Back to Tech Index
Back to Home
AC Power: Color Codes, And How To Avoid Killing Yourself
By Mike Morris WA6ILQ
|A note to folks arriving here from the Wikipedia page on USA Electrical wiring:|
This web site, www.repeater-builder.com, is oriented towards amateur radio, commercial 2-way radio, Civil Air Patrol, and other two-way radio topics.
This article on AC Power has attraced a large number of readers from the Wikipedia site, and those folks that have very little, if any, exposure to the Federal Government licensed amateur radio service. This article is oriented towards educating the folks that are supervising the power wiring for a two-way radio base station or repeater station, which are frequently at remote or mountaintop radio sites.
I live in the Los Angeles area, and a while back I made several business trips to do software upgrades - thirty sites in six states in less than twenty days. Most were in Washington, Oregon, Idaho, Nevada or Arizona so it wasn't as bad as it could have been... I scheduled one trip across a 4-day weekend and used the days off to help out a group of teenage hams that were setting up a new repeater site. They had built a 8 foot by 8 foot (2.5m by 2.5m) concrete block building and installed an adjacent 30 foot (9m) tower - a completely new site on private property that will initially have a UHF autopatch repeater using an NHRC-10 controller and later on the tower will grow to 60 feet and sport an Icom IC-706 HF/6m/2m remote base. The slab was poured and the building was built from the ground up, using volunteer labor (the kids called it "slave labor"). Construction supervision was by one of the kids uncles, he was a retired mason (bricklayer) who also knew how to pour concrete and he insisted that all permits be done properly. They had even trenched to the nearest power panel and laid a 2 inch conduit for 240vAC mains power, and an additional 3/4 inch conduit for telephone (autopatch) and ethernet (IRLP) wiring, plus a extra 3/4 inch conduit for future use or in case they forgot something. I was drafted by one of the hams I know for my knowledge of repeaters, but I ended up probably saving someone's life when I took over supervising the AC power wiring installation, and in the process teaching a few young hams and their folks how not to commit suicide by electricity. I'm not going to embarrass anyone by stating who he, she or they are or even what state they are in.
In the electronics world, color codes rule - everyone knows the resistor color code (the clean version is "Big boys race our young girls but Violet generally wins gold or silver"), and back in the tube era there were standard color codes for capacitors and for transformer windings.
Well, transistors have replaced tubes as the predominant technology, and the "red is always +12vDC and black is always ground" mentality has taken over.
Well, it's not that way in the electrical power world, where the black wire can (and will) kill you. This fact was brought home to me in a painful way thirty years ago when I discovered two adjacent metal workbenches in a school electronics lab were 120vAC hot to each other, and again later on when I was working at NASA / JPL.
USA AC POWER WIRING FACTS:
1) Mains power in the USA is fed from the utility to the building power entrance on three wires (mentally picture the secondary of a center-tapped transformer, with the center tap being ground / neutral). This allows full voltage from end wire to end wire and half voltage from either end to center tap / neutral / ground. Other countries do it differently (and you will find the local color code is different as well - for example common european colors are brown for hot, blue for neutral, and green/yellow for "earth" (safety ground)).
Note that there is no such thing any more as "110 volts" or "220 volts" although those numbers are still commonly used. Residential single-phase voltage is supposed to be 120 / 240 VAC, however you will find 120 / 208 VAC in light commercial, some apartment complexes, and some condominium buildings (from two phases of a three-phase distribution system). This means that standard 240 volt appliances (like clothes dryers) are going to be underperforming when run on 208 volts. Some equipment has a wiring option, others can be ordered specifically for 208 volts.
