Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Monitor malfunctions and electrical problems

I think the confusion is what we mean by an "office environment."  Most
suburban offices get single-phase drops just like a residence.  Even
non-factory office buildings which have three-phase drops will rarely
carry it around into office space, and may only have it easily available
in some place like a boiler room.

In a large office building, the drop will be three-phase, but the lighting
circuits themselves are not.  There may well be three separate legs, taken
from the three-phase drop, but -- and this is where the confusion lies --
the power is single-phase from the perspective of any particular light
fixture.  There is effectively no such thing as three-phase office lights.

The important point is that, in a multi-phase system, the neutral is
common to all legs.  However, the total current into loads is always going
to equal the total current out of loads -- this is actually Kirchoff's
Law.  One of the main purposes of multi-phase service is to "timeshare"
the neutal so that instantaneous current never exceeds its instantaneous
ampacity rating.  If the loads have a phase changing effect, which is to
say that they are highly inductive or capacitive, then the current on the
neutral can become excessive and this, in turn, can lead to low voltage at
the loads because of the resulting drop.  Nevertheless, computer power
supplies are strictly power-consuming devices and it is something of a
misnomer to talk about them as "generating" current, harmonic or
otherwise, although almost no computer power supplies are linear.

Voltage supplied to the monitor which is too low can cause shake.  
However, it will also usually cause all sorts of other problems that show
up before shake, such as pincushion distortion and dimness.  Also, the
voltage must drop quite far, usually below 90V, before noticeable problems
occur.  This is highly specific to each monitor: large monitors will be
more susceptable to low voltage than small monitors, good brand-name
monitors seem to be less susceptable than off-brand monitors, and so on.  

I do disagree that low voltage to this extent will leave computers
unaffected, and some brands -- especially Compaq -- may even refuse to
turn on.  In any enviroment where a UPS is commonly used, low-voltage will
result in an immediate warning long before it reaches a stage where actual
performance of either computers or monitors will be affected.

While what this electrician says is all technically accurate, I stil think
it is very unlikely that an actual case of monitor shake is caused by
building power problems, and the most likely solution is to put some
ferrite material on the cables to filter hash from the computer.

-- Mike

On 2000-05-05 at 12:06 -0400, Bill Horne wrote:

> I asked for a professional opinion on the shaking monitor thread 
> and followups.  Herewith:  the word from my favorite Master 
> Electrician. Ahem.
> Bill Horne
> -----Original Message-----
> From:	Thomas D. Horne
> Sent:	Friday, May 05, 2000 11:11 AM
> To:	Bill Horne
> Subject:	Re: FW: last on X for now
> > Very few offices have three-phase power. This is only used to 
> operate heavy machinery with
> > large motors.  It would be uncommon in an office environment, 
> although sometimes HVAC or
> > refrigeration equipment needs it.  No light fixtures ever run 
> on three-phase power, at
> > least not unless you are talking about things like stage 
> lighting.
> >
> [snip]  In multistory buildings ALL power
> is three phase: this allows for very large cost savings in 
> circuit
> layout and installation.  The power to the lighting circuits in 
> large
> office buildings in the US is not only multiphase but very often 
> a
> higher voltage than the receptacle circuits.  The reason that is 
> true is
> that the watts available to provide lighting is the product of 
> the
> ampacity of the circuit times the voltage of the source.  A 120 
> volt
> circuit breakered at twenty amps can provide a theoretical 
> maximum of
> 2400 watts.  The same circuit supplied from a 277 volt circuit 
> can
> supply 5540 watts.  One four wire, three phase, twenty amp, 
> circuit can
> supply 16620 watts of power.  This is why three phase power is 
> the rule
> rather than the exception in large office buildings.  The 
> construction
> cost savings are too great for any competent electrical engineer 
> to pass
> up.    [snip]
> > The most common cause of monitor shake is RF hash.  I would 
> doubt very
> > seriously that there is an serious electrical problem causing 
> it.
> > Sometimes a ground loop can do it, especially on an unbalanced 
> network
> > topology (coax Ethernet).
> >
> One relatively rare cause of monitor malfunction is high voltage 
> on the
> neutral of a multi phase circuit.  This can be brought on by 
> excessive
> voltage drop in the neutral caused by harmonic currents 
> generated by
> things like computer power supplies.  Since the neutral, or 
> grounded
> conductor, has no over current protection, these harmonic 
> currents can
> raise the current flow to a level higher than the circuit can 
> carry
> without excessive voltage drop.  The excessive drop can raise 
> the
> voltage on the neutral, at the outlet, to twenty or more volts. 
>  Since
> the voltage available to the monitor is the difference between 
> the phase
> conductor and the neutral conductor voltage any excessive 
> voltage drop
> on the neutral will lower the apparent voltage at the outlets 
> effected.
> This can cause erratic monitor performance without noticeably 
> effecting
> the computer's performance.
> Tom

Subcription/unsubscription/info requests: send e-mail with
"subscribe", "unsubscribe", or "info" on the first line of the
message body to discuss-request at (Subject line is ignored).

BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!

Boston Linux & Unix /