Philips and Sylvania F32T8 having different power usage

[ITSME.ULTIMATE]

Maybe the lamp specialists here could comment on my observation.

From a small number of samples, I noticed the F32T8 Philips lamps have a
higher arc voltage than electrically equivalent Sylvania lamps.

Just among the lamps I have around here, Philips lamps always cause a
higher power draw (implying higher arc voltage considering the ballast
is more or less a constant current source)

On a Sylvania/Motorola rapid start ballast a pair of Sylvania F32T8 32W
makes the input power be around 60W.
Under the same conditions, a pair of Philips Alto F32T8 makes the input
power around 64W, which I think is a fairly substantial difference when
they're using F32T8 (30W) lamps to save power.

Ballast: QTP/2X32T8RSND120V
Lamp type 1: Philips F32T8/TL841 (green endcaps)
type 2: Sylvania FO32/841/ECO (Sylvania equivalent of
environmentally friendly series)

Ambient: 20C ish, explosed lamp luminiaire
voltage: ~122v. Ballast is of auto-regulating type that maintains
constant output.



Neither is of krypton filled "ES" type.

Different ballast/lamp makes a difference, but there's a trend of
Philips lamps having a higher arc voltage than comparable Sylvania lamps
from my experience.

Is it just me or can you guys reproduce this trend too? What might cause
them to have a different arc voltage?

[Victor Roberts]

On Sun, 29 Jan 2006 20:35:40 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
Maybe the lamp specialists here could comment on my observation.
From a small number of samples, I noticed the F32T8 Philips lamps have a
higher arc voltage than electrically equivalent Sylvania lamps.
Just among the lamps I have around here, Philips lamps always cause a
higher power draw (implying higher arc voltage considering the ballast
is more or less a constant current source)
On a Sylvania/Motorola rapid start ballast a pair of Sylvania F32T8 32W
makes the input power be around 60W.
Under the same conditions, a pair of Philips Alto F32T8 makes the input
power around 64W, which I think is a fairly substantial difference when
they're using F32T8 (30W) lamps to save power.
Ballast: QTP/2X32T8RSND120V
Lamp type 1: Philips F32T8/TL841 (green endcaps)
type 2: Sylvania FO32/841/ECO (Sylvania equivalent of
environmentally friendly series)
Ambient: 20C ish, explosed lamp luminiaire
voltage: ~122v. Ballast is of auto-regulating type that maintains
constant output.
Neither is of krypton filled "ES" type.
Different ballast/lamp makes a difference, but there's a trend of
Philips lamps having a higher arc voltage than comparable Sylvania lamps
from my experience.
Is it just me or can you guys reproduce this trend too? What might cause
them to have a different arc voltage?

This is very strange. Both are rated for 32 watts, and the
rated power for both should be confirmed on the same
reference ballast. My experience is that there is usually
very little power variation between lamps when properly
measured.

Both are rated for either rapid start or instant start. The
Philips lamp is rated for 2950 initial lumens but the
Sylvania web site is not cooperating (or they will not let
me in because I don't remember my registration info right
now) so I can't get a comparative figure.

However, the only thing that should matter is that both of
these 4-foot T8 lamps are rated for the same power on the
same reference ballast.

Considering the fact that the lamps have the same diameter
and length and both should use only argon and mercury, the
only factors that can change the voltage are the argon
pressure and the mercury vapor pressure. You can't change
the argon pressure but perhaps the factory messed up. Are
you sure that both lamps are fully warmed up? How long did
they burn before you made the voltage measurement?

Perhaps Jeff has an answer to this.
--
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.

[ITSME.ULTIMATE]

In article <***@4ax.com>, ***@lighting-
research.com says...

Post by Victor Roberts
On Sun, 29 Jan 2006 20:35:40 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
Maybe the lamp specialists here could comment on my observation.
From a small number of samples, I noticed the F32T8 Philips lamps have a
higher arc voltage than electrically equivalent Sylvania lamps.
Just among the lamps I have around here, Philips lamps always cause a
higher power draw (implying higher arc voltage considering the ballast
is more or less a constant current source)
On a Sylvania/Motorola rapid start ballast a pair of Sylvania F32T8 32W
makes the input power be around 60W.
Under the same conditions, a pair of Philips Alto F32T8 makes the input
power around 64W, which I think is a fairly substantial difference when
they're using F32T8 (30W) lamps to save power.
Ballast: QTP/2X32T8RSND120V
Lamp type 1: Philips F32T8/TL841 (green endcaps)
type 2: Sylvania FO32/841/ECO (Sylvania equivalent of
environmentally friendly series)
Ambient: 20C ish, explosed lamp luminiaire
voltage: ~122v. Ballast is of auto-regulating type that maintains
constant output.
Neither is of krypton filled "ES" type.
Different ballast/lamp makes a difference, but there's a trend of
Philips lamps having a higher arc voltage than comparable Sylvania lamps
from my experience.
Is it just me or can you guys reproduce this trend too? What might cause
them to have a different arc voltage?