2) Once you get past the main entry panel (usually where the power meter is located) the ground wire is not the same as the neutral / zero volts wire. The solid green wire (or green with a yellow stripe or stripes), green screws, and round hole in power outlets is the safety ground (also known as the "Bonding Conductor" or the "Equipment Grounding Conductor", or internationally the "protective earth" conductor) and connects the housing of all equipment to ground at the electrical service entrance point completely independently of any other wiring. The water lines, natural gas lines, metallic air conditioning lines, air conditioning ducts, etc. is also supposed to be connected to the safety ground. Internationally the standard for the "protective earth" conductor uses green insulation with a yellow stripe or stripes where it used to be solid green. In the USA the solid green insulation is still very common, or it can be a bare uninsulated copper wire. Even when it is a bare copper wire (usually #12 or #10) you will hear electricians refer to it as the "green wire" - don't let that bother you. Whatever form it takes it can never ever be interrupted by any type of switching, be it a circuit breaker, a switch, a fuse or anything else. This "green wire" is intended to permanently "bond" all of the normally unenergized parts of the equipment / appliance / fixture to the ground bus back at the power service entrance. Under normal circumstances there is no current in the green wire - in other words this wire is strictly for handling fault current, and never, ever carries load current.
A comment from a friend that is a retired building inspector:Personally, I don't trust the conduit / flex to be the bonding ground - in addition to making sure that the conduit / flex is grounded, I always pull a green wire along with whatever is going inside the conduit and treat the conduit / flex only as a construct for the physical protection of the enclosed wiring. In fact, I've been called anal when folks see me use nothing but EMT (solid metal tubular conduit) or flex, metal outlet boxes, and lots of green pigtails to ground every outlet box, metal pull box and metal switch box (in addition to the outlet or switch itself) to the green wire. My response usually is: "I am doing this job as if MY mother or daughter will be living here". And I have NEVER had a problem with the building inspector (except for the fact that he thinks I overdo things).
In older wiring systems (where "old" is defined as the mid-nineteen-seventies and before), the conduit or flex may be ASSumed to be the bonding system. Some of the time, this is an un-founded ASSumption.
In case you are curious, the international standard green-yellow marking of the ground conductors was introduced to reduce the risk of confusion by colorblind installers. Between 7% and 10% of men cannot clearly distinguish between red and green, which is a particular concern in older schemes were red marks a live conductor and green marks safety ground. If you are purchasing wire for a new wiring project, and you have a choice between green / yellow and solid green, please pick the green / yellow.
3) The "zero volts" or the return wire is white (which in some circumstances and for various reasons can sometimes be gray). The white / grey wire, the silver screws (commonly called the white screws), the wide slot on power outlets and the threaded sleeve of the light bulb sockets are all the neutral connection. This is insulated from ground everywhere EXCEPT at the service entrance where it is ALSO tied to the ground buss. Yes, green and white / grey are tied together at the service entrance, and only at the service entrance. The hot wire is the power source, the white / grey is the power return. As with the green wire safety ground above, the white / grey neutral wire must never be interrupted by any disconnecting mechanism. The switch should ALWAYS be in the hot lead, so that when it is off the fixture or outlet is electrically DEAD... but I've seen too many residential bedroom ceiling light circuits or bathroom light circuits wired with the switch in the neutral lead... after all, it's a series circuit, and it doesn't matter, right? Note that while the white / grey wire is at zero volts, it is never considered as or ever referred to as "ground"! It is always referred to as "neutral" or "the neutral lead".
One personal pet peeve: Over the years I've seen several situations where a residential bathroom light fixture body is hot to the faucet because:
(a) the fixture body and mounting box wasn't grounded,
(b) the switch was in the neutral lead and
(c) the hot wire insulation inside the box had frayed and the copper wire was touching the fixture.
Repeating the above: the switch should ALWAYS be in the hot lead, so that when it is off the fixture is electrically DEAD.
Do not trust the plumbing to be at ground. Plastic fittings and PVC or ABS pipe are too prevalent for this ASSumption to be valid.
ALWAYS make sure that the electrical fixture or outlet is green-wire-grounded, with a pigtail if need be, and always make sure the shell of the light bulb socket(s) i.e. the threaded base(s) are at neutral potential. Yes, this last point is impossible on table lamps and other household equipment that uses simple two-wire AC cords unless you deliberately use a polarized plug, (which you should). In the 35+ years since the mandatory electrical overview class by a NASA / JPL electrician and a safety supervisor (discussed later) I've not purchased or installed a non-polarized 2-wire plug. And I've cut several non-polarized plugs off the ends of various cords (like on kitchen applicances) and installed polarized ones.
