Maker Pro
Maker Pro

quality of incandescent lamps downhill

J

JM

Jan 1, 1970
0
The quality of standard 60/75/100 watt residential incandescent lamps from GE
and Syvlania is rather poor lately. I have been having bad luck with them
lately. I just contacted Sylvania about their 2,000 hour bulbs as not
lasting as such. They sent me some new ones. I took a peek inside some of
the new ones with my laser pen. The filaments are already loose and floppy,
and no filament supports were included. Now, shouldn't a long life bulb have
at least one filament support? Taking a look inside of some of the previous
early failures, often what's left of the filament is stretched and sagged
very unevenly. Sylvania seems to use a filament design that is not well
suited for extended service: rather long, vertical filament in respect to
base down, and no filament supports. I even visited a local store and
purchased a pack of SLI Extralife bulbs. But while I was there, I took a
quick peek at some cheapo Feit Electric and some other cheapo no name long
life bulbs. I quickly put them down as they were basically just 130v lamps
rated at 120v and hence they had rather low efficiancy. I got home and took
a peek inside of the SLI bulbs. They have a shorter length of filament
oriented horizontally in respect to base down. And yes one filament support
was included in each bulb, and all for $1.19 . I have used SLI's bulbs in
the past with good results. Even GE's normal 1,000 hour bulbs didn't do too
well when I've used them. GE uses the same filament design as sylvania.
Does anyone else have similar experiences? Perhaps is there some Sylvania
and GE workers here would like to look into this?
 
D

Don Klipstein

Jan 1, 1970
0
The quality of standard 60/75/100 watt residential incandescent lamps from GE
and Syvlania is rather poor lately. I have been having bad luck with them
lately. I just contacted Sylvania about their 2,000 hour bulbs as not
lasting as such. They sent me some new ones. I took a peek inside some of
the new ones with my laser pen. The filaments are already loose and floppy,
and no filament supports were included. Now, shouldn't a long life bulb have
at least one filament support? Taking a look inside of some of the previous
early failures, often what's left of the filament is stretched and sagged
very unevenly. Sylvania seems to use a filament design that is not well
suited for extended service: rather long, vertical filament in respect to
base down, and no filament supports. I even visited a local store and
purchased a pack of SLI Extralife bulbs. But while I was there, I took a
quick peek at some cheapo Feit Electric and some other cheapo no name long
life bulbs. I quickly put them down as they were basically just 130v lamps
rated at 120v and hence they had rather low efficiancy. I got home and took
a peek inside of the SLI bulbs. They have a shorter length of filament
oriented horizontally in respect to base down. And yes one filament support
was included in each bulb, and all for $1.19 . I have used SLI's bulbs in
the past with good results. Even GE's normal 1,000 hour bulbs didn't do too
well when I've used them. GE uses the same filament design as sylvania.
Does anyone else have similar experiences? Perhaps is there some Sylvania
and GE workers here would like to look into this?

I have noticed that "Big 3" lamps with the roughly 1 inch (2.5 cm)
coiled-coil filament lack filament supports. I expect these will mostly
give close to rated life if your line voltage is not high (measure it -
highish line voltage apears to me somewhat common nowadays) and if
vibration is low.
If you have significant vibration, you need lamps other than
incandescents with unsupported filaments.

Longlife often means mainly designing the filament to run at a slightly
lower temperature to slow down major aging mechanisms that vary greatly
with temperature. Resistance to vibration and mechanical shock is
something different from this, although in my experience most 120V
incandescents designed to have life expectancy of 3500 hours or more have
supported filaments.

I expect that most incandescents with supported filaments also have the
filament being longer, giving more heat conduction loss to the gas and
less efficacy as a result. I also expect that most incandescents with
supported filaments are designed for longer life and this requires a
compromise in efficacy.
I further expect that most incandescents with supported filaments have
multiple supports and a coiled filament as opposed to the coiled-coil
filament. The "denser" design of coiled-coil I expect to reduce radiating
ability of the tungsten, so the wire has to be slightly thicker and longer
to have same effective radiating area, and I expect such thicker wire to
operate at a slightly higher temperature for same life in comparison to
singly coiled filament wire of same wattage and voltage. As a result, I
expect this to be another reason for lower efficacy of multi-supported
filaments.
An even further loss for filament supports is the heat that they conduct
from the filament.

I do expect that a single-support coiled-coil filament has as its only
loss compared to an unsupported one being the heat conducted by the
filament support - and I guess this accounts for 20-40 lumens of light,
based on 1710 lumens being a common figure for 100 watt lamps with one
support back in the 1970's when supports were more common and 1750 lumens
being a highish-to-common figure for 100 watt lamps with no supports.

