Re: The insane progress nobody is talking about

On Mon, 1 Jul 2024 16:04:09 +0100, Robert Carnegie
<rja.carnegie@gmail.com> wrote:

On 25/06/2024 16:34, Paul S Person wrote:

On Tue, 25 Jun 2024 08:56:17 -0500, "Michael F. Stemper"
<michael.stemper@gmail.com> wrote:
 

On 20/06/2024 11.21, Paul S Person wrote:

On Thu, 20 Jun 2024 02:59:25 GMT, Random <random@who.cares> wrote:

>

A conventional bulb's filament is not sensitive to AC voltage fluctionations,
where the conversion electronics is. My guess is that your wiring to that
light is causing voltage dips and is stressing the electronics in the LED bulb
base.

>
I've seen a similar claim about timers that actually count
cycles-per-second: if those vary then the timer misperforms.

>
This isn't really an issue in North America. According to NERC[1], frequency in
the Eastern Interconnection (EI) is above 59.972 Hz 95% of the time. This is
99.95% of nominal frequency, or an error of 28 mHz.
>
If frequency sat that low for a 24 hour period (which it doesn't; frequency error
regularly crosses zero), it would be a loss of 40 seconds in a day.
>
As a matter of fact, a few years back, the EI had an ongoing problem with frequency
being high: 3 mHz fast as a sustained average over several years. This was considered
a significant enough issue to require an investigation[2].
>
If uncorrected, it would have caused clocks to gain over 4 seconds per day. This
led to regularly implementing Time Error Correction. In this case, that meant a
coordinated (across the EI) reduction in generation to reduce frequency until the
time error crossed zero again.
>
Four seconds error per day is considered a problem. If you live in North America,
you can count[3] on your analog clock.

 
Three things:
 
-- I was talking about a timer, not a clock. In particular, a timer
controlling our porch light, when went wonky after about 10 years. It
could, of course, have simply aged out, or my switch to CFLs may have
been responsible, who can say? And the timer was put in in the 90s.
Still works fine as manual switch, though.
 
-- My memory is that our City Light was reducing the CPS deliberately
-- or allowing it to fluctuate as it chose -- for some reason.
 
-- Seattle is not in the Eastern Interconnection. Not that I think
whatever Interconnection it /is/ in does any different than the EI.
 
Summary:
-- wrong device
-- possible misapplication of modern conditions to the past
-- wrong part of country

>
In the 1990s, I'd expect your device to have
its own microprocessor clock,,but I suppose
you'd know.

This is an Intermatic Programmable Wall Switch EI341. It looks a lot
like this:
<https://www.ebay.com/itm/116149597163?chn=ps&_trkparms=ispr%3D1&amdata=enc%3A1I9uzWWJSRfWjHKDBgPhM7Q98&norover=1&mkevt=1&mkrid=711-213727-13078-0&mkcid=2&itemid=116149597163&targetid=4580634177211616&device=c&mktype=&googleloc=&poi=&campaignid=604202271&mkgroupid=1232553588462791&rlsatarget=pla-4580634177211616&abcId=9427713&merchantid=51291&msclkid=c8574f2cf4701b2941c129ae2126ea4c>
but has only one ON/OFF cycle.

To program it, you set the start time, move the slider to ON, and push
the button. This should cause the indicator to start blinking. You
then set the end time and push the button. The indicator is now a
solid red, indicating that the switch is "programmed".

The device pictured looks very much like my model, but it can store 24
ON/OFF pairs, while mine can only handle one. The documentation for
mine says nothing about a microprocessor clock.

It was sold to us and installed by an electrician (well, a company in
the business of providing electricians to homeowners to perform a
specific task). There is no telling when it was manufactured.

When the timer function is OFF, the push button works fine as a wall
switch. Well, except that you have to look outside to see if the light
is actually on or off, depending on what you wanted.

I should note that our even older furnace thermostat uses a 24-hour
clockface and little blue and red pins for programming. Mechanical
timers work quite well.

I think I heard that fluorescent and CFL lamps
do create a sort of anti-signal that confuses
an electricity meter - this may be not true at
all, but I can see it upsetting a clock, as well.
You could try LED.

It is true that the documentation requires a "Class P Protected
Ballast" for any fluorescent lamp controlled by it. As I noted, this
may or may not have been the cause of the problem.

I think I've got one CFL, bathroom, that won't
die.  It's slightly annoying, as it's the type
that takes a while to be bright enough to read by.

I've had tubes do that, when they get old and slow. The CFL on the
front porch can take a while to get started when it is cold out but
it's not actually intended to be used outside.
--
"Here lies the Tuscan poet Aretino,
Who evil spoke of everyone but God,
Giving as his excuse, 'I never knew him.'"