4) Any color except green or white / grey is a HOT wire I will repeat myself in fewer words: ALL OTHER COLORS ARE HOT. Engrave that in your brain. Deeply. It may save your life or that of a family member, friend or co-worker.
5) Brass screws, the narrow slot on 2-prong or 3-prong power outlets, and the center button / center contact of the light bulb sockets are always the hot wire (but I've seen too many incorrectly rewired table lamps that have the shell hot and the button as neutral, even with a polarized plug). This is why some quality "zip" cord has a brass colored conductor and a silver colored conductor – the polarized plug goes on one end with the wide blade on the silver color, and the lamp socket goes on the other with the shell on the silver. Other types / manufacturers of zip cord have a "rib" molded into the insulation over one wire so you can tell the two conductors apart.
When wiring a structure electricians are supposed to use one color, usually black to signify an unswitched circuit, and another hot color to signify a switched circuit. As an example of this usage, the hot wire from the power disconnect breaker to a wall-mounted light switch should be black, and the switched hot wire (from the switch to the outlet or light fixture) should be another color: blue, brown, yellow, orange, red, etc. Personally, when I do AC wiring, it's not "should be" but "is". In situations where you have multiple circuits fed to somewhere else through a single conduit or flex I've seen blue, brown, yellow, orange and red frequently used as unswitched feeders downstream from a breaker panel (and don't forget to follow the NEC and bump up the diameter of the neutral appropriately if you run multiple hot circuits with shared neutral(s) in one run of flex or of conduit).
This article was proofread by two friends before posting... one is a retired building inspector (25+ years), another is a retired electrician with 20+ years of experience in both residential and commercial environments. My electrician friend told me a story that he had read in an electrical trade magazine many years ago: a child (in bare feet) was electrocuted when he contacted a metal portion of an older freestanding illuminated business sign. The sign was at the child's shoulder level and mounted on a wooden wall and surrounded by a flower bed. As part of the coroners investigation the sign was dismantled to see what had happened. It turned out that:
a) sprinkler water and rain water had migrated inside the sign (which was over 20 years old),
b) the sign was constructed in such a way that water pooled in the bottom,
c) the green wire was fastened to the bottom sheet metal of the sign with a single common sheet metal screw,
d) the submerged end of the green wire had corroded to the point it had broken,
e) the hot wire had been routed inside the top of the sign and the insulation became brittle due to a combination of age and the rising heat from the light bulbs,
f) the brittle insulation had flaked off the wire in chunks and those pieces were found in the bottom of the sign,
g) the now exposed hot lead had come into contact with the metal housing of the sign.
None of the above had been noticed since the sign still worked, but the combination resulted in 120vAC on the shell of the sign. A child in bare feet standing in a damp flowerbed coming in contact with with the housing resulted in a fatality.
The article suggested extending the green wire to a position that is on the side or top of the housing to prevent pooled water from causing a disconnected green wire. It also suggested bonding together all sections of a multisection metal housing sign with green jumpers.
In short, the green wire has one very important purpose - to trip the breaker or blow the fuse by being a fault current pathway BEFORE a persons body touching a charged surface becomes the path of least resistance to ground.
6) The most common mains power usage of the wire color red is to signify the other hot leg (i.e. non-black) of a 208 / 240 volt circuit. On the other side of the circuit breakers you may find it used to signify switched circuits as mentioned above. I have, and frequently. Personally, I always use black on the source side of a switch, and a non-black color (anything but red or orange is my first choice) on the load side.
7) The USA National Electrical Code (NEC) allows for a situation where an electrician runs out of a wire color while on the job and you can find the answer at any electrical supply store and at most construction supply stores - electrical tape in various colors (sometimes called "phase tape"). A band of colored tape wrapped around a wire signifies a "wrong color" - for example, white tape wrapped around black (actually any color) insulated wire marks that wire as being a "white" (i.e. neutral) wire. Tape wrapped around the end of a wire - even black tape - indicates that the tape color takes precedence over the wire insulation color. If you ever need to use color / phase tape, use it liberally: a couple of overlapping layers for at least 2 inches, at both ends of the wire run, and in every splice box and / or pull box in-between (a lot of electricians skip the intermediate splice and pull boxes, and the inspectors let them get away with it). Personally, I put a length of clear heat shrink over the phase (color) tape. My local electrical contractors supply house and even the local Home Depot store stocks colored tapes in red, orange, yellow, green, blue, violet, brown and white. And note that larger wire sizes are available only with black insulation, and that phase tape is mandatory in that situation. You should use a phase tape marker on all the black conductors - even to the point of using a black piece of tape on a black wire.
