vintage electronics.

RESPECT FOR ELECTRICITY - Creating a Safe Workplace

Your work bench should be constructed out of a non-conductive material. It should be sturdy enough so that it doesn't move when accidentally bumped. If you make your own work bench, use wood or heavy plastic, do not use metal for anything other than fasteners. My own bench has a 2X4 frame with a 3/4" particle board work surface it is very heavy and was assembled in its final resting place. I have used it for 27 years. If you use an old (or new) piece of furniture for a work space it should also fit these criteria. Any metal should not be exposed, and in a location where it can not be touched.

Mount a power strip (surge suppressor) with a switch in a location where it can be accessed by you and any other person in the room. All power to circuits under test should come from this source. Check it to make sure that the HOT circuit is actually being broken, not the neutral. I have my entire workbench and all test equipment connected in this manner. Make sure all persons that reside in your house (or visit often) know where this switch is and how to kill all power in case of accident. One of those big red emergency stop buttons would also be a good idea. Get into the habit of turning your workbench OFF when you are not using it. This is especially important if you have children at home. I also use a second power strip to power up the unit under test. This offers an additional method of killing power to the unit under test, and the option to power up the bench (and test equipment) without energizing your device under test. My computer, TV, and VCR are on the end of the same bench but on a separate switch so that they can be operated independently.

Your work area should be protected by a Ground Fault Current Interrupter (GFCI or GFI) type circuit breaker. This can be installed in the breaker panel that powers your work area, or a GFCI type outlet can be installed in the electrical box where your workbench is plugged in. Understand that a GFCI may save your life if you come in contact with the power line and an external path to ground. It is USELESS in all other electrical shock situations. Nevertheless this is a fairly common electrical hazard (especially if you work on vintage electronics), so it is worthwhile protection.

The work space area should be dry and well lit. A basement is a poor choice. Many basements have a high humidity and bare concrete floors. This can create an unsafe condition. If your work area has a high humidity consider a de-humidifier or a room air conditioner. Install good lighting. Proper lighting can help to avoid accidents. I have two fluorescent shop lights installed overhead, and connected to the bench power. That way it is obvious that the bench is ON.

The floor area under your workspace should be covered in a non - conductive material. If you are furnishing your own work space consider a thick plastic barrier under the carpet. There are plastic materials made for use in electrical work areas with insulative and moisture barrier properties. If you must use a basement for a work area consider this a requirement. It is also suggested where the floor material is concrete or metal. Dry wood floors are generally OK even without carpet.

Make sure that your work space can not be accessed by untrained people or pets when you are not present. I have a dedicated room in my house for my electronics workspace. I always keep the door closed when I am not in there working. My daughter was taught at an early age that that room was off limits when I was not present. The two cats were trained not to enter. I always turned the bench off when leaving the room, and made sure the cats had not snuck in before turning it back on. I no longer have children or pets at home but I still turn the work bench OFF when not actually using it.

The chair for your work area should also be free of metal. Any metal used in its construction should not be exposed.

The bench top itself should be uncluttered and free of unnecessary objects. I would discourage the use of an anti - static mat since the components used in vacuum tube circuits are generally not static sensitive (MOSFETs ARE static sensitive). A grounded anti - static mat provides an intentional high resistance path to ground. It was intended for use around low voltage computer electronics and may be unsafe when working with high voltages. If you also work on digital electronics as I do, remove your anti - static mat when working with any high voltages. I do not use an anti - static mat since static electricity is rare in south Florida due to the humidity.

Check your work area for any conductive object that you can touch at the same time as your circuit under test. If any of these objects are connected to a power source or grounded. Remove them from the work area before working on a live circuit. THIS INCLUDES YOUR SOLDERING IRON, and any powered or rechargeable tools that are plugged in.

The idea is to create a work environment where if you were (please don't actually do this) to hold a live wire in one hand, no part of your body would be able to touch any conductive object capable of completing a path for current to flow. Since you must have TWO points of contact to get fried, you must make sure that you don't start out (or provide) ANY. This concept GREATLY reduces your risk. If all of your circuit probing takes place with one hand and you accidentally touch ONE hot circuit with that hand and THERE IS NO OTHER PATH FOR CURRENT TO FLOW, you won't get shocked. If you simultaneously touch something else with your other hand you will get shocked, and possibly killed because the path for current goes through your chest. If you are working with one hand and touch TWO points in a live circuit with that ONE hand you will get shocked, but since the path of current REMAINS in that hand you will probably not be killed.

