Notes on the Troubleshooting and Repair of Small Household Appliances and Power Tools


  15.17) Window air conditioner doesn't cool

This means the fan runs but you do not hear the compressor kick in.

It could be several things:

* If you hear the 'click' of the thermostat but nothing happens (Your room
  lights do not dim even for a second) and there is no other sound, it could
  be bad connections, bad thermostat, dirty switch contacts, bad compressor,
  etc.  Or, you have it set on fan instead of cool.  Try cycling the mode
  selector switch a couple times.

* If you do not hear a click at all, then the thermostat is probably bad or
  it is cooler in your room than you think!  Try tapping on the thermostat.
  Sometimes they just stick a bit after long non-use.

* If you get the click and the lights dim and then a few seconds later there
  is another click and the lights go back to normal, the compressor, or its
  starting circuitry is bad.  It is trying to start but not able to get up to
  speed or rotate at all.

Except for a bad compressor, all these are repairable relatively inexpensively
but if it is real old, a new high efficiency model may be a better solution.

  15.18) Air conditions freezes up

When this happens, airflow is reduced greatly since ice is blocking the
evaporator.  Turning the unit off for a while or running it on fan-only
will clear the ice but this may indicate the need for maintenance or an
actual problem.  Similar comments apply to window and central air conditioners
as well as heat pumps.

The three major causes of an air conditioner freezing up are:

1. Reduced airflow due to a dirty filter or clogged evaporator.  If you are
   not aware that there is a filter to clean, this is probably the cause :-).

2. Low Freon.  While your intuition may say that low Freon should result in
   less cooling, what happens is that what is there evaporates too quickly
   and at the input end of the evaporator coils resulting in lower temperatures
   than normal at that end (which results in condensed water vapor freezing
   instead of dripping off) but part of the evaporator will likely be too warm.

   You cannot fix this yourself without specialized equipment.  For a room
   air conditioner that isn't too old, it may be worth taking it in to a
   reputable shop for an evaluation.  For a central air conditioner, you will
   have to call an HVAC service company for repairs.

   The fact that the Freon is low means that there is a leak which would also
   need to be repaired.  Freon does not get used up.

3. Outside and/or inside temperature may be very low.  The unit may not be
   designed to operate below about 65 degrees F without freezing up.

If it is 90 degrees F and you have full air flow with the fan set on high and
still get the freezup on a part of the evaporator, then low Freon is likely.

  15.19) Comments on electric clothes dryer problems and repair

For quite a lot of useful information, do a web search for 'appliance 
repair'.  There are a couple of decent sites with DYI information.

(From: Bernie Morey (bmorey@aardvark.apana.org.au)).

I've repaired our electric dryer several times over the years and kept it 
going well beyond its use-by date.

My main problems have been:

1. Mechanical timer failure.  Easy fix.

2. Leaking steam damaging the element.  Have replaced element twice -- fairly
   easy job.  Had to replace some stainless stand-offs at the same time.  
   Elements readily available and equivalent of USD24 each.

3. Bearing replacement -- have to be done carefully or they don't last.

4. Belt replacement.  (Make sure you center the belt with respect to the idler
   and rotate the drum by hand to double check it before buttoning things up.
   Else, it may pop off the first time the motor starts.  --- sam).

5. Exhaust fan bearing replacement.  This was the trickiest, although far 
   from impossible.  It is a sealed unit subject to high heat and dust 
   contamination -- not a good environment.

The only problem for the past two years has been the dryer throwing the 
exhaust fan belt.  Cleaning up the fluff fixes it for another year.

Did all these without any guide -- just carefully inspecting the work before 
starting and making diagrams of wiring and ESPECIALLY the main drum belt.  I 
generally have to get my wife to help me with the main belt -- hard to get 
the tensioner in position while stopping the belt slipping down the far side 
of the drum.

These things are mechanically and electrically pretty simple -- if it's not 
working the fault is usually obvious.

  15.20) Dryer shuts down after a few minutes

There are multiple thermostats in a dryer - one that sets the air temperature
during normal operation (and controls power to the heating element) and one or
more that sense fault conditions (and may shut everything down) such as those
described below.

(From: Bernie Morey (bmorey@aardvark.apana.org.au)).