I don't have to go far to see an example of phase tape: I lived in a house that was built in 1939 and completely rewired somewhere abound 1958 to 1960... There was a meter and main breaker panel on the side of the garage that fed a 4-breaker sub-panel just for the garage and workshop. The main panel also fed a 2" underground conduit (to the house) containing three black feeders for the main power. One of the black wires had no tape (not a good idea). The second had several turns of red tape around each end. The third had several turns of white that has faded to grey. There was no green - the buried conduit was the gound. Every bedroom ceiling light switch had a black wire coming into it, and a black wire with yellow tape going out of it. The kitchen and dining room were remodeled about 1962 to 1964 and the switched outlets and ceiling lights in those two rooms have no identifying tape - interesting, no? The bathrooms were redone a few years ago (under my supervision) and they have proper colored wire to start with.
While we are mentioning tape, don't use the bargain-bin no-name cheap black stuff on any AC power wiring that you are responsible for (or anywhere else, for that matter). It loses its "stickyness" with age and unravels, or it melts and turns into a gooey mess. Use the real Scotch "Super 33" tape. Scotch "Super 88" is even better as it's thicker, has a better adhesive, and it has better UV resistance (which only counts if it's outdoors). The price difference between the cheap stuff and the Scotch 33 is not worth having a splice "unwind" itself and possibly cause a short or worse, an electrical fire. Have a local union electrician show you how to tape up a wirenut. There is a non-obvious trick involving stretching the tape in the middle and not stretching the tape at the ends that you need to know in order to have your work pass inspection. While you are talking to him, have him show you how to mark a wire with phase tape. He will probably use more tape that you think is necessary, but he knows what the local building inspector(s) want to see, and a few inches more tape on each phase marker, on each wirenut and / or splice is cheap compared to being forced to go back and do it over.
8) The NEC is revised frequently, especially after major incidents. One such incident was the First Interstate Bank fire on May 4, 1988 in downtown Los Angeles that started on the 12th floor (of a 62 floor building). The fire was caused by an overloaded AC power outlet, and a lack of inspection after wiring changes were made by building electricians to support a new minicomputer installation two years earlier. At the time it was the worst high-rise fire in history and was the incident that "wrote the book" on fighting high-rise fires. The 1974 movie "Towering Inferno" was strangely prophetic - in the movie the fire is caused by electrical overloads...
Another incident that resulted in a change in the NEC was the 1980 Las Vegas MGM Grand Hotel fire that killed 85 people. The after-incident investigation found that the cause was a piece of flex cable that powered the lights in a display case in a restaurant. The cable was damaged somehow, and a hot-to-conduit short developed. But one of the clamps in the conduit path had worked loose, resulting in a non-zero resistance between the display case and "real ground" that was low enough to draw substantial current but not enough to trip the breaker. So the current kept flowing, and the heat kept building until it started a fire. I'm pretty sure that THIS was the event that prompted the NEC to ban the reliance on conduit grounds and require that green wires be pulled on every circuit.
The organization behind the NEC does excellent work, but in my personal opinion, the NEC's reliance on the green wire to trip a circuit breakers is too focused on just one method. If you want to protect against hot-wire shorts to ground, then use the device that's specifically designed to detect them and open the circuit -- the ground fault interupter! If one had been on that MGM Grand circuit, it would have immediately tripped even without a green wire in the conduit. But having a green wire AND a GFI is better.