The concepts presented in the preceding paragraph are generalizations and become less valid as the voltage goes up. Following those procedures is always a good idea, and may save your life when working with the voltages typically found in tube equipment ( 200 - 450 Volts). All bets are off when working on high powered amplifiers using transmitting tubes that operate at voltages above 1000 volts. Some of these amplifiers operate on voltages as high as 2500 volts (I have one under construction). At these voltage levels even a single point contact can be lethal since any "insulation" will break down due to moisture absorption. An amplifier of this type should not be attempted without experience and training with high voltage.


The first and foremost practice to adopt is to always keep a circuit in the UNPOWERED state except when actually making measurements on it. NEVER leave a circuit powered up unattended. Replace all protective covers as soon as repairs are made. If the device under test is new, make sure that all exposed electronics is inaccessible to the user before placing the unit into service.

Make sure that a circuit is COMPLETELY DEAD before touching it. Verify that there is NO connection to the power source. This does not mean that the power switch is off, this means that the unit is UNPLUGGED. I will pull the power cord (all of my amplifiers use detachable power cords) from the back of the unit AND turn OFF the power strip that it is plugged into. This means TWO actions are needed to power the unit back up. Wait 5 minutes before touching the unit and then the first move you must make is to check for residual electrical charges. ALL capacitors are capable of storing an electrical charge, that is what they are designed to do. The better quality capacitors can store charge for a longer time. If you designed the equipment all of the power supply capacitors have bleeder resistors, right? If you haven't verified their presence and condition assume that they are missing. You must make sure that all capacitors are discharged BEFORE touching the unit. It is a good idea to have a digital voltmeter always connected to the main high voltage power source of your circuit under test. Verify that the meter reads the expected voltage when the unit is on, and decays to near zero when the unit is turned off. Verify that the meter reads near zero before touching the unit, but do not rely on it as your only indication of a dead circuit. See the page on safe meter use.

Get used to thinking about each and every action that you take when there is a live circuit on your bench. Take the time to mentally visualize the consequences of everything you do BEFORE you do it. I have been working on electronics since I was 10 years old (solid state was not common yet). I still stop to think about what I am doing whenever the power is ON. The 845SE amplifier that I built (1200 volts) still scares the SH** out of me. This is GOOD. You MUST be conscious of the fact that electricity has the power to KILL you if you disrespect it. DON'T EVER FORGET IT. If you ever get mad it or frustrated because your circuit won't work, WALK AWAY. Come back again another day. Mistakes can kill you.

As mentioned previously remove any and all metal jewelry. Metal jewelry creates a good contact point for current to enter or exit the body. Remove it. It also can short out your circuit, and as mentioned before actually melt right on your body. The filament transformer in most tube amplifiers can melt small jewelry chains.

Keep your work area as clean as practical. It is especially important to remove any extra wires, cables, tools and other conductive items that are not being used.

The old axiom says to always work on a live circuit with one hand in your pocket. This is a good idea. I can't seem to do that but I do always keep my left hand on my waist or behind my back when probing in a live circuit. The concept is to get in the habit of avoiding situations where electrical current could flow through your body.

Wear some type of footwear. This will help provide insulation where it is needed. Shoes with rubber or plastic soles are preferable over leather since they are less likely to absorb moisture. If you are a habitually barefoot person (like me) keep your feet off the ground when sitting and keep a pair of slip on sandals under your chair. Bare feet on carpet is hazardous, bare feet on concrete is DEADLY. It may seem like concrete would be a poor conductor but it conducts well enough to kill you. I read somewhere that standing barefoot on concrete grounds you about as well as standing in a bathtub full of water. In this situation touching a single powered electrical point would have disastrous results. Both of these conditions kill people every year. A GFCI may save your life in this situation if contact was made with the power line voltage, this is why they are required by code in bathrooms. The concept again is to get in the habit of avoiding situations where electrical current could flow through your body. Current from hand to foot is especially deadly since it passes through the entire torso.

If you are in a humid climate, or a humid environment (like a basement) consider a dehumidifier or room air conditioner. Lowering the room humidity and/or the temperature lessens moisture absorption by porous materials. It also reduces sweating by the body. This will greatly increase the skin resistance which will reduce the severity of a shock if it happens. I live in South Florida, and I tend to sweat a lot. for this reason I installed a window air conditioner in my work room, even though my house has central air. Both are usually on when I am working. You can get a single room air conditioner now for under $100. Mine was $79 at SAM's club.

Digital meters are now available for under $10 each from surplus dealers and wholesalers like Harbor Freight ($20 at Radio Shack). It is a good idea to purchase several (I have 7) of these (find one with detachable leads, that covers your anticipated voltage ranges). That way you can connect several meters into the circuit under test at once, while the circuit is OFF. Then you can power the circuit up and read the meters WITHOUT TOUCHING ANYTHING except the power switch. If you are using a power strip you can be a safe distance away from the circuit, turn it ON, wait for the readings to stabilize, read the meters, Turn the power OFF, wait for all meters to read near zero, verify that the circuit is DEAD, and then you may approach it. If you follow this procedure it is very hard to get shocked. On a new and previously untested circuit, especially a high powered one, I add a second measure of safety by standing behind a thick Pexiglass sheet in case anything explodes (it DOES happen, and it has happened to me). I will also have one meter on each power supply point, if the voltage readings do not come close to the anticipated values after the tubes warm up POWER OFF immediately. See the page on safe meter use.

On that subject keep a good fire extinguisher in your work area at all times. The best type for electrical fires is a CO2 (or other inert gas) extinguisher, since they will put out a small fire without leaving a mess. They are getting hard to find however. You can also use a dry chemical extinguisher that is rated for electrical fires, but these will make a mess. DO NOT USE any type of liquid extinguisher, since the liquid is likely to be conductive.

It is a good idea not to work on dangerous electronics when you are the only person at home. If you should become incapacitated due to an electric shock, you will not be able to call for help. All people in your house should know where your master switch is, and how to kill the power. They must know not to touch you or ANYTHING ELSE until that master switch is OFF. Provide some visual means to reassure them that it is OFF. That is why I have the overhead lights connected to this switch. If you are going to work on electronics often, someone in the house should know CPR.

Label the breaker that feeds your work area in your homes electrical panel. Make sure that all persons residing in your home know where it is, and how to kill the power to your work area. This provides a backup means to kill power in case of an accident.

Make sure the device under test is mechanically stable. It should be incapable of falling over (or off of the table) if bumped. Route any wires (power cable, speaker leads, interconnect cables) off to the rear of your bench. Keep any test leads, or scope probes routed away from the edge of the table. The idea is to avoid dropping a live circuit in your lap or on the floor if your chair (or anything else) gets tangled in one of the cables.

Make sure that all tools and other unnecessary conductive objects are removed from the immediate area. Keep all test leads that are connected into a live circuit away from you and any other conductive objects. Ideally there should be only the device under test and one or more meters on the bench top. I have all of my test equipment on a shelf above the work bench to allow for a clean test area.

Remember that components can (and DO) fail in an unexpected and often violent manner. An overloaded resistor can build up internal heat and pressure causing it to explode or burst into flames. They will usually (but not always) smoke or emit an odor of burning plastic before failing. Overloaded capacitors, especially electrolytics, will often explode violently without warning. The newer types have a safety vent (usually on the top of the can). A capacitor of this type can violently spew its guts out through this vent when severely overloaded. The contents will be extremely hot and contains a caustic substance. The material can be sprayed a few feet. The size and magnitude of the disaster is dependent on the available power. The worst case is an electronic device that operates DIRECTLY from the power lines like computer power supplies, UPS's, and those fluorescent light bulbs that screw into a conventional bulb socket. I saw one of those, which had been working normally, suddenly explode quite spectacularly. I autopsied the device and found that there is a diode bridge connected to a 150 volt electrolytic cap connected directly to the power line with NO fuse. Our power here runs about 120 to 122 volts which puts about 155 volts on the 150 volt cap - BANG.

When working on a new project, or a piece of vintage electronics, it is a good idea to remain a few feet away from the unit when first powering it up. Use a safety barrier if possible. After it has been on for a few minutes turn it OFF (even if it is working normally), wait for things to discharge, then touch the case of each capacitor carefully. If any are very hot find the cause before proceeding. The old metal can caps found in vintage electronics will explode violently with considerable force, check them carefully. They will often develop a slow leakage that increases with time and temperature. Consider replacing all of the capacitors when restoring an piece of old equipment. It will result in a safer and more reliable piece.

Always wear eye protection when working on a live circuit. Parts can explode without warning.