As for why lack of supports - my belief is that the reason is that
publicly traded companies are run by people who get hired and fired for
what they achieve in "The Bottom Line", all too often on an excessively
short term basis. My suspicion is that somebody had to tell the
big institutional shareholders that they could get away with eliminating
the support and the cost of putting it in, or else get replaced by someone
who will in order to keep the company having its shares no less attractive
than ones of a competitor willing to do the same.

Alternatives: You may want to use compact fluorescent lamps. They
don't work well in all situations but they may work well for you in place
of at least some of your incandescents. I discuss this more in:
http://www.misty.com/~don/cfapp.html

My compact fluorescent "top page" is http://www.misty.com/~don/cfx.html

- Don Klipstein ([email protected])
 
V

Victor Roberts

Jan 1, 1970
0
As for why lack of supports - my belief is that the reason is that
publicly traded companies are run by people who get hired and fired for
what they achieve in "The Bottom Line", all too often on an excessively
short term basis. My suspicion is that somebody had to tell the
big institutional shareholders that they could get away with eliminating
the support and the cost of putting it in, or else get replaced by someone
who will in order to keep the company having its shares no less attractive
than ones of a competitor willing to do the same.

I hate to blow your conspiracy theory, but I believe the reason that
most ordinary incandescent lamps are built without filament supports
is that the supports reduce efficacy.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
V

Victor Roberts

Jan 1, 1970
0
| On Thu, 19 May 2005 03:16:09 +0000 (UTC), [email protected] (Don
| Klipstein) wrote:
|
|> As for why lack of supports - my belief is that the reason is that
|>publicly traded companies are run by people who get hired and fired for
|>what they achieve in "The Bottom Line", all too often on an excessively
|>short term basis. My suspicion is that somebody had to tell the
|>big institutional shareholders that they could get away with eliminating
|>the support and the cost of putting it in, or else get replaced by someone
|>who will in order to keep the company having its shares no less attractive
|>than ones of a competitor willing to do the same.
|
| I hate to blow your conspiracy theory, but I believe the reason that
| most ordinary incandescent lamps are built without filament supports
| is that the supports reduce efficacy.

He did say to that effect that this does happen. But they are doing things
that do reduce the lifetime of bulbs marketed as long life bulbs.

Are you saying that you have run life tests under standard conditions
with statistically a significant number of samples to justify your
claims? If not, do you have access to other data that supports your
claims? Have you ever seen the life test facilities maintained by the
three major lamp companies?
They are
being dishonest about the product being sold, and in effect are not giving
consumers a choice between product types.

In fact, in the US the Federal Trade Commission (FTC) does a pretty
good job insuring that the lamps meet the published performance
numbers. Also, are you saying that someone is holding a gun to your
head and making you purchase products produced by these companies?

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
D

Daniel J. Stern

Jan 1, 1970
0
The quality of standard 60/75/100 watt residential incandescent lamps
from GE and Syvlania is rather poor lately.

Here we go again...
 
V

Victor Roberts

Jan 1, 1970
0
On 19 May 2005 22:53:41 GMT, [email protected] wrote:

[snip]
|>| I hate to blow your conspiracy theory, but I believe the reason that
|>| most ordinary incandescent lamps are built without filament supports
|>| is that the supports reduce efficacy.
|>
|>He did say to that effect that this does happen. But they are doing things
|>that do reduce the lifetime of bulbs marketed as long life bulbs.
|
| Are you saying that you have run life tests under standard conditions
| with statistically a significant number of samples to justify your
| claims? If not, do you have access to other data that supports your
| claims? Have you ever seen the life test facilities maintained by the
| three major lamp companies?

I'm saying I don't need to. Too many bulbs were being replaced too
often. I'm not talking about 2000 hour bulbs averaging 1800 hours of
life; I'm talking about 2000 hour bulbs averaging 300 hours of life.
It's dramatically noticed in many cases; I don't need no steeking test
lab.

If 2000-hour lamps were operating for 300 hours I would think that
there would be far more people complaining than you. I don't but any
long life lamps except for the 3000-hour Philips Halogina, and I have
no complaints about their life. Most of my incandescent sockets use
normal life, that is 1000-hour or 750-hour, depending upon the
wattage, incandescent lamps and they last long enough that I can't
remember when I put the lamp in.

If you are relating your real experience instead of just blowing off
steam, then I suggest you measure your line voltage. As you know, lamp
are rated at 120 volts and even a small increase in line voltage can
have a dramatic effect on lamp life.
|>They are
|>being dishonest about the product being sold, and in effect are not giving
|>consumers a choice between product types.
|
| In fact, in the US the Federal Trade Commission (FTC) does a pretty
| good job insuring that the lamps meet the published performance
| numbers. Also, are you saying that someone is holding a gun to your
| head and making you purchase products produced by these companies?

They obviously are not buying them in the stores I buy them in, and
testing them in real homes, or else they are like the FDA and on the
take from those who they are supposed to be protecting us from.

I don't know where they buy their lamps, but I do know lamps are not
tested in "real homes." Like most products they are tested under
standard conditions which are published and can be replicated in any
test facility. The most important of these conditions for standard
incandescent lamps in the US is that the line voltage is held at 120
volts.

While I'm sure that the FDA has made some mistakes, I seriously doubt
the FDA and the FTC are "on the take" as you so delicately put it. I
know of two specific instances when the FTC came down on lamp
manufacturers for what they, the FTC, believed was misrepresentation
of light output data.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
B

Bob

Jan 1, 1970
0
The comments below about the 120v line voltage are now invalid in many parts
of the country, and may explain why many people are griping about somewhat
shorter lamp life. Many utilities - the one I work for included - have a
distribution voltage regulation target of 125.5 volts.

This http://n2qew.dyndns.org/da/da.PNG is a screenshot from one of my
substations. That S in a circle is a regulator. The voltages before and
after it are what's expected at the secondary of a 13,200 to 120/240
distribution transformer. As you can see, the regulator's target is 126
volts. This is done for efficiency - our equipment, as well as the
customer's. It has no negative effects on electrical, or electronic
equipment as the ANSI standard is 120 volts +/- 10% which means you could
see from 108 to 132 volts at your outlet, and the utility is within specs.
We try to run + 5% which keeps the saturation high in motors (air
conditioners, etc) and transformers. This reduces the current drawn
dramatically, which reduces line losses. You pay by the watt-hour, and end
up paying the same, or perhaps a little less because there's less power lost
to heat in reactive appliances - most things in your house. The thing that
suffers though is using 120 volt bulbs where you've consistently got 125 -
126 volts. I suppose that's why there is 130 volt bulbs available.

BTW: The currents in amps that you see in the screen shot are at 13,200
volts.
 
V

Victor Roberts

Jan 1, 1970
0
The comments below about the 120v line voltage are now invalid in many parts
of the country,

My comments refer to standard test conditions and are therefore NOT
invalid since the standard test conditions remain 120 volts.


Because it makes it hard to follow a discussion.
Why is top posting a bad idea?

Vic Roberts
 
C

Clive Mitchell

Jan 1, 1970
0
Victor Roberts said:
If you are relating your real experience instead of just blowing off
steam, then I suggest you measure your line voltage. As you know, lamp
are rated at 120 volts and even a small increase in line voltage can
have a dramatic effect on lamp life.

Check if they are being exposed to high vibration or bumps too. And
lamp orientation can affect life as well.
 
B

Bob

Jan 1, 1970
0
Victor Roberts said:
My comments refer to standard test conditions and are therefore NOT
invalid since the standard test conditions remain 120 volts.


Because it makes it hard to follow a discussion.
Why is top posting a bad idea?

Vic Roberts
OK, I can post to your newsgroup your way.

Anyway, I need to explain further. I agree that testing 120 volt rated light
bulbs at 120 volts is a perfectly valid way to test the average life when
the bulb is run on 120 volts. However, as many people here have noticed,
when you take a 120 volt bulb, and power it with a higher voltage, the live
of that bulb, on average, will be shorter.

My comments were about what people should expect to see as line voltage, not
anything related to standardized test conditions for light bulbs.

What does remain is that people need to determine what their line voltage is
in their are - it should be relatively stable - and decide whether they want
little more light than rated from a bulb, with shorter life, or getting a
bulb that matches their line voltage.
 
V

Victor Roberts

Jan 1, 1970
0
| On Thu, 19 May 2005 23:30:42 -0400, "Bob" <[email protected]>
| wrote:
|
|>The comments below about the 120v line voltage are now invalid in many parts
|>of the country,
|
| My comments refer to standard test conditions and are therefore NOT
| invalid since the standard test conditions remain 120 volts.

Since the actual voltage standard is not exactly 120 volts, that cannot
be a valid test condition. As I showed in my previous post, merely
deviating the voltage even while keeping a median at 120 volts
completely changes the dynamics as they affect overall sales. Single
voltage testing simply does not give an accurate portrayal of reality
even if it is at the median voltage, which as Bob points out before,
could very well be above the midpoint of the standard.