Another personal opinion: the minimum electrical wire sizes specified in the NEC for amperage are totally inadequate and inappropriate as a wire size guideline for proper operation. The NEC considers the purpose of the circuit breaking mechanism (a breaker or a fuse) to be protecting the wiring (to prevent fires), not to protect the load. In other words, they don't care about proper performance of the load (your equipment) with minimal voltage drop, just the safety and integrity of the power panel, wiring and receptacle feeding it. Unfortunately too many architects and remodeling managers allow the construction electricians to bid a job only based on local code requirements. And the electricians that do the work don't care because (a) the smaller diameter wire is faster and easier to run (less work), (b) the smaller diameter wire costs them less (i.e. lower material costs raise their profits), (c) they get paid the same hourly rate for running thin wire as thick wire, and (d) they won't have to live there, nor will they have to do the future maintenance (they are the construction electricians).
The older Radio Amateur's Handbook published by the ARRL (older being defined here as late 60s / early 70s) had a VERY nice one-page wire table in them that included a column that listed the resistance of the wire per one hundred feet. It's worth hunting down a copy and xeroxing that page for your personal file cabinet. And remember, at two amps of load current a wiring resistance of 5 ohms per 100 feet is equal to a thirty volt drop at the load 150 feet away - fifteen volts outbound, and fifteen volts on the return. This means that your 250 watt soldering iron is running on 90vAC instead of 120vAC and at 180 watts just might not get hot enough to solder that PL-259 connector to the RG-8 coax. And that's just a 250w load - how about that RV or 5th wheel trailer that has an electric clothes dryer in it? Low voltage causes even more problems on electric motors - you can actually kill a motor with too much voltage drop. Using conductors that are just one size larger can cut that voltage drop in half. (This is why the kids mentioned in the story at the top of the page used a two-inch conduit from their new building to the nearest panel - they ran a ground, a neutral, and two hot wires. And they were "fat" wires (deliberately oversize in the conductor diameter).
The NEC specifies that residential electrical must be a #14 gage wire or
larger (smaller numbers indicate a larger wire size) for use on a 15 amp
circuit, but only up to about 50 feet of wire length (and outlets are
completely forbidden on #14 gage circuits). The NEC specifies #12
for a residential 20 amp circuit, but only up to about 50 feet. For
a 30 amp circuit the NEC wants you to use #10 - but again, only up to about
The NEC handles commercial buildings differently from residential - for example, in commercial environments the smallest wire size allowable is #12.
Personally, I think that is too small. Very few people realize that the wire that feeds a common 3-wire outlet can be carrying as much as 24 amps in a worst-case scenario. The NEC allows the standard NEMA 15R outlet (the common "duplex" 3-wire outlet) to supply up to 12 amps of current to each of the two outlets (which is why the vacuum cleaner wars a few years ago stopped at 12 amps of "cleaning power"). This means that you can plug in two loads of 12 amps each into one outlet.
If you have any questions about what wire sizes or what types of insulation are legal for your conditions and building environment you need to check with the local building code authorities before you get started. It's way cheaper to ask first rather then to be forced to rip everything out and start over. If the locals have different rules than the NEC, they are usually tougher and they are the ones that you have to follow to get the local inspector to sign off on the work... (as long as you use the proper type of wire - i.e. THHN, or whatever they want to see - you will never have a problem if you go larger in wire size (conductor diameter), or better (in insulations specifications).
The actual NEC is big, expensive and pretty dry reading, but overview books are available at most good bookstores and at web-based booksellers like Amazon. Or there are the specialty technical-only booksellers.
A more accessible source of information is the overview books written for do-it-yourselfers that available at Home Depot and similar places. My favorite is "Ugly's Electrical Reference" by George Hart, it's about 200 pages and you can buy it at Home Depot. And don't forget your local public library - they sometimes have copies of the current or a relatively recent NEC in the reference section, and the some have overview books and older issues of the NEC in the circulation section. And check the second-hand stores - I saw a five or six year old copy of the NEC in a local thrift store (for US$2) just last year. And you occasionally find the $25 and $35 Home Depot overview books for $1 or so at church rummage sales. Used bookstores are interesting places as well.
I am not a professional electricial, nor do I play one on TV or on this web page. But I do know how to read, and I have read sections of the NEC, and several other books. I've also had long discussions with union journeyman electricians and building inspectors. Every bit of AC power wiring that I personally do uses wire sizes that are at least NEC commercial - and in many cases I go even further (I've been known to use #8 wire for a 75-foot run to a 20-amp air conditioner since low voltage does nasty things to AC motors). Personally, in the last 10 years I've never installed any branch circuit smaller than #12 (even for a two-amp bathroom ceiling fan).