This is very strange. Both are rated for 32 watts, and the
rated power for both should be confirmed on the same
reference ballast. My experience is that there is usually
very little power variation between lamps when properly
measured.
Both are rated for either rapid start or instant start. The
Philips lamp is rated for 2950 initial lumens but the
Sylvania web site is not cooperating (or they will not let
me in because I don't remember my registration info right
now) so I can't get a comparative figure.
However, the only thing that should matter is that both of
these 4-foot T8 lamps are rated for the same power on the
same reference ballast.
Considering the fact that the lamps have the same diameter
and length and both should use only argon and mercury, the
only factors that can change the voltage are the argon
pressure and the mercury vapor pressure. You can't change
the argon pressure but perhaps the factory messed up. Are
you sure that both lamps are fully warmed up? How long did
they burn before you made the voltage measurement?
Perhaps Jeff has an answer to this.

My testing procedures aren't standardized, but I believe it reflects on
more of real world conditions.

Each set of lamps were warmed up for approximately 20 minutes. The
lamps are not new. They all have about a thousand hours of use. Line
voltage variation is a possibility, but as I said earlier, the ballast
is a regulating constant power using a boost conveter topology. I have
verified that varying the input voltage from 115 to 125 changes the
voltage and current inversely proportionately, but input power remained
the same(which I think is nice).

It would interseting if someone else here could compare Philips and
Sylvania lamps the same way I did and post the result.

[Jeff Waymouth]

I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?

Jeff Waymouth

Post by ITSME.ULTIMATE
research.com says...

Post by Victor Roberts
On Sun, 29 Jan 2006 20:35:40 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
Maybe the lamp specialists here could comment on my observation.
From a small number of samples, I noticed the F32T8 Philips lamps have a
higher arc voltage than electrically equivalent Sylvania lamps.
Just among the lamps I have around here, Philips lamps always cause a
higher power draw (implying higher arc voltage considering the ballast
is more or less a constant current source)
On a Sylvania/Motorola rapid start ballast a pair of Sylvania F32T8 32W
makes the input power be around 60W.
Under the same conditions, a pair of Philips Alto F32T8 makes the input
power around 64W, which I think is a fairly substantial difference when
they're using F32T8 (30W) lamps to save power.
Ballast: QTP/2X32T8RSND120V
Lamp type 1: Philips F32T8/TL841 (green endcaps)
type 2: Sylvania FO32/841/ECO (Sylvania equivalent of
environmentally friendly series)
Ambient: 20C ish, explosed lamp luminiaire
voltage: ~122v. Ballast is of auto-regulating type that maintains
constant output.
Neither is of krypton filled "ES" type.
Different ballast/lamp makes a difference, but there's a trend of
Philips lamps having a higher arc voltage than comparable Sylvania lamps
from my experience.
Is it just me or can you guys reproduce this trend too? What might cause
them to have a different arc voltage?

This is very strange. Both are rated for 32 watts, and the
rated power for both should be confirmed on the same
reference ballast. My experience is that there is usually
very little power variation between lamps when properly
measured.
Both are rated for either rapid start or instant start. The
Philips lamp is rated for 2950 initial lumens but the
Sylvania web site is not cooperating (or they will not let
me in because I don't remember my registration info right
now) so I can't get a comparative figure.
However, the only thing that should matter is that both of
these 4-foot T8 lamps are rated for the same power on the
same reference ballast.
Considering the fact that the lamps have the same diameter
and length and both should use only argon and mercury, the
only factors that can change the voltage are the argon
pressure and the mercury vapor pressure. You can't change
the argon pressure but perhaps the factory messed up. Are
you sure that both lamps are fully warmed up? How long did
they burn before you made the voltage measurement?
Perhaps Jeff has an answer to this.