Well, even "Bob" admits that 120 volts is still the center of the
range. So 120 volts IS still the mean, with an allowed variation
around the mean. In fact at my house, line voltage is very close to
121 volts and not 125 volts.

But, more importantly, a "standard test" is just that - a test that is
conducted under published and standardized procedures so it can be
replicated in any laboratory and the results will be the same - or as
much the same as is possible given the Uncertainty Principle and other
such things :) Since one ANSI standard line voltage in the US is 120
volts +/- 10% it seems reasonable to me to run the standard test at
120 volts. The data showing how lamp performance will vary is
available. Perhaps we need to print the life, power and lumen curves
as a function of line voltage on each lamp box :) but how many people
know what the line voltage at their house really is? (The data given
by Bob may apply to the calculated voltage at the step-down
transformer on the pole and may be higher then 120 volts to account
for losses between the pole and the house.)

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
B

Bob

Jan 1, 1970
0
Victor Roberts said:
On 20 May 2005 11:46:26 GMT, [email protected] wrote:

Well, even "Bob" admits that 120 volts is still the center of the
range. So 120 volts IS still the mean, with an allowed variation
around the mean. In fact at my house, line voltage is very close to
121 volts and not 125 volts.

Chill out. This is not the "battle of the bulbs". I don't know where you get
"even "Bob" admits that 120 volts..." when I don't recall denying anything
here. And, yes, Bob is my name, so there's no pseudonym there either. I
don't recall saying anything negative about the validity of lamp testing a
120 volt lamp at 120 volts. The problem is that these lamps, while tested at
120 volts produce x lumens, and last y hours, do not perform the same when
placed in service and operated at a different voltage than what is specified
on the packaging. The consumer (home lighting engineer) is mis-applying the
product. I would hazard a guess that a 120 volt lamp operated at +10% or 132
volts is going to have a greater than 10% loss of life, while a lamp
operated at -10% or 108 volts will have more than a 10% increase in life.
What it really boils down to is that if someone is experiencing shorter than
rated lamp life consistently, then maybe they should figure out what their
mean voltage is, and buy lamps rated accordingly. I have seen 120 volt, 125
volt, and 130 volt lamps.
 
V

Victor Roberts

Jan 1, 1970
0
Chill out. This is not the "battle of the bulbs". I don't know where you get
"even "Bob" admits that 120 volts..." when I don't recall denying anything
here. And, yes, Bob is my name, so there's no pseudonym there either. I
don't recall saying anything negative about the validity of lamp testing a
120 volt lamp at 120 volts. The problem is that these lamps, while tested at
120 volts produce x lumens, and last y hours, do not perform the same when
placed in service and operated at a different voltage than what is specified
on the packaging. The consumer (home lighting engineer) is mis-applying the
product. I would hazard a guess that a 120 volt lamp operated at +10% or 132
volts is going to have a greater than 10% loss of life,

Operating a standard 120-volt incandescent lamp at 132 volts would
result in about 29% of rated life, that is a loss of 71% of rated
life.
while a lamp
operated at -10% or 108 volts will have more than a 10% increase in life.

When operated at 108 volts, that same lamp would last for about 3.9
times the rated life.
What it really boils down to is that if someone is experiencing shorter than
rated lamp life consistently, then maybe they should figure out what their
mean voltage is, and buy lamps rated accordingly. I have seen 120 volt, 125
volt, and 130 volt lamps.

125-volt lamps may be hard to find, but 130-volt versions are made by
all the major lamp manufacturers.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
V

Victor Roberts

Jan 1, 1970
0
On 20 May 2005 11:34:51 GMT, [email protected] wrote:

[snip]
Let's take a look at the economics of this. The bulbs could be made to
last 5 times longer than they do now. But that would only mean 1/5 of
the sales. By having them last 1/5 of the time, sales could potentially
be 5 times as much. A few people will switch brands, but mostly
everyone is doing it, so you're going to gain back from the other
brands what you lost anyway. A few people might find some decent
products or maybe just decide to live with less light. That might
shave a bit off the sales, but even still, bulbs lasting 1/5 the time
generate at least 4 times the sales due to relamping.

You can already buy long life lamps. The standard life 1000-hour and
750-hour lamps actually save you money since they are more efficient
than longer life versions and the electricity used to operate lamps
costs more then the lamps themselves.
| If you are relating your real experience instead of just blowing off
| steam, then I suggest you measure your line voltage. As you know, lamp
| are rated at 120 volts and even a small increase in line voltage can
| have a dramatic effect on lamp life.

Last time I had a really serious problem with multitudes of bulbs
rapidly burning out (about 5 of them lasting less than one hour) I did
check the voltage: 120.3 volts

What kind of lamps are your purchasing that last for less than one
hour?

[snip]
These laboratory conditions will not show the problems. Let's assume
for a moment than the median voltage in US homes is 120 volts. That
means for every home in which the voltage is low there is another home
in which the voltage is high. Now what does that do to the lamp life?

These curves are available and have been discussed here a number of
times.
One home might have bulbs last 5 times longer while the other has bulbs
last 1/5 of the time. If everything were equal, let's say these
homeowners are buying 10 replacement bulbs per year. But with these
voltage variations, one of them is buying only 2 replacement bulbs per
year, while the other is buying 50. What would be 20 sales per year
for these 2 homes is now 52 sales per year.

This is just physics of incandescent lamps - I don't know how we
change that without the installation of expensive voltage regulators.
Or perhaps move to CFLs with universal input voltage ballasts.
So you can see that even with the median voltage held to 120 volts, a
disparity of voltage is going to increase the sales of replacement
bulbs even though the median is held.

The line voltage is controlled by the power companies, and to the best
of my knowledge, they have no financial interest in selling lamps. The
lamp companies in turn have no control over the line voltage in your
home.
So these labs that test bulbs at exactly 120 volts are NOT doing a
realistic test at all. What they SHOULD be doing is running multiple
parallel tests at the various lesser and greater voltages that actually
exist in homes. And further, these tests should include real life
factors such as the occaisional surge and sag that commonly do exist.

These tests have been run and the data published. Operation at
non-standard conditions can be related to tests at standard
conditions. So, testing lamps under standard conditions insures that
the quality is being met and all the other data you are asking for can
be extrapolated, except that you will need to define "occasional
surge" so this is an event that can be replicated.
Maybe we should start by finding out just how much deviation in voltage
actually exists. I've heard reports on some DIY forums of voltages in
various parts of the US as low as 108 and as high as 130.

The line voltage is 120 volts +/- 10% which means 108 volts to 132
volts.
So at least
one test bank in the lab should be running those voltages. What we
need to know is just what portion of homes have each of the various
voltage levels (1 volt steps is probably sufficient). Then the labs
can set up appropriately balanced testing and compare the relamping
rates of the overall test against a control group of the perfect and
surge/sag-free 120 volt test.

As stated above, this data is already available.
| While I'm sure that the FDA has made some mistakes, I seriously doubt
| the FDA and the FTC are "on the take" as you so delicately put it. I
| know of two specific instances when the FTC came down on lamp
| manufacturers for what they, the FTC, believed was misrepresentation
| of light output data.

Track the number of people that leave government jobs in these
regulatory agencies and move on to jobs in the very industries they
just regulated before, and you'll see a whole different picture. The
FDA is perhaps the worst. The FTC isn't quite as bad, but their name
has come up a few times in the past.

I don't know of a single person who moved from the FTC to a job with a
lamp manufacturer. Can anyone provide a name? The entire lamp industry
probably produces less profit than one drug.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
J

JohnR66

Jan 1, 1970
0
JM said:
The quality of standard 60/75/100 watt residential incandescent lamps from
GE
and Syvlania is rather poor lately. I have been having bad luck with them
lately. I just contacted Sylvania about their 2,000 hour bulbs as not
lasting as such. They sent me some new ones. I took a peek inside some
of
the new ones with my laser pen. The filaments are already loose and
floppy,
and no filament supports were included. Now, shouldn't a long life bulb
have
at least one filament support? Taking a look inside of some of the
previous
early failures, often what's left of the filament is stretched and sagged
very unevenly. Sylvania seems to use a filament design that is not well
suited for extended service: rather long, vertical filament in respect to
base down, and no filament supports. I even visited a local store and
purchased a pack of SLI Extralife bulbs. But while I was there, I took a
quick peek at some cheapo Feit Electric and some other cheapo no name long
life bulbs. I quickly put them down as they were basically just 130v
lamps
rated at 120v and hence they had rather low efficiancy. I got home and
took
a peek inside of the SLI bulbs. They have a shorter length of filament
oriented horizontally in respect to base down. And yes one filament
support
was included in each bulb, and all for $1.19 . I have used SLI's bulbs in
the past with good results. Even GE's normal 1,000 hour bulbs didn't do
too
well when I've used them. GE uses the same filament design as sylvania.
Does anyone else have similar experiences? Perhaps is there some Sylvania
and GE workers here would like to look into this?
I've tried the GE Reveal bulbs. They lasted 200 hours. From my other post:

In my case, it was bulb quality. I bought 2 four
packs of GE Reveal bulbs. One pack was 60W and one was 40W. They were
installed in two separate fixtures (4 bulbs per fixture) the same time. In
6-8 weeks 7 of the 8 bulbs have blown and the last one hung on for another
three weeks. The bulk blew with around 200 hours of use. Far short of
expected. The bulbs I have used prior to and after the GE ones have a normal
life span. It is safe to say the GE Reveal bulbs are JUNK. The filament
design is probably the same for standard GE bulbs. There are no vibration
issues and my line voltage is 116-122 depending on the season.

The Sylvania bulbs (long life) seem to last as rated, but I have not not
purchased any in a while as I tend to stock up. I bought the Reveal bulbs to
see how the light output was.
John
 
B

Bob

Jan 1, 1970
0
| On 20 May 2005 11:34:51 GMT, [email protected] wrote:
|
| [snip]
I'm going to be switching to low voltage lighting, anyway.

Does a 10% increase in voltage have the same effect on a low voltage bulb as
a line voltage bulb?
Maybe we need to push for tighter voltage regulation?
I can tell you that you would wind up paying ALOT more for that in the form
of electric bills than you would save in light bulbs. Regulators are LARGE
lossy mechanical monsters. Voltage is regulated on the primary side in
typically .5% steps relative to 120/240 volts. But remember, we, and other
utilities try to run +5% for efficiency.
This http://n2qew.dyndns.org/da/DSCF1231.JPG is a picture of 3 single phase
400 kva regulators. It (they) control the voltage on a single feeder leaving
the sub.
This http://n2qew.dyndns.org/da/DSCF1215.JPG is a closer picture. The dial
shows the current state (boost or buck), and the range of extremes. These
are computer controlled, and can regulate pretty closely AT THE SUBSTATION.
At the voltages we run, load on the line really isn't a big factor. 1 amp is
equal to 13,200 watts. A 10 percent voltage change would be1,320 volts at
primary voltage (13,200 volts). If you were drawing 200 amps from each both
legs feeding your house 48,000 watts (not likely, by the way), the feeder to
your house only needs to provide 3.6 amps. Not much voltage drop for many
miles. Consequently, substation regulaition works pretty well - if feeders
are properly engineered, and monitored - ours are.
http://ka9wgn.ham.org/ |
--------------------------------------------------------------------------
---
 
V

Victor Roberts

Jan 1, 1970
0
| Operating a standard 120-volt incandescent lamp at 132 volts would
| result in about 29% of rated life, that is a loss of 71% of rated
| life.

Is that specified on the package in as clear a lettering as the number
of hours for 120 volts?

Not on the boxes I just checked, but it is printed directly on the
lamp, which some might consider even better.
|>while a lamp
|>operated at -10% or 108 volts will have more than a 10% increase in life.
|
| When operated at 108 volts, that same lamp would last for about 3.9
| times the rated life.

I'm glad you have some statistics. You did the tests already?

The data I gave above comes from standard published equations.
However, when I taught at the LRC my class ran a test of eleven
1000-hour incandescent lamps all operated at the voltage that should
reduce their life to 100 hours. (We didn't have time to run the test
to 3000 hours.) All the lamps failed between 74 and 132 hours. The
mean life was 103 hours and the standard deviation was 16.42.
|>What it really boils down to is that if someone is experiencing shorter than
|>rated lamp life consistently, then maybe they should figure out what their
|>mean voltage is, and buy lamps rated accordingly. I have seen 120 volt, 125
|>volt, and 130 volt lamps.
|
| 125-volt lamps may be hard to find, but 130-volt versions are made by
| all the major lamp manufacturers.

Do they make them in every variety of bulb type as they make the 120 volt
versions?

No.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
V

Victor Roberts

Jan 1, 1970
0
| What it really boils down to is that if someone is experiencing shorter than
| rated lamp life consistently, then maybe they should figure out what their
| mean voltage is, and buy lamps rated accordingly. I have seen 120 volt, 125
| volt, and 130 volt lamps.

The NNNN hours of lifetime rating should not be used in marketing because
the majority of homes don't have the exact conditions where they will get
that lifetime. A table of hours for various voltages is what should be
given. Or else manufactures will need to put the voltage in large letters
on the bulb packaging, and stock a variety of voltage choices at retail
locations. Maybe you can look for bulbs in the 127 volt aisle.

In the best of circumstances the published performance of all products
sold is based on standardized tests. You can then compare products
from different manufacturers even if the performance is not exactly
what you will get in your own application. When there is no
standardized test the manufacturers just make up their own tests,
which leads make it impossible to compare products from different
manufacturers.

I agree that consumers don't understand how variations in line voltage
will change the operating life of their lamps. I have no objection to
publishing the life vs. voltage curves, but how many consumers know
the line voltage in their homes? And - as I stated in the previous
reply, the lamp voltage is listed on the lamp.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
V

Victor Roberts

Jan 1, 1970
0
| On 20 May 2005 11:46:26 GMT, [email protected] wrote:
|
|>| On Thu, 19 May 2005 23:30:42 -0400, "Bob" <[email protected]>
|>| wrote:
|>|
|>|>The comments below about the 120v line voltage are now invalid in many parts
|>|>of the country,
|>|
|>| My comments refer to standard test conditions and are therefore NOT
|>| invalid since the standard test conditions remain 120 volts.
|>
|>Since the actual voltage standard is not exactly 120 volts, that cannot
|>be a valid test condition. As I showed in my previous post, merely
|>deviating the voltage even while keeping a median at 120 volts
|>completely changes the dynamics as they affect overall sales. Single
|>voltage testing simply does not give an accurate portrayal of reality
|>even if it is at the median voltage, which as Bob points out before,
|>could very well be above the midpoint of the standard.
|
| Well, even "Bob" admits that 120 volts is still the center of the
| range. So 120 volts IS still the mean, with an allowed variation
| around the mean. In fact at my house, line voltage is very close to
| 121 volts and not 125 volts.

I never said 120 volts wasn't the center. My point is that real life
is not precisely at 120 volts everywhere.


| But, more importantly, a "standard test" is just that - a test that is
| conducted under published and standardized procedures so it can be
| replicated in any laboratory and the results will be the same - or as
| much the same as is possible given the Uncertainty Principle and other
| such things :) Since one ANSI standard line voltage in the US is 120
| volts +/- 10% it seems reasonable to me to run the standard test at
| 120 volts. The data showing how lamp performance will vary is
| available. Perhaps we need to print the life, power and lumen curves
| as a function of line voltage on each lamp box :) but how many people
| know what the line voltage at their house really is? (The data given
| by Bob may apply to the calculated voltage at the step-down
| transformer on the pole and may be higher then 120 volts to account
| for losses between the pole and the house.)

I don't buy light bulbs to be able to replicate tests. Replicating tests
doesn't help me. If someone challenges the test you did, of course you
can say "The test was at 120 volts, so you can test it yourself at 120
volts to verify it".

If the standard is 120 volts +/- 10% then it is MORE reasonable to me to
run the test many different voltages in the range. By doing a test only
at 120 volts, you are ignoring the 10% part entirely. If that were 5% or
20% it would change the number of light bulb failures per year (there
would be a greater number of failures as the percentage swing increases).
You're simply not testing reality when you test at a single voltage.

Tests should be run at 110 volts, 114 volts, 117 volts, 119 volts, 120
volts, 121 volts, 123 volts, 126 volts, and 130 volts. Test enough in
each to make statistics reliable. Then weight the results based on a
study of just what portion of homes actually has what voltage. If it
turns out the median voltage actually being served is different than
120 volts, you might need to change the set of voltages tested.

Then a second round of tests should vary the voltage on each light bulb
accounting for day and seasonable changes in service voltage. Further
tests would apply special conditions like surges, sags, and vibration
at a number of different frequencies and G-force.

There just isn't much value in testing at a single voltage for other
than watching and detecting certain changes in manufacturing patterns.
A decrease in lifetime can indicate a problem in manufacturing. But an
average 3000 hour life in the lab test at a single voltage just does not
translate to an average 3000 hour life in the home.

The tests at voltages from 108 to 132 have been run and the results
have been used to create a set of equations that show how lamp life,
power, output, and color temperature change with operating voltage.
These changes are fixed by the physics of the lamp and there is no
reason to run these tests on every batch of lamps.

Tests are run at standard conditions to insure that the production run
from which the sample tested was taken is normal and does not have a
manufacturing defect. Once the life at nominal voltage is assured, the
life at other voltage is fixed by physics and does not have to be
validated on every batch.

I hope you also understand that a life test is destructive. Life tests
are run on samples of the production run, and the rest of that
production run is then sold to consumers. The lamp that _you_
purchased has not been life tested, because that is impossible - it
would destroy the lamp. So you have to have some faith that the lamp
you purchased is similar to the lamps that were tested.

You propose that the lamp manufactures determine the median line
voltage, which would be a daunting task, as it varies with power
company, distance from the substation, distance from the pole
transformer and many other variables. Every appliance in your home is
labeled 120 volts. I do not believe it is unreasonable for the lamp
companies to measure lamps at 120 volts, which is supposed to be the
center point of the voltage supplied to homes. In any case, I don't
think you would be satisfied with the median line voltage - only with
tests run at the voltage in _your_ house, which I believe you said was
very close to 120 volts.

If you continue to disagree with me, you are free to make your case to
the Federal Trade Commission.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
V

Victor Roberts

Jan 1, 1970
0
On 20 May 2005 19:10:26 GMT, [email protected] wrote:

[snip]
|>These laboratory conditions will not show the problems. Let's assume
|>for a moment than the median voltage in US homes is 120 volts. That
|>means for every home in which the voltage is low there is another home
|>in which the voltage is high. Now what does that do to the lamp life?
|
| These curves are available and have been discussed here a number of
| times.

They need to be put on product packaging. Otherwise the stated lifetime
is a form of false advertising. It's really a marketing issue.

I disagree that this is false advertising since the published
performance is based on standardized, published test protocols that
are accepted by all the lamp companies, are well known to the
government and are run at the official mean line voltage for
residences in the US.

I do agree that perhaps more consumer education is warranted on the
effect of line voltage on lamp life - but as I have stated, I suspect
that only the most technically sophisticated consumers have any idea
what the line voltage is in their homes.

[snip]
|>So you can see that even with the median voltage held to 120 volts, a
|>disparity of voltage is going to increase the sales of replacement
|>bulbs even though the median is held.
|
| The line voltage is controlled by the power companies, and to the best
| of my knowledge, they have no financial interest in selling lamps. The
| lamp companies in turn have no control over the line voltage in your
| home.

But they can at least know what the reality is and be honest about it
by marking the correct RANGE of lifetime on the bulb packaging.

Perhaps you should take this up with your power company and stop
dinging the lamp companies.
|>So these labs that test bulbs at exactly 120 volts are NOT doing a
|>realistic test at all. What they SHOULD be doing is running multiple
|>parallel tests at the various lesser and greater voltages that actually
|>exist in homes. And further, these tests should include real life
|>factors such as the occaisional surge and sag that commonly do exist.
|
| These tests have been run and the data published. Operation at
| non-standard conditions can be related to tests at standard
| conditions. So, testing lamps under standard conditions insures that
| the quality is being met and all the other data you are asking for can
| be extrapolated, except that you will need to define "occasional
| surge" so this is an event that can be replicated.

When the design changes, like using fewer supports, the full range of
test needs to be run again.

They are. The tests at 120 volts will detect any changes due to
design. The variations at other voltages are mostly determined by
physics and do not have to be done on every lamp design.
|>Maybe we should start by finding out just how much deviation in voltage
|>actually exists. I've heard reports on some DIY forums of voltages in
|>various parts of the US as low as 108 and as high as 130.
|
| The line voltage is 120 volts +/- 10% which means 108 volts to 132
| volts.

Right.

Maybe we need to push for tighter voltage regulation?
Perhaps.



|>So at least
|>one test bank in the lab should be running those voltages. What we
|>need to know is just what portion of homes have each of the various
|>voltage levels (1 volt steps is probably sufficient). Then the labs
|>can set up appropriately balanced testing and compare the relamping
|>rates of the overall test against a control group of the perfect and
|>surge/sag-free 120 volt test.
|
| As stated above, this data is already available.

Not on the bulb packaging.

|>| While I'm sure that the FDA has made some mistakes, I seriously doubt
|>| the FDA and the FTC are "on the take" as you so delicately put it. I
|>| know of two specific instances when the FTC came down on lamp
|>| manufacturers for what they, the FTC, believed was misrepresentation
|>| of light output data.
|>
|>Track the number of people that leave government jobs in these
|>regulatory agencies and move on to jobs in the very industries they
|>just regulated before, and you'll see a whole different picture. The
|>FDA is perhaps the worst. The FTC isn't quite as bad, but their name
|>has come up a few times in the past.
|
| I don't know of a single person who moved from the FTC to a job with a
| lamp manufacturer. Can anyone provide a name? The entire lamp industry
| probably produces less profit than one drug.

So no one at GE used to work for the FTC? Or are you just counting those
in the lamp division?

I only said I don't know of anyone from the lamp companies who worked
at the FTC. I obviously can't speak for all of GE or even all of the
lamp division. I do, however, know a person who worked at the FTC. He
is not an engineer and never worked for GE.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
 
Top