Most circuits in the modern home in the United States are wired with non-metallic sheathed (i.e. "NM") 14‑2 or NM 12‑2 cable (often referred to by the name "Romex" - another name that has become a generic). This type of cable is the least expensive for a given size and is appropriate for dry indoor applications. The designation NM XX‑Y indicates, respectively, the type of sheathing (in this case, non-metallic), the size of the main conductors, and the total number of circuit conductors not including the green safety grounding conductor. For example, NM 14‑2 cable contains a 14 gauge black wire, a white wire and a bare wire (for ground). NM 14‑3 cable contains four conductors, a black, a white and a red plus the bare wire ground. When NM XX‑2 cable is used on 220v circuits (like to a window air conditioner), the one black, one white, and a bare grounding conductor conflicts with the NECs rule about the white conductor identifying the neutral conductor (on a 220v circuit there isn't a neutral). The NEC allows the white wire to be used as the second hot conductor, but mandates the white be marked with phase tape on each end.
9) In commercial / industrial (i.e. non-residential) wiring in the USA, watch out for the orange wire. IF it is correctly used it is at 277 volts AC above local neutral and ground. That stings. To avoid confusion for the guy that follows you, especially in a commercial installation, when you do AC wiring in a commercial environment kindly move the color orange to the bottom of your list of available colors. Better yet, leave the commercial / industrial to the licensed professionals.
By the way, electrical color codes are much, mmuch more extensive that
anything I've listed above. See the web page at
for an excellent overview. Wikipedia has a decent writeup on
three-phase power that includes color codes at
The mathematics of three phase power is covered in a different article at http://en.wikipedia.org/wiki/Three-phase.
There is also some information on single phase power at http://en.wikipedia.org/wiki/Single_phase_electric_power.
At the new radio site mentioned above the construction supervisor listened when I pointed out my concern (mentioned above) about high power loads in both sides of a duplex outlet so they used #10 on every outlet (a few folks thought it was serious overkill for the load involved). The kids wired every branch circuit with #10, and every quad outlet box was fed with two branch circuits, one from each phase, so that in the future 220vAC can be easily obtained at any location in the room if needed. The kids pulled a #10 black, a #10 white, and one more #10 (blue, brown or yellow) to each outlet box (plus the green/yellow safety ground).
The overhead lighting circuits were wired from two different breakers (one on each phase) with #12 wire - a green/yellow, a white, and two hot wires (brown, blue or yellow). They mounted two light switches by the door, feeding two four‑foot single tube fluorescent fixtures. The two phases, two breakers, two switches and two fixtures on the lighting insured that there would be light even if one of the lighting switches or breakers failed, or if they lost a phase from the upstream breaker panel. Doing things that way is just good planning and design... when I help wire a house I put three circuits to each room: the first powers the ceiling light, the second for the outlets on two adjacent walls and the third for the outlets on the other two adjacent walls. No matter what breaker you have to switch off to repair something, if you need light you just turn on the ceiling light, or plug your work light into an outlet on the opposite wall - you won't need an extension cord that runs to the next room or to the far side of the house.
Tools that you might not think of: A 100' Greenlee brand nylon fish tape is rather expensive, but worth it as it will last you for many, many years (like my dad used to say, "Buying good tools only hurts once"). Add to your shopping list a squeeze bottle of Scotch #77 wire lube and a black Sharpie marking pen (to label the circuits - just use the Sharpie to write on the inside of the outlet boxes as to where the wire is from, i.e. "FRM CB#1" ("From circuit breaker #1"), or on the inside of the light fixture you could write "FROM CB #4 VIA LEFT SWITCH BY DOOR". The next gentleman that has to work on the wiring will bless your thoughtfulness.