Never intentionally touch any part of a live circuit to see if it is hot (thermally). The insulation may fail. Turn the unit OFF wait for things to discharge, then touch it. If it was too hot, it will still be hot.

Never enter your work area when you are not in a completely alert and sober state. Even very mild intoxication GREATLY increases the risk of accident since you MUST ALWAYS think about the consequences of each action when working with a live circuit. This seems totally obvious to me, but some people need to be reminded. The same applies to being tired, distracted, or under the influence of some over the counter drugs. I have found that cough medicine makes me stupid, so I stay out of the room (except to use the computer, the work bench remains OFF) when I have a cold. Remember YOU are RESPONSIBLE for your own electrical safety, AND the safety of those who enter your work area.

Always remember that electricity can be the source of many rewarding hobbies and careers but it MUST BE RESPECTED. It has the power to kill you. This information is not meant to frighten the newcomer away from the hobby, it is meant to be educational, and to possibly save their life. Statistically you are much more likely to be killed driving a car.

What Happens if Someone Gets Shocked

Despite the best safety procedures, accidents still do occur. Most of the time, these accidents are the result of not following safe practices. But however they may occur, they can still happen, and anyone working around electronics should be aware of what needs to be done for the victim of electrical shock.

If you see a shock victim, the very first thing to do is shut off the power. If a shock victim's breathing and heartbeat are paralyzed by electric current, their survival time is very limited. If the shock current is of sufficient magnitude, their flesh and internal organs may fried by the electrical current. The electrical current must be disabled quickly. This is why I insist on a master bench switch. If someone touches the shock victim, there may be enough voltage present to shock them too. Never touch a victim of electrical shock unless you are sure that the power is OFF.

Once the victim has been removed from the source of electric power, the immediate concern should be breathing and pulse. If they are not occurring, begin CPR immediately while calling for help. Continue CPR until help arrives. If the victim is conscious, it is best to have them lie still until the emergency personnel arrive. The victim may appear OK but suffer after effects at a later time. There is the possibility of the victim going into a state of shock (physical shock as opposed to electrical shock). There is also an elevated risk of heart attack or cardiac arrest for several hours after the incident. Since the body's central nervous system actually operates on tiny electrical impulses, a large electrical impulse (shock) may disrupt the system balance for a while. They should be observed (professionally) for several hours after a serious incident.

RESPECT FOR ELECTRICITY - Designing Safe Equipment

The most important thing to understand is that when your project is finished, the user can not come in contact with ANY conductor that isn't grounded. What does this mean? It means that the user can't touch any of the circuitry that contains ANY voltage, and ANY CONDUCTIVE MATERIAL that the user can touch (knobs, connectors, transformers, cabinet if metal) is GROUNDED.

When you have finished that cool new project check for continuity between the ground prong on the power cord, and any conductive material that the user can touch. If continuity is not present, find out why and fix it before putting the device in service.

The current trend in electric appliance design today is "double insulation". The device is usually encased in plastic, and has a two wire power cord with no ground. This may work OK on low voltage devices that do not have external connections. This practice does not work on vacuum tube equipment that is connected to external devices that are handled by the user, such as the turntable or CD player. A failure in the insulation of a transformer could put a high voltage on the external devices WITHOUT AFFECTING THE AMPLIFIERS OPERATION. The equipment could work normally with this fault for years until an unlucky user touched the turntable and a grounded object at the same time. That user could be killed.

Sounds far fetched? I found an working amplifier with 300 volts on the speaker leads of one channel. This amplifier was built by a friend of mine who brought it over for testing and it would blow the fuse every time I connected it up to my equipment. The output transformer had developed a short between the primary and the secondary. This did not affect the operation of the amplifier with speakers attached since speaker terminals were not grounded. He had used output transformers that were over 50 years old, and the insulation had broken down over time. This amplifier would have never worked in the first place if everything was properly grounded.

Two wire power cords were common on vintage electronic equipment. I would highly recommend replacing this with a three wire power cord where possible. This is especially important on vintage guitar amplifiers. I have found several vintage guitar amplifiers with enough leakage current to give the user a nasty shock if he touched the guitar and a grounded object like a microphone.

All new equipment should have a three prong power cord with good continuity from the ground prong to ANY CONDUCTIVE MATERIAL that the user can touch. I use the ground terminal on the power receptacle as the star grounding point for the amplifier. See the section on AMPLIFIER GROUNDING for further details on this subject