The dryer is likely cutting out because a thermostat is tripping.  The
fundamental reason is probably that the exhaust air is too  hot.  And the air
flow is probably too hot because it is restricted -- lower  volume of air at
higher temperature.  Check these things out:

1. Lint filter.  Although these can look clean (and I assume you do clean 
   it after every load!) the foam variety can gradually clog up with very fine 
   dust and restrict air flow.  If it's a foam disk, a new one is fairly cheap.

2. Can you feel the exhaust air?  If not, the exhaust fan belt may be worn 
   broken or slipping.  The exhaust fan bearing could be partly seized -- try 
   turning the fan by hand and check for stiffness.

3. Air outlet blockage.  Lint and dust may have built up in the exhaust 
   side of the machine.  Check for restrictions.  Our machine just vents up 
   against the laundry wall as it is too difficult to vent it to the outside. 

   Outside vents are often plastic tubing with a spiral spring steel coil for 
   stiffness -- check for kinks or obstructions.

4. 'Clutching at straws' Dept #1:  Element may have developed a hot-spot 
   near a thermostat.  Involves dismantling the machine and checking the 
   element.  NB -- if you dismantle the machine, make a diagram of how the 
   drive belt fits over the drum, motor, and idler!

5. 'Clutching at straws' Dept #2:  The drum may be restricted from turning 
   freely.  This would slow the motor and hence the exhaust fan.  Check for 
   socks, women's knee-highs (these thing seem to breed everywhere!) & caught 
   near the bearings (probably the front).

You cannot completely check the thermostat with a meter -- they are either
open or closed.  To test it properly you would have to know the temperature 
at which it opens (from the manufacturer's specs), and then measure the 
temperature of the exhaust air with a probe while watching the thermostat. 

  15.21) Why has my dryer (or other high current) plug/socket burned up?

This sort of failure is not unusual.  The brass (or whatever) corrodes a
bit over time and/or the prongs loosen up.  It doesn't take much resistance
at 20 or 30 Amps to produce a substantial amount of heat.  The hotter it gets,
the more the resistance goes up, heating increases, it loosens more, and so
on until something melts.  The power is I*I*R (where I is current and R
is the resistance) so at 20 A, a .1 ohm resistance at the contact results
in 40 W - think of the heat of a 40 W light bulb.

An exact cause would be hard to identify.  However, only the plug and
receptacle are involved - this is not a case of an outside cause.  Such
a failure will not normally blow a fuse or trip a breaker since the current
does not increase - it is not a short circuit.

It is definitely wise to replace both the plug and receptacle in such
cases since at the very least, the socket has lost its springiness due to
the heating and will not grip well..  Make sure that the prongs on the new
plug make a secure fit with the socket.

On plugs having prongs with a pair of metal strips, spreading them out a bit
will make much better contact in an old receptacle.

In general, if a plug is noticeably warm, corrective action should be taken
as it will likely get worse.  Cleaning the prongs (with 600 grit sandpaper)
and spreading the metal strips apart (if possible) should be done first but
if this does not help much, the plug and/or socket should be replaced.
Sometimes, the original heating problem starts at the wire connections to
the plug or socket (even inside molded units) - loose screws, corroded wires,
or deteriorated solder joints.

  15.22) Four year old gas dryer just started popping GFCI

Why is it on a GFCI in the first place?  A grounded outlet is all the
protection that is needed and any type of appliance with a motor or transformer
could be a potential nuisance tripper with a GFCI (though not always).

As to why it is now different, I assume that this is a dedicated outlet so
nothing else you added could affect it.  Thus you are left with something
changing in the dryer or the GFCI somehow becoming overly sensitive.

It is possible that there is now some electrical leakage in the dryer wiring
just from accumulated dirt and grime or dampness.  This could be measured
with an AC milliamp meter or by measuring the resistance between the AC wires
and the cabinet.  If this test shows up nothing, I would recommend just
putting on a grounded outlet without a GFCI.  It could also be that due
to wear, the motor is working harder at starting resulting in just a tad
more of an inductive current spike at startup.

  15.23) Checking dishwasher solenoids

(From: Filip "I'll buy a vowel" Gieszczykiewicz (filipg@repairfaq.org)).