The school electronics lab workbenches I mentioned above? They were sitting on a linoleum floor, which was on top of a second, older, asbestos tile floor, each of which acted as an insulator. One bench had been added to the rear of the room a number of years after the lab was built, and placed in a back corner of the room to be used as an equipment repair bench (electronics students are rough on equipment...). The story was that the back bench AC outlets had never worked because they had never been hooked up, hence it was used as a storage bench and library table. I accidentally discovered its exterior was hot one day when I leaned on it and touched an adjacent bench. Thirty seconds (with a VOM) later I determined that the shell of the bench was 120vAC hot to either of the nearest two benches, and 5 minutes later I knew that it wasn't just high resistance leakage current - you could light a 120v 100 watt light bulb to full brilliance by connecting one side to the "dead" workbenches AC outlet ground screw and the other side to the adjacent benches ground screw. Upon dismantling the bench, we discovered that it was connected with the black wire to the metal framework, and green to the hot wire of the outlets, after all black is ground, right? The breaker feeding the "hot" bench was quickly located and switched off, and locked out until the wiring was corrected by and signed off by the school physical plant department. Yes, we could have swapped green and black ourselves, but to quote our instructor "Never get a union electrician angry at you - and especially when your paycheck is signed by the same person as signs his paycheck". Besides, I think he wanted the campus electrician to see the reversed wiring - it had been that way for at least eight years (can you say "liability"? ... I knew you could).
About five years later I was working at NASA / JPL and was involved
in a similar situation. Our group was charged with the regular maintenace
of all of the electronics equipment used in two buildings dedicated to the ongoing
operation and support of the spacecraft missions. As such, in many cases,
JPL designed and built it's own equipment. One rack mounted system used
36 power supply shelves of four separate power supplies each (+12v, -12v, +5v
and -5v DC). The tech that had wired them had done so in the "electronics"
fashion. He had treated the black wire of the AC power cord as "zero volts",
the white as "hot", and the green as chassis ground, after all black is ground,
right? Both the AC fuse and the AC power switch of each power supply shelf
were wired in the white lead, which left exposed terminals in the power supply
hot to the chassis when the power was switched off - not a good thing when you
have your hand inside the equipment to change one of the eight hard-to-get-at
fuses (one for the AC side and a second for the DC side of each supply), or to
press a reset button on one of the circuit boards.
Note: NASA/JPL always wired their equipment with the hot wire going first to the switch, then to the the fuse holder, and then the load. The idea was that when you switched the power off, everything inside the box is dead.
Unfortunately the equipment was installed in a long row of six-foot-tall equipment cabinet racks in the main computer room of the Space Flight Operations Facility (commonly called the SFOF, pronounced "ess-fof"). This was the main scientific data center for all of JPL, and was used to support multiple simultaneous spacecraft missions. The biggest problem was that the data center had several missions in progress, the most visible of which at that time was Viking... and on a absolutely unchangeable deadline: Viking One was over half way to Mars, Viking Two was not far behind, and simple physics tells you that neither were going to wait for the folks back home to get their equipment rewired! Remember folks, physics makes the universe go 'round. Fortunately the wiring error was discovered (by one of my co-workers) before someone was hurt, and after several discussions between us, our manager, the designer, the SFOF electricians, their supervisor and the Lab safety supervisor it was decided to simply cut the 36 AC power plugs off the equipment power cords (this was in the days of wired-in power cords with molded-on plugs) and to mount 36 brand new plugs with the black and white wires deliberately reversed on the screws of the plugs. Established electrical safety protocol(s) required a red cardboard trouble tag to be duct-taped to each door of each rack, 36 more were taped to the front of each power supply tray, and 36 more were tied to the plug end of each of the 36 power cords. Yes, a bit of overkill, but when two facility electricians, the electricians supervisor, the lab electrical safety specialist, his supervisor, his department manager, the equipment designer, his supervisor and the local federal OSHA representative are breathing down the back of your bosses neck pointing out his workers lives are involved, the cardboard tags are cheap... Each and every tag identically described the same identical deliberate wiring error in the plug, the same identical reason for it, provided the same three contact names and phone numbers, and listed the same will-be-repaired-before date... we should have used a rubber stamp kit... (yes, this was the redundancy proliferation and reduction group in the department of redundancy department in the redundancy control section of the redundancy management division - typical NASA). Plus we were instructed to lock the rack cabinets until the problem was repaired - and locked equipment cabinets was something that was never, ever done in the SFOF equipment rooms. There was a reason that there were CO2 extinguishers hung on wall brackets every 20 feet on every wall. So we cheated - a little - we locked the cabinets but every tech was issued a key the next day (all 26 of us).
