Re: Current mid-range technology?

[nobody@xxxxxxxxxxxx (Dave Houston)]

Russ in San Diego <russgoog@xxxxxxxxx> wrote:

>On Jul 5, 6:43 pm, Neil Cherry <nche...@xxxxxxxxxxxx> wrote:
>> On Mon, 06 Jul 2009 00:21:33 GMT, Dave Houston wrote:
>> > Russ in San Diego <russg...@xxxxxxxxx> wrote:
>>
>> >>On Jul 4, 12:47 pm, nob...@xxxxxxxxxxxx (Dave Houston) wrote:
>> >>> Russ in San Diego <russg...@xxxxxxxxx> wrote:
>>
>> >>> >The reason I'm reserving judgment on XTB is that, while it ought to
>> >>> >ensure devices turn on or off when commanded, I don't see how it would
>> >>> >help with one of the worst X10 problems, that of devices switching on
>> >>> >or off when NOT commanded, i.e., after having recognized a code within
>> >>> >line noise.
>>
>> >>> Noise, per se, is seldom the reason for random switch ons. Collisions
>> >>> between two valid codes, spikes on the powerline (e.g. UPB data, fluorescent
>> >>> ballasts, motors) and brownouts (e.g. sags from motors starting) causing the
>> >>> microcontroller to reset are the most likely causes. The probability that
>> >>> random noise can create a valid Manchester code is near zero.
>>
>> >>I sure wish I could figure out what's causing my problem, then. I've
>> >>got a pair of sconce lights (CFLs) switched by an appliance module
>> >>(not a dimmer module) that, just this morning, were turning on by
>> >>themselves.  Nothing else in the house should be generating any codes,
>> >>and there are no code collisions.  No other fluorescents were turned
>> >>on in the house.
>>
>> >>There's no brownout or spikes -- I've got a photovoltaic system that
>> >>produces a very clean, very precise power waveform (necessary for a
>> >>grid-interactive inverter).
>>
>> >>There are no motors in our house that would be starting when the
>> >>lights go on -- the only motors in the house are the vacuum cleaner
>> >>(not in use) and refrigerator compressor, which is nearby, but the
>> >>activity of which is NOT correlated with this behavior.
>>
>> >>This is kind of frustrating.
>>
>> > The trickle current in X10 switches and modules (including appliance
>> > modules) can cause some CFLs to turn on. Have you tried incandescents in the
>> > same fixtures? This will tell you whether it's CFL related.
>>
>> I agree with Dave, I'm also having trouble with a CF (I thought I cut
>> the diode but I'll have to check).
>>
>> --
>> Linux Home Automation         Neil Cherry       nche...@xxxxxxxxxxxxxxx://www.linuxha.com/                        Main sitehttp://linuxha.blogspot.com/ ;                  My HA Blog
>> Author of:      Linux Smart Homes For Dummies
>
>Ah, good point.  I've been assuming that an appliance module would be
>enough to prevent these sorts of problems.
>
>You would think that, by now, someone would have modified the CFL
>design so it would be compatible -- or more likely, modified the home
>automation modules so they are compatible with CFLs!

There's no "standard" design for CFLs. Ones you buy today may not be the
same as ones you buy next month even if the PN is the same.

X10 designed their modules long before CFLs became common (but at about the
time CFLs were introduced in the late '70s). They won't offer a madification
as they do not want the liability of electrofried hackers.

>These lights are on too often to switche them to incandescents, other
>than for a brief test period. I may try it, though.

I was only suggesting a test period. It might be wise to test before
modifying the modules.

>Can anyone point me toward instructions for modifying an appliance
>module to defeat the auto-detect feature that may be causing my
>problem?

http://idobartana.com/hakb/index.htm