Greetings. Well, since it's a moist/damp environment... I'd suspect a
bad connection first. You will need to pop off the front bottom panel
and get at the wires that actually connect the solenoid to the timer
motor (and/or wire harness). You will need an ohmmeter to check the
resistance of the coil - if it's OK (20-200 ohms I would guess), that's
not the problem. Well, that leaves you with pretty much the wires that
connect the timer motor (a MULTI-contact switch driven by a timer motor
like those found in old clocks that plugged into outlets) and the switch
itself. I hope the dishwasher is unplugged... Since the dishwasher
operates as a closed system (because of the "darned" water :-) it will
be difficult to test it in circuit. I suggest that you try to trace
the wires that come off the solenoid to their other ends... and then test
the wires themselves. If you feel this is too much for you, call the
repair folks - ask around... see if anyone else knows a particular
service that has a good record...

Chapter 16) Electrical Wiring Information and Problems

  16.1) Safe electrical wiring

This chapter is in no way intended to be a comprehensive coverage of wiring
issues but includes a discussion of a few of the common residential wiring
related questions.  For more information, see the official Usenet Electrical
Wiring FAQ or a DIY book on electrical wiring.  The NEC (National Electrical
Code) handbook which is updated periodically is the 'bible' for safe wiring
practices which will keep honest building inspectors happy.  However, the NEC
manual is not what you would call easy to read.  A much more user friendly
presentation can be found at the CodeCheck web site:

* http://www.codecheck.com/

This site includes everything you always wanted to know about construction
codes (building, plumbing, mechanical, electrical) but were afraid to ask.

In particular, the following series of sections on Ground Fault Circuit
interrupters is present at the CodeCheck web site and includes some nice
graphics as well.

  16.2) What is a GFCI?

A Ground Fault Circuit Interrupter (GFCI) is a device to protect against
electric shock should someone come in contact with a live (Hot) wire and
a path to ground which would result in a current through his/her body.  The
GFCI operates by sensing the difference between the currents in the Hot and
Neutral conductors.  Under normal conditions, these should be equal.  However,
if someone touches the Hot and a Ground such as a plumbing fixture or
they are standing in water, these currents will not be equal as the path is
to Ground - a ground fault - and not to the Neutral.  This might occur
if a short circuit developed inside an ungrounded appliance or if someone
was working on a live circuit and accidentally touched a live wire.

The GFCI will trip in a fraction of a second at currents (a few mA) well
below those that are considered dangerous.  Note that a GFCI is NOT a
substitute for a fuse or circuit breaker as these devices are still
required to protect equipment and property from overloads or short
circuits that can result in fire or other damage.

GFCIs can be installed in place of ordinary outlets in which case they
protect that outlet as well as any downstream from it.  There are also
GFCIs that install in the main service panel.

Note that it may be safe and legal to install a GFCI rated at 15 A on a
20 A circuit since it will have a 20 A feed-through.  Of course, the GFCI
outlet itself can then only be used for appliances rated 15 A or less.

Many (if not most) GFCIs also test for a grounded neutral condition where a
low resistance path exists downstream between the N and G conductors.  If such
a situation exists, the GFCI will trip immediately when power is applied even
with nothing connected to the protected outlets.

  16.3) GFCIs, overloads, and fire safety

A GFCI is NOT a substitute for a fuse or circuit breaker (unless it is a
combined unit - available to replace circuit breakers at the service panel).

Therefore, advice like "use a GFCI in place of the normal outlet to prevent
appliance fires" is not really valid.

There may be some benefit if a fault developed between Hot and Ground but that
should blow a fuse or trip a circuit breaker if the outlet is properly wired.
If the outlet is ungrounded, nothing would happen until someone touched the
metal cabinet and an earth ground simultaneously in which case the GFCI would
trip and provide its safety function.  See the section: "Why a GFCI should not be used with major appliances" for reasons why this is not generally desirable
as long as the appliance or outlet is properly grounded.

However, if a fault occurs between Hot and Neutral - a short in the motor, for
example - a GFCI will be perfectly happy passing almost any sort of overload
current until the GFCI, wiring, and appliance melts down or burns up - a GFCI
is not designed to be a fuse or circuit breaker!  That function must be
provided separately.

  16.4) How does a GFCI work

GFCIs typically test for the following condition:

1. A Hot to Ground (safety/earth) fault.  Current flows from the Hot wire to
   Ground bypassing the Neutral.  This is the test that is most critical for

2. A grounded neutral fault.  Due to miswiring or a short circuit, the N and G
   wires are connected by a low resistance path downstream of the GFCI.  In
   this case, the GFCI will trip as soon as power is applied even if nothing
   is connected to its protected (load) circuit.

To detect a Hot to Ground fault, both current carrying wires pass through the
core of a sense coil (transformer).  When the currents are equal and opposite,
there is no output from its multiturn sense voltage winding.  When an imbalance
occurs, an output signal is produced.  When this exceeds a threshold, a circuit
breaker inside the GFCI is tripped.

GFCIs for 220 VAC applications need to monitor both Hots as well as the
Neutral.  The principles are basically the same: the sum of the currents in
Hot1 + Hot2 + Neutral should be zero unless a fault exists.

To detect a grounded neutral fault, a separate drive coil is continuously
energized and injects a small 120 Hz signal into the current carrying
conductors.  If a low resistance path exists between N and G downstream
of the GFCI, this completes a loop (in conjunction with the normal connection
between N and G at the service panel) and enough current flows to again
trip the GFCI's internal circuit breaker.

GFCIs use toroidal coils (actually transformers to be more accurate) where the
core is shaped like a ring (i.e., toroid or doughnut).  These are convenient
and efficient for certain applications.  For all practical purposes, they are
just another kind of transformer.  If you look inside a GFCI, you will find a
pair of toroidal transformers (one for H-N faults and the other for N-G faults
as described above).  They look like 1/2" diameter rings with the main current
carrying conductors passing once through the center and many fine turns of
wire (the sense or drive winding) wound around the toroid.

All in all, quite clever technology.  The active component in the Leviton
GFCI is a single chip - probably a National Semiconductor LM1851 Ground Fault
Interrupter.  For more info, check out the specs at National'a web site at: http://www.national.com/pf/LM/LM1851.html.

  16.5) More on how the GFCI detects a N-G short

To detect a Neutral to Ground fault there is a second transformer placed
upstream of the H-G sense transformer (see the illustration of the internal
circuitry of the GFCI at: http://www.national.com/pf/LM/LM1851.html).  A small
drive signal is continuously injected via the 200 T winding which induces
equal voltages on the H and N wires passing through its core.

* If N and G are separate downstream (as they should be), no current will be
  flow in either wire and the GFCI will not trip.  (No current will flow in
  the H wire as a result of this stimulus because the voltage induced on both
  H and N is equal and cancels.)

* If there is a N-G short downstream, a current will flow through the N wire,
  to the G wire via the short, and back to the N wire via the normal N-G
  connection at the service panel.  Since there will be NO similar current in
  the H wire, this represents a current unbalance and will trip the GFCI in
  the same manner as the usual H-G short.

* Interestingly, this scheme automatically detects a H-H fault as well.  This
  unlikely situation could occur if the Hots from two separate branch circuits
  were accidentally tied together in a junction box downstream of the GFCI.
  It works the same way except that the unbalance in current that trips the
  GFCI flows through the H wire, through the H-H fault, and back around via
  the Hot busbar at the service panel.  Of course if the two Hots are not on
  the same phase, there may be fireworks as well :-).

  16.6) GFCIs and safety ground

Despite the fact that a Ground Fault Circuit Interrupter (GFCI) may be
installed in a 2 wire circuit, the GFCI does not create a safety ground.
In fact, shorting between the Hot and Ground holes in the GFCI outlet
will do absolutely nothing if the GFCI is not connected to a grounded
circuit (at least for the typical GFCI made by Leviton sold at hardware
stores and home centers). It will trip only if a fault occurs such that
current flows to a true ground.  If the original circuit did not have a
safety ground, the third hole is not connected.  What this means is that
an appliance with a 3 prong plug can develop a short between Hot and the
(supposedly) grounded case but the GFCI will not trip until someone
touches the case and an earth ground (e.g., water pipe, ground from
some other circuit, etc.) at the same time.

Note that even though this is acceptable by the NEC, I do not consider it
desirable.  Your safety now depends on the proper functioning of the GFCI
which is considerable more complex and failure prone than a simple fuse or
circuit breaker.  Therefore, if at all possible, provide a proper Code
compliant ground connection to all outlets feeding appliances with 3 wire

  16.7) Where are 3 wire grounded outlets required?

If you move into a house or apartment where some or all of the outlets are the
old 2 prong ungrounded type, don't panic.  There is no reason to call an
electrician at 2:00 AM in the morning to upgrade them all at great expense.

You don't need grounded outlets for two wire appliances, lamps, etc.  They
do essentially nothing if the third hole isn't occupied :-).  A GFCI will
provide much more protection!

You should have grounded outlets for the following:

* Computers in order for the line filters and surge suppressors to be most

* High-end entertainment gear if it uses 3 prong plugs for similar reasons.

* Microwave ovens.  For safety, these really should be on a grounded circuit.
  (A GFCI will not protect against a fault on the high voltage side of a
  microwave oven, though this sort of fault is extremely unlikely).

* Large appliances including refrigerators, clothes washers and dryers,
  dehumidifiers, window air conditioners, etc.

In most cases, there will only be a few circuits where this is needed and only
these need to be upgraded.  To what extent the wiring plan of your residence
separates lighting type circuits from those with outlets that will be used for
3 wire equipment will determine how easy it is to upgrade only those outlets
that are affected.  It may be cheaper to just add new branch circuits for
specific equipment needs.

  16.8) Why you should NOT connect G to N

The question often arises: "Why can't I just connect the G to the N if my
outlets are only two prong?"

For one reason, consider the 'appliance' below:

             |                 |        Open Fault
  Hot o---------o-o----/\/\---------+------ X -----o Neutral
             | Switch  Load    |    |
             |  (On)           |----+ Case should be G but is connected to N

With the appliance 'on', current passes through the internal wiring/motor/etc.
of the appliance to the N but this is now connected to the case as well.  If
the house wiring opens (or even if the plug is loose, it is possible to have
line voltage on the case.

  16.9) Testing installed GFCIs

The built-in tester is designed to actually introduce a small leakage
current so its results should be valid.  Therefore, testing a single
GFCI outlet with an external widget is not really necessary except for
peace-of-mind.  However, such a device does come in handy for identifying
and testing outlets on the same circuit that may be downstream of the GFCI.

An external tester is easy to construct - a 15 K ohm resistor between H and G
will provide a 7 mA current.  Wire it into a 3 prong plug and label it "GFCI
Tester - 7 mA".  The GFCI should trip as soon as you plug the tester into a
protected outlet.  On a GFCI equipped for grounded neutral detection (as most
are), shorting the N and G conductors together downstream of the GFCI should
also cause it to trip.

Note that such a tester will only work for GFCI protected outlets that are on
grounded (3 wire) circuits (unless you add an external ground connection).
Thus, just using a commercial tester may falsely indicate that the GFCI is bad
when in fact it is simply on an ungrounded outlet (which is allowed by Code in
a retrofit situation).

The test button will work whether or not the circuit includes a safety ground
because it passes an additional current through the sense coil between Hot and
Neutral tapped off the wiring at the line side of the GFCI and therefore
doesn't depend on having a safety Ground.

I suppose you can purchase suitable low cost testers as well (but they are
subject to the same must-be-grounded restrictions).  Try your local home
center or electrical supply distributor.

  16.10) John's comments on the use of GFI breakers

(From: John Grau (affordspam@execpc.com)).

I personally would not feed a subpanel with a GFI breaker.  Here are just a
few of the reasons:

1. GFI breakers for personnel protection are set to trip at 5 mA (1/1000ths of
   an Amp). The longer the circuit conductors, the greater the potential for
   leakage. If you subfeed a panel, you would have the cumulative distances of
   all circuits connected to that panel to contend with and hope that the
   breaker would hold.

2. You would not be able to connect any thing to that subpanel that would be a
   critical load. e.g. freezer, sump pump, well pump, furnace, etc. An
   unnoticed nuisance trip, could mean that you would come home to a thawed
   freezer, frozen pipes, flooded basement, etc.

3. Using breakers to achieve GFI protection has 2 downsides: expense, and
   usually, an inconvenient location to reset the tripped device. A GFI outlet
   at the point of usage, is usually more convenient to reset, should it trip.
   Here in Wisconsin, I can buy about 6 GFI outlets for the cost of 1 breaker. 

There is no compulsory language in the National Electrical Code the forces an
update to current code standards, unless you repair, replace or update the
affected component. Not all changes in the 1996 code made sense, and I would
not update the wiring in my own home (built in 1995) to current standards.

  16.11) Phantom voltage measurements of electrical wiring

When making measurements on household wiring, one expects to see one of three
voltages: 0, 115 VAC, or 230 VAC (or very similar).  However, using a typical
multimeter (VOM or DMM) may result in readings that don't make sense.  For
example, 2 VAC between Neutral and safety Ground or 40 VAC between a Hot wire
(with its breaker off) and Neutral or safety Ground.

The most likely reason for these strange readings is that there is E/M
(electromagnetic) coupling - capacitive and/or inductive - between wires which
run near one another - as inside a Romex(tm) cable.  Where one end of a wire
is not connected to anything - floating, the wire acts as an antenna and picks
up a signal from any adjacent wires which are energized with their 60 (or 50)
Hz AC field.  There is very little power in these phantom signals but due to
the very high input resistance/impedance of your VOM or DMM, it is picked up
as a voltage which may approach the line voltage in some cases.

Another possibility is that the you didn't actually walk all the way down to
the basement to shut off power completely and the circuit is connected to a
high tech switch (such as one with a timer or an automatic dimming or off
feature) or a switch with a neon light built in.  There will be some leakage
through such a switch even if it is supposed to be off - kill power completely
and test again.

Putting any sort of load between the wires in question will eliminate the
voltage if the cause is E/M coupling.  A small light bulb with test probes can
be used to confirm this both by serving as a visual indication of significant
voltage (enough to light the bulb, if weakly) and to short out the phantom
voltage for testing with the multimeter.

There can be other causes of such unexpected voltage readings including
incorrect or defective wiring, short circuits in the wiring or an appliance,
and voltage drops due to high current in a circuit.  However, the E/M coupling
explanation is often overlooked when using a multimeter.

  16.12) Determining wiring of a 2-wire outlet

Connect a wire between one prong of a neon outlet tester and a known ground -
cold water pipe if copper throughout, heating system radiator, ground rod, etc.

(Experienced electricians would just hold onto the other prong of the tester
rather than actually grounding it.  Their body capacitance would provide
enough of a return path for the Hot to cause the neon to glow dimly but
you didn't hear this from me :-).  Yes, they survive without damage and don't
even feel anything because the current is a small fraction of a mA.  DON'T
try this unless you are absolutely sure you know what you are doing!)

With one prong grounded, try the other prong in the suspect outlet:

* The Hot should glow brightly and the Neutral should not light at all.  This
  is the normal situation.

* If neither side glows, the fuse is blown, the circuit breaker is tripped,
  this is a switched outlet and the switch is off, or there is a wiring
  problem elsewhere - or your ground isn't really ground.

* If both sides glow and using the tester between the slots results in no glow,
  then you have an open Neutral and something else on the circuit that is on is
  allowing enough current to flow to light the neon tester.

* If both sides glow and using the tester between the slots results in an even
  brighter glow, the outlet is wired for 220 V, a dangerous violation of the
  NEC Code unless it is actually a 220 V approved outlet.  It is unlikely you
  will ever see this but who knows what bozos worked on your wiring in the

  16.13) Outlet wiring screwed up?

So your $6 outlet tester displays a combination of lights that doesn't
make sense or one or more lights is dim.  For example, all three lights
are on but K and X (see below) are dim.

The three neon bulbs are just between what should be (The first letter is
how the light is marked on mine):

K Hot to Ground (GROUND OK).
O Hot to Neutral (HOT OK?).
X Neutral to Ground (HOT/NEUTRAL REVERSE - should not light).

I suspect at the very least that your ground is not connected at the
service panel.  I may run from some/all the outlets but ends somewhere.
You are seeing capacitive/inductive pickup between the floating ground
and the other wires in the circuit.  Your N and H may be reversed as well
but this cannot be determined without checking with a load between H/N
and a proper ground.

I would recommend:

1. Determining if the ground wire for those 3 prong outlets does indeed
   go anywhere.

2. Determining if the Hot and Neutral polarity is correct by testing between
   each of the prongs and a confirmed ground (properly connected 3 prong
   outlet, service panel, or a cold water pipe in an all metal water system)
   with a load like a 25 W light bulb.  The neon lamps in the tester or a high
   impedance multimeter can be fooled by capacitance and other leakage paths.

For a computer or other 3 wire appliance, you should really install a
proper 3 prong outlet wired correctly.  Otherwise, any power line filters
and surge suppressors will not have the safety ground (which a GFCI does
NOT create).  Some UPSs may get away without one but then their surge
suppressor and/or line filters will not work correctly.

Some appliances like microwave ovens MUST have a proper safety ground
connection for safety.  This not only protects you from power line
shorts to the case but also a fault which could make the case live
from the high voltage of the microwave generator.

  16.14) 220 V outlet reads 0 VAC between slots

"I have a 220 outlet that I need to plug an AC unit into.  The AC unit works
 fine in another outlet, but not in this specific outlet.  I pulled out my
 handy dandy meter and checked the voltage across the two line slots - the
 meter read 0.

 But when I tried one line and the ground I got 125 V.  Similarly, when I
 tried the other line and the ground I also got 125 V.  What's the scoop?  Why
 does the meter, and obviously the AC, think that there isn't 220 V coming in?
 Any help is greatly appreciated - as this room is stinking hot right now!"

Did it ever work?  It sounds like both slots are being fed from the same
phase of the power from the service panel.  Check with a load like a 100
W light bulb between each slot and ground.  This could have happened during
the original installation or during renovation.

Another possibility is that there is some other 220 V appliance on the
same line with its power switch in the ON position (and not working either)
AND one side of the line has a tripped breaker or blown fuse.

Yet another possibility:

(From: David L. Kosenko (davek@informix.com)).

My load center is GE unit.  They make both full height and half height breakers.
If you use a half height breaker set for a 220 line, you must be careful to
install it across the two phases.  It is very easy (especially if you don't
know about 220) to place the ganged breakers into a single full height slot in
the load center, giving you both lines off the same phase line.

  16.15) Testing for fault in branch circuit

This may trip the breaker or blow a fuse - or trip a GFCI if so protected.
The procedure below is specifically for GFCI tripping.  You will need a

* First, unplug everything from the circuit and see if it still trips.  If
  it now does not trip, one of the appliances was the problem.  Try them one
  at a time to see which is the problem and then check the section for that
  or a similar appliance elsewhere in this document.

Assuming the circuit is at fault:

* You need to determine whether this is a H-G leakage fault (which is what
  most people think is the only thing GFCIs test for) or a shorted G-N fault.

* A H-G fault that doesn't trip the normal breaker might be due to damp
  wiring (an outside outlet box that gets wet or similar) or rodent damage.

* A shorted G-N fault means that G and N are connected somewhere downstream
  of the GFCI - probably due to incorrect wiring practices or an actual
  short circuit due to frayed wiring or wires touching -  damage during
  installation or renovation.

Assuming the line is separate from any other wiring:

* With the line disconnected from the service panel (all three wires), first
  test between each pair of wires with the multimeter on AC to make sure it is
  truly dead - there should be virtually no voltage.  H-G, N-G, and H-N should
  all be close to 0 (say, less than a volt).

* If this passes, test across the dead line's H and G for leakage on the
  resistance range.  It should be greater than 15 K ohms (it should really
  be infinity but to trip the GFCI requires around 15 K ohms or less).

* Then, test for resistance between H and G - this too should be infinity.

One of these will show a fault - possibly the N-G test indicating a short or
improperly wired outlet since this would not result in any operational
problems until a GFCI is installed (though it does represent a safety

Go to [Next] segment
Go to [Previous] segment

Go to [Table 'O Contents]

Written by Samuel M. Goldwasser. | [mailto]. The most recent version is available on the WWW server http://www.repairfaq.org/ [Copyright] [Disclaimer]