My testing procedures aren't standardized, but I believe it reflects on
more of real world conditions.
Each set of lamps were warmed up for approximately 20 minutes. The
lamps are not new. They all have about a thousand hours of use. Line
voltage variation is a possibility, but as I said earlier, the ballast
is a regulating constant power using a boost conveter topology. I have
verified that varying the input voltage from 115 to 125 changes the
voltage and current inversely proportionately, but input power remained
the same(which I think is nice).
It would interseting if someone else here could compare Philips and
Sylvania lamps the same way I did and post the result.

[ITSME.ULTIMATE]

In article <***@comcast.com>, jfwaymouth3
@comcast.net says...

Post by Jeff Waymouth
I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?
Jeff Waymouth

There is a repeatable increase in system wattage of 5% or so when
Philips lamps are swapped in place of Sylvania lamps.

The testing is under a real world condition. The measurement instrument
might not be that accurate, but what counts is that going back 'n forth,
the Philips lamps consistently cause the system to draw more power.

[Victor Roberts]

On Mon, 30 Jan 2006 04:51:24 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
@comcast.net says...

Post by Jeff Waymouth
I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?
Jeff Waymouth

There is a repeatable increase in system wattage of 5% or so when
Philips lamps are swapped in place of Sylvania lamps.
The testing is under a real world condition. The measurement instrument
might not be that accurate, but what counts is that going back 'n forth,
the Philips lamps consistently cause the system to draw more power.

I'm a bit confused. If you want to understand if the power
draw is different for the two brands you have to move a bit
away from "real world." It would be nice but not necessary
to use a reference ballast. However, you do have to insure
that the line voltage, ambient temperature and air velocity
are the same when testing the two lamp types. Also, you MUST
use the same ballast, not just the same ballast type. (This
is because you are not using a reference ballast.) I believe
you are using the same ballast and are probably testing the
lamps in such short time scale that the line voltage and air
temperature are probably constant, but I just wanted to
check.
--
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.

[ITSME.ULTIMATE]

In article <***@4ax.com>, ***@lighting-
research.com says...

Post by Victor Roberts
On Mon, 30 Jan 2006 04:51:24 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
@comcast.net says...

Post by Jeff Waymouth
I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?
Jeff Waymouth

There is a repeatable increase in system wattage of 5% or so when
Philips lamps are swapped in place of Sylvania lamps.
The testing is under a real world condition. The measurement instrument
might not be that accurate, but what counts is that going back 'n forth,
the Philips lamps consistently cause the system to draw more power.

I'm a bit confused. If you want to understand if the power
draw is different for the two brands you have to move a bit
away from "real world." It would be nice but not necessary
to use a reference ballast. However, you do have to insure
that the line voltage, ambient temperature and air velocity
are the same when testing the two lamp types. Also, you MUST
use the same ballast, not just the same ballast type. (This
is because you are not using a reference ballast.) I believe
you are using the same ballast and are probably testing the
lamps in such short time scale that the line voltage and air
temperature are probably constant, but I just wanted to
check.

What good does it do for power bill reduction if it used more power in
real world application despite the same lab result?

The testing was done in an ordinary room, around 70F and the line
voltage wasn't exactly 120V, but it does not vary much overtime. Further
more, the model of ballast I used for this testing is regulated and does
not produce a change in input power from 114 to 126 volt.

The testing was done using the same ballast, as in the same ballast, not
just the same model, in the same fixture.

Both types of lamps were allowed to warm up for approx 15 min.

By the way, Philips lamps do have cathode guards and it looks like
they're the only one to use them.
http://www.nam.lighting.philips.com/us/ecatalog/fluor/pdf/p-5415.pdf

[Victor Roberts]

On Mon, 27 Feb 2006 19:24:42 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
research.com says...

Post by Victor Roberts
On Mon, 30 Jan 2006 04:51:24 -0800, ITSME.ULTIMATE

Post by ITSME.ULTIMATE
@comcast.net says...

Post by Jeff Waymouth
I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?
Jeff Waymouth

There is a repeatable increase in system wattage of 5% or so when
Philips lamps are swapped in place of Sylvania lamps.
The testing is under a real world condition. The measurement instrument
might not be that accurate, but what counts is that going back 'n forth,
the Philips lamps consistently cause the system to draw more power.