At some point in the above discussions the federal OSHA representative casually suggested that the JPL Safety folks have a staff electrician walk the electronics technicians (especially those in the design / construction and maintenance groups) through an AC Power orientation class. A mix of "you're absolutely right", "good idea", some bobble-head-style head nodding and other similar affirmations were heard and seen. By the time the casual suggestion filtered up the three or four management layers, across the bureaucratic morass, and back down again to our group it had turned into two identical formal training classes, one in the morning for 1/2 of the day shift plus the entire night shift, and a second one in the late afternoon for the other 1/2 of the day shift plus the entire evening shift. Attendance at one of them was mandatory which meant that the evening shift and night shift folks got overtime pay (our group provided 24x7x366 support). Plus it would be videotaped for new employees, i.e. due to employee turnover - the newbie could watch the tape and get the full benefit of the class (or so they say...) The first version of this article was written from those 1976 class notes...
Months later, after the first Viking lander was successfully down on Mars (touchdown was on July 20, 1976 at 0512 Pacific Time, and I was in the SFOF watching the video monitors, along with what seemed like every lab employee whose badge gave them access to the building) and the background noise level around the SFOF had quieted back down to something approaching "normal" (we had a little time before Viking Two was due to land), I scheduled progressive equipment outages with the appropriate folks and I budgeted one shift per day across fifteen working days to remove, rebuild and reinstall the power supply shelves. As long as we were doing major rebuild work on the system we also decided to relocate all eight fuses per shelf to indicating fuseholders mounted in the front panel of each power supply shelf, plus we added an external guarded reset button to the front of the logic shelf (in parallel with the one on the logic card... we were tired of removing a cover panel and then twisting our hands into a form of the Vulcan Nerve Pinch to press the secret magic reset button). Another senior tech and I modified two power supply shelves at a time, six shelves per rack, for all six racks of equipment. Plus, as good hams always do, we added a few other necessary improvements, then red-lined a new set of as-built drawings and submitted them back to Documentation Control. As we progressed down the line of cabinets we stripped the red trouble tags off the equipment and gladly tossed them in the trash (I should have kept one of them). We finished in twelve and a half days (our crew of 26 prided itself on always delivering early and under budget).
As a final note, improper residential AC wiring is far too common, and if you
don't have to pull new wire it is usually easy to fix. Recently a good friend
was transferred by his employer to an area near me. He and his wife had been
shown over thirty houses by various agents, and together they trimmed the final list
to five. My friend asked for my help in making the final selection - a fresh
point of view, so to speak. As we walked through the five houses I took about
20-30 digital pictures per house plus written notes on the visible problems / defects
in each house, and I also bent over and plugged an outlet wiring tester into every
AC outlet that was easy to get to (the tester is a rubber plug with three neon lamps
under a clear cap - it is available from most electricial supply houses). Over
40% of the outlets in the five houses we looked at had either open grounds or reversed
hot and neutral wires... this is not a good percentage! And in cases like this
an open ground usually indicates that the selling homeowner has simply replaced the
old 2-wire outlets with new 3-wire outlets to make the house look like it's been
Click here for photos of an outlet tester and of an outlet and GFI tester (sometimes called a GFCI tester). If you are going to buy one new, get the GFI tester version. And note that the neon lamps only need a few milliamps to light up - an outlet can test good but be functionally dead (i.e. not delivering usable power due to a high resistance connection somewhere).
Permission to any group to use this writeup is available for inclusion in any newsletter or bulletin for noncommercial, nonprofit, educational and public safety use within the scope of the U.S. Copyright Law. Just drop me an emailed note.
Copyright © Michael R. Morris WA6ILQ July 2003
wa6ilq /at/ repeater-builder /dot/ com
Back to the top of this page
Back to Tech Index
Back to Home page
I've had several requests for permission to reprint this article, so...
Permission to use this writeup as mentioned above is granted with four trivial conditions: