weird power problem

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

weird power problem

Post by philba » Wed Jan 16, 2008 9:20 am

I recently build a PC with, among other things, a 450W power supply. The PC seems to work fine but has one very weird side effect. When I turn the PC on, the circuit breaker on A DIFFERENT circuit pops. This is pretty reliable but maybe 1 in 20 times, the breaker doesn't pop. The breaker for circuit that the PC is on DOES NOT POP. I do get lights on that circuit flickering.

Both circuits are 20A with pretty heave gauge wire (#10??) and similar length runs. I've run some pretty heavy loads on the PC's circuit though they were resistive (1200W heat gun, for example). The PC load is obviously inductive. I do see the lights flicker when I run heavy loads but the breaker on that circuit hasn't popped once since the wiring was installed in 2001. The circuit that pops, is in the same room as the PC circuit so probably shares the same run.

I presume some sort of wiring problem (L & N reversed??) but have no idea how to track it down or fix it. any help would be greatly appreciated. I'm not opposed to hiring an electrician but would like to diagnose the problem first before starting the meter running.

FOB
Posts: 15
Joined: Thu Nov 29, 2007 9:54 am
Location: Eastern Iowa
Contact:

Wierd Power

Post by FOB » Wed Jan 16, 2008 10:22 am

Did the circuit work OK before the PC was installed? Will the breaker reset and stay set after it trips or do you have to shut off your PC?
You can buy a plug in device that will tell you if the wiring is correct for less than $10 at Home Depot or any electrical supplier.
Do you have any friends interested in Ham radio? you may be introducing RF into the line and that will make circuit breakers act wierd. That may be hard to track down without a O-scope. Try putting a line filter in the wall socket and plug the pc into that. Or try putting a TV near the PC when you turn it on does the picture do strange stuff? Lines and Squiggles? If it does maybe some RF is getting out of the cabinet. Do you have any long lines going to any periferal devices? Sometimes they act as antennas.
Just a few thoughts to try before you rewire the place.
FOB

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Post by Robert Reed » Wed Jan 16, 2008 12:17 pm

"I presume some sort of wiring problem (L & N reversed??) but have no idea how to track it down or fix it."
If you have NEMA standard outlets and you view it from the front, the right side bkade jack is hot to the ground jack and also the left blade jack. The left blade jack will show no voltage to the ground jack, with no load current flowing. Some very small voltage will show under heavy load.This may be a split circuit run from box to that room, in which I beleive they share a common neutral (the idea being that under normal loads, very little current will flow in the neutral) which may cause some problems. One other thing is are these standard magnetic breakers - or do they have electronics associated with them as in the case of GFI?

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Post by philba » Wed Jan 16, 2008 12:44 pm

Lots of good questions:
- yes, standard breakers
- yes, the circuit breaker resets just fine and stays reset, even with the PC running.
- the circuit that pops does have GFIs on it but those do not pop - the standard breaker at the distribution panel in the basement pops.
- neither circuit has had any problems to date.
- the PC is completely contained within the case. it's an aluminum case so it's probably better than most for interference. The only things attached are the obvious ones (kbd, mouse, ethernet cable, display and audio). I see nothing odd on those connections. I'll try the TV test.

If it is the PC (power supply?), how could that cause a problem on a seperate circuit??? it would take a pretty good surge and wouldn't it trip the PCs circuit as well?

I'll get a tester, probably should have one anyway.

Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: weird power problem

Post by Bigglez » Wed Jan 16, 2008 12:47 pm

philba wrote: When I turn the PC on, the circuit breaker on A DIFFERENT circuit pops. This is pretty reliable but maybe 1 in 20 times, the breaker doesn't pop. The breaker for circuit that the PC is on DOES NOT POP. I do get lights on that circuit flickering.
Greetings Philba,

Are either of the branch circuits (for your new PC and the
one that trips) fitted with CGFI devices? That would be my
first suspect. (They may be working and your new PC or
another device has high unbalanced current, possibly a
leakage to ground or turn-on inrush current to ground).

There is a rare but annoying (and potentially dangerous)
wiring fault in US domersitc "split-phase" 120/240V systems
called "open neutral".

Normally the 240V feed from the street is centre-tapped
at the pole or underground transformer to create two
120V phases (L1, L2). If the loads in the home are
symmetrical, current does not flow in the neutral (N).

Out of balance loads cause a current to flow in the N
conductor to restore balance.

Using a DMM read the voltage of the two busses (L1, L2)
in the breaker panel, with respect to N. (You will need
to remove the safety cover inside the panel, and the
busses are hot - be careful).

Try to load one or other of the two suspect branch
circuits. If the N is bonded correctly both L1 and L2
will be very close to the same voltage.

We had an open neutral in our present home, and
only found it after the first Christmas, when adding
40 - 60 Amps for decorations caused an unbalance
that drove L1 up to 140V and L2 down to 100V -
enough to burn up the garage door motor on L2,
and blow out many of the lamps on the L1 side.

Prior to finding this problem we had similar unexpected
breaker popping and light flickering as you described.

Comments Welcome!

Peter

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Post by philba » Wed Jan 16, 2008 1:11 pm

that's interesting, peter. I'm not sure I can contrive a scenario to measure the open neutral problem as it's a rather large house with a lot of electrical circuits - I have 3 distribution panels in the basement. Is there another way to determine if I have an open neutral?

The circuit that pops does have GFCIs on it but they don't pop. How would that cause the breaker in the panel to pop?

Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Post by Bigglez » Wed Jan 16, 2008 1:45 pm

philba wrote: Is there another way to determine if I have an open neutral?
Greetings philba,

Yes. Any difference in potential between the N from the
street and a local N is a problem, as they should
be bonded. Measuring the bus voltages at the main
panel, and again at the sub-panels will show up any
open N. The measurement needs two loads (first on
L1, then on L2). Your 1200W heat gun or similar
heavy 120V load will do the trick.
philba wrote:The circuit that pops does have GFCIs on it but they don't pop. How would that cause the breaker in the panel to pop?
Okay, I was wondering if you had CGFI breakers
in the main (or sub) panel(s)? A disturbance on one
phase (L1 or L2) can induce a trip of a CGFI on the
other phase.

With CGFIs in outlets or portable power strips things
are a little fuzzy and harder to diagnose.

Should be interesting to read your test results when
you get an outlet tester (a very good tool as suggested
by someone else).

Final thought, breakers fatique and if heavily tripped
will likely not reset properly or show up as nuisance
tripping. These are easy to replace by a homeowner.

Comments Welcome!

Peter

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Post by philba » Wed Jan 16, 2008 7:50 pm

I was wrong, the circuit that trips is lighting only. However, I now believe it to be a GFCI circuit breaker. The clue should have been the little yellow test button on the breaker. sigh.... sorry guys. I don't understand why the electricians installed a GFCI on a lighting only load.

I got the circuit tester and the circuit that the computer is on shows up as wired correctly. I measured the circuit that trips at a switch (since no outlets). Black to White (L to N, I believe), measures 121VAC. L to Gnd - 121 VAC, N to Gnd - 0 (actually about 50 mV). At the panel, both circuits appear to have L coming from the breakers. So, I think both circuits are wired correctly.

Anyway, is it possible the GFCI breaker has gone bad? It's a square D, Type HOM, Series 1, labeled as 20A. I couldn't find the specific breaker on the square D web site. Since it's a lighting only load, I assume that it would be ok to replace with a standard (i.e. non-GFCI) breaker. Thoughts?

User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Post by jollyrgr » Wed Jan 16, 2008 8:30 pm

philba wrote:I was wrong, the circuit that trips is lighting only. However, I now believe it to be a GFCI circuit breaker. The clue should have been the little yellow test button on the breaker. sigh.... sorry guys. I don't understand why the electricians installed a GFCI on a lighting only load.

{SNIP}

Thoughts?
The circuit that is LIGHTING ONLY, does it happen to be lighting in one or more bedrooms? Can you read the part number off the circuit breaker? If so, post or PM this information. DO NOT replace this breaker just yet!

Try this: Turn the PC on making sure all the lights are OFF on the circuit that trips. Turn the computer on and off a couple times. You don't even have to let the computer fully boot, just turn it off before the OS begins to load. After each time of cycle on and off of the computer, check to see if the breaker is on or tripped. (Turn a light on that circuit and turn it back off before trying the computer again.)

Then go turn on all the lights on that circuit. Turn on and off the PC several times. I'm going guess that it only trips the breaker with the lights on and when you turn on the computer.

Just a guess without seeing it, I don't believe the breaker is a GFCI even though it may look like one. I won't venture my thoughts until I read your answers. By chance, is the house relatively new (within the last five years or so)?
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Post by Robert Reed » Wed Jan 16, 2008 8:54 pm

Philba
I just went thru a weird experience with my panel mounted Square D GFI breaker, although not as weird as yours. Since you mentioned Square D, and it does sound like a defective breaker, I thought I would toss this at you. First of all and to my surprise, a Home Depot replacement would not fit the bus bar tabs properly due to the fact that there are several different styles even in the S D brands. Mine needed to be a 'QO' series and none available locally (could be ordered with a $75 price tag). A search on E-Bay and two days later UPS dropped off a brand new one at my door step for $24 dollars including shipping. If you have that odd series and need one I think I can still find the E-Bay adress. But better yet, just purchase a standard breaker w/o GFI. That feature is not needed for your puposes. BTW, its quick and easy to temporarily sub in another breaker from another circuit in your box so as to prove your suspicions.

User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Post by jollyrgr » Wed Jan 16, 2008 10:00 pm

PLEASE NOTE



I suspect something else is up. DO NOT replace the breaker with a normal one just yet. This may very well not be a GFCI breaker. Even if it is there is probably a CODE reason that is there.
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!

SETEC_Astronomy
Posts: 582
Joined: Tue May 09, 2006 12:44 am
Contact:

Post by SETEC_Astronomy » Wed Jan 16, 2008 10:18 pm

Is it possibly an Arc fault breaker and the inductance of the PC supply is tripping it? I'm not sure how long they've been around for home use but I'm pretty sure it's now code in some areas to include Arc fault breakers for certain or all breakers in new homes. I'm just tossing something out there.......

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Thu Jan 17, 2008 2:14 am

Hi there,

Just wondering, did you try to plug it into another outlet using a long
extension cord to see what happens? This might tell you something.
LEDs vs Bulbs, LEDs are winning.

User avatar
haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup » Thu Jan 17, 2008 11:26 am

Well, if you have the circuit tester, plug it in, press the test button and if the breaaker trips, its a GFCI. If it does not, its probably an AFCI.

AFCI is now required in bedrooms while GFCI is required in bathrooms and kitchens. If I recall, AFCI have a yellow test button and the GFCI had a red test button (but I need to check that when I get home tonight in my box). Since GFCI is cheaper and easier to retrofit in a wall receptacle, we tend to find them there but AFCI only have breaker versions and are only found in the panel.

my AFCI has never tripped but I have heard that some can be sensitive. Try moving the breaker to the other phase (i.e 240V in gives two 110V phases). In most cases this will be the next slot up or down in a square D panel. you may have to swap with another breaker if you're full. Just be careful to keep the hots on a 3 wire cable on opposite phases to balance the load. In fact if your two effected circuits come out on the same 3 wire cable, that may be your problem right there.

An easy test to see if the branches are on different phases without opening the breaker box is to just measure the voltage difference between the hot wires on each branch. It will either be 0V or 220V for same or different phases. Anyway, changing phases is not a guaranteed fix but its an easy thing to try for no $.

Oh and #10 for 20A is overkill, you probably have #12 which is the typical size for a 20A branch.

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Post by philba » Thu Jan 17, 2008 2:38 pm

It's probably a GFCI. It has the letters GFCI on it (in incredibly tiny lettering). The lighting is in both a bedroom (commandeered as my office) and a bathroom. Electrical was done in 2001 so I'm pretty sure it's not an arc fault circuit interrupter which became code is 2002. It does, however, have a yellow test button.

Jollyrgr, the circuit almost always trips. It seems to be invariant whether there is a load or not. I made sure all the lights were on and powered up the computer - the breaker tripped. I then reset the breaker, turned all the lights off and powered up the PC - the breaker tripped. The info from the breaker is pretty much what I posted earlier (Type HOM, Series 1, labeled as 20A).

Al, I will try using an outlet on another circuit.

Post Reply

Who is online

Users browsing this forum: Google [Bot] and 37 guests