Re: naughty devices

["S. Hieber" <shieber at yahoo_com>]

Roger Miller said, in part:

> I just built the lights for my new tank using 6 55 watt PCs from
> ahsupply.  
> Now I have a naughty device.  I found fairly quickly that running
> these 
> lights on the same household circuit with my computer renders the
> internal 
> modem useless.  

Electronic ballasts produce harmonic distortion that feeds back onto
the house power line -- as I recall up to 10% is allowed, which is a
whopping number.  It's part of the reason that you are advised to
ground the case of the ballast -- a wire should be connected to the
ballast and to the (wire that connects to) the third hole on the wall
socket.  Other things put distortion onto the house power lines.   Ever
notice that every time you vacuum, you get those scratch dots buzzing
across your TV screen and or the radio static increases?  that's one
reason they call the distortion "noise."  Okay, maybe you don't vacuum,
but your power drill probably has the same effect on other items on the
same circuit.

Generally, this isn't a GFCI problem unless a device tries to filter
out the distortion by using a capacitor connected to ground.  The
capacitor "bleeds" the distortion to ground, which some GFCIs will pick
up as current leaking to ground, i.e., a ground fault.  Newer GFCIs are
less prone to tripping due to noise problems -- older and cheaper GFCIs
often just don't work as well.

Devices (motors, flourescent lights, and others) produce radio waves
that can be picked up as noise, static, interference by other devices. 
It could be that one of the elements on the modem is sensitive to the
RFI produced by the flourescents -- although 1) good grounding
(including the refelctor) usually keeps the RFI from being a problem
and 2) that problem probalby wouldn't change much by changing what
circuit the lights are plugged into.

> Odd.  The rest of the computer seems to be working
> fine.  

I guess that means you aren't running Windows ;-)   Seriously, though,
what's liekly hear is that the modem isn't filtering out the noise but
the other parts probably are or else aren't sensitive that particular
frequencies involved.  Distortion and RFI are wild whacky stuff.

> I 
> can plug the lights into a different circuit without problems, but
> that 
> involves snaking an extension cord to another part of the house, so
> it isn't 
> a permanent solution.

Changing what is connected to what is probably the only permanent
solution.  Certainly, the simplest.  But I'll be happy to be proved
wrong here.
> 
> There is audible static in the connection tones.  Dial tone on the
> line 
> sampled through a telephone sounds fine and there are no other
> telephone 
> disturbances that I've found.  There is a little AM radio
> interference, but 
> nothing a strong station doesn't power through.

That's probably the small amount of RFI the lights produce.

>  TV and cell phones
> are fine. 

Well, it's probably not a set of frequencies that involves them  -- The
operational freqs for both were picked for their being mathematically
far from common sources of RFI -- espeically so the cell phones.

>  I got a new surge protector with 60 db RFI protection, and that does
> not 
> help.

It probably filters out some freqs of power line distortion, does
nothing for radio waves in the air -- does not filter out the line
distortion freqs causing you a problem.  Got an oscilliscope -- you
could find out what freqs they are and make a filter to "trap" or
attentuate them.  Not necessarily easy.  Better done by an electronics
buff.

> 
> So this leads me to several questions re naughty devices.
> 
> 1)  Has anyone else had similar problems, and if so how did they
> solve it?

Change what is wired on which circuit -- sometimes, change the curcuits
themselves.  It's always nice to dedicate circuits that have
flourescents or motors because those devices tend to be the naughtiest.

> 
> 2) Any idea what may cause the ballasts to behave badly?  Ahsupply
> has not 
> (yet) responded to my question.  My installation is stock and per
> their 
> instructions except that I wired all three ballasts in parallel on a
> single 
> power cord instead of using three power cords.

It's normal electronic ballast operation, only the amount of distortion
varies from design to design, not whether there is distortion.
> 
> 3)  Given no solution to 2) is there a device that can be used to
> isolate the 
> lights  and their problems?  I've seen small plug-in devices for
> isolating 
> devices from X10-equipped circuits.  I suspect that I would need
> something 
> similar.

The right kind of filter between the lights and the house receptacle
could "trap" the distortion if it was designed for the right
frequencies.  Or the filter could be between the computer devices and
the AC power supply.  Acutally, most computer power supplies (the
internal can of stuff converts the AC 120 to lower voltages) have some
filtering built into them -- for that matter, they usually have more
surge protection than most office supply store Surge Protectors offer. 

Sorry I can't be more help,
Scott H.

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com