I'm a bit confused. If you want to understand if the power
draw is different for the two brands you have to move a bit
away from "real world." It would be nice but not necessary
to use a reference ballast. However, you do have to insure
that the line voltage, ambient temperature and air velocity
are the same when testing the two lamp types. Also, you MUST
use the same ballast, not just the same ballast type. (This
is because you are not using a reference ballast.) I believe
you are using the same ballast and are probably testing the
lamps in such short time scale that the line voltage and air
temperature are probably constant, but I just wanted to
check.

What good does it do for power bill reduction if it used more power in
real world application despite the same lab result?

If you are trying to find out if A uses more power than B it
is necessary to make sure that every variable other than A
and B are constant when you switch from one to the other. If
you do not do this, then you will never know if the
different power draw is due to a difference between A and B
or some other variable that changed during the swap. That
is the reason that performance testing of all types of
devices is done under controlled conditions - what some
would call laboratory conditions.

Post by ITSME.ULTIMATE
The testing was done in an ordinary room, around 70F and the line
voltage wasn't exactly 120V, but it does not vary much overtime. Further
more, the model of ballast I used for this testing is regulated and does
not produce a change in input power from 114 to 126 volt.
The testing was done using the same ballast, as in the same ballast, not
just the same model, in the same fixture.
Both types of lamps were allowed to warm up for approx 15 min.

It does seem like the other obvious variables were pretty
well controlled.

Post by ITSME.ULTIMATE
By the way, Philips lamps do have cathode guards and it looks like
they're the only one to use them.
http://www.nam.lighting.philips.com/us/ecatalog/fluor/pdf/p-5415.pdf

Well, that might be the reason why the Philips lamps draw
more power, but their print and on-line catalogs say they
draw the normal amount of power - so we really don't have a
solution.
--
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.

[Victor Roberts]

On Mon, 30 Jan 2006 06:56:37 -0500, Jeff Waymouth

Post by Jeff Waymouth
I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?

Jeff - I added the reference ballast to this discussion in
the context of the fact that both lamp types should draw 32
watts when measured on the same reference ballast in an
ambient of 25 C. That seems to imply that the two different
brands have the same electrical impedance (at least at 25
C). And so they should.

When placed on any other ballast that is not the specified
reference ballast, they should each draw the same amount of
power, though that power may no longer be 32 watts, since
they have the same impedance - at least at 25 C.

Is it possible that some design feature, perhaps related to
the low Hg content, causes these two different brands to
have different impedances when they are operate in free air
that is not at 25 C?

The OP needs to use a voltmeter to insure that the line
voltage is the same when testing each lamp type. Also, I
assume he used the same ballast, not just the same type of
ballast to confirm the power difference. And, the ballast
temperature must be the same for the two tests since the
ballast performance is a function of the ballast
temperature. After doing the above he should try to run at
least one comparison test with the lamps in still air at 25
C to see if the ballast input power is closer between the
two types.
--
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.

[Jeff Waymouth]

Vic,

To your quetion as to whethe low mercury dosing could cause different
impedence, the short, honest, answer is "I don't know". I'll do some
checking around

Jeff Waymouth

Post by Victor Roberts
On Mon, 30 Jan 2006 06:56:37 -0500, Jeff Waymouth

Post by Jeff Waymouth
I was going to ask if this was done under reference ballast conditions,
but this latest post shows that the test was not. I wonder how muchg of
this difference was natural variation?

Jeff - I added the reference ballast to this discussion in
the context of the fact that both lamp types should draw 32
watts when measured on the same reference ballast in an
ambient of 25 C. That seems to imply that the two different
brands have the same electrical impedance (at least at 25
C). And so they should.
When placed on any other ballast that is not the specified
reference ballast, they should each draw the same amount of
power, though that power may no longer be 32 watts, since
they have the same impedance - at least at 25 C.
Is it possible that some design feature, perhaps related to
the low Hg content, causes these two different brands to
have different impedances when they are operate in free air
that is not at 25 C?
The OP needs to use a voltmeter to insure that the line
voltage is the same when testing each lamp type. Also, I
assume he used the same ballast, not just the same type of
ballast to confirm the power difference. And, the ballast
temperature must be the same for the two tests since the
ballast performance is a function of the ballast
temperature. After doing the above he should try to run at
least one comparison test with the lamps in still air at 25
C to see if the ballast input power is closer between the
two types.

[Victor Roberts]

On Tue, 31 Jan 2006 06:47:50 -0500, Jeff Waymouth

Post by Jeff Waymouth
Vic,
To your quetion as to whethe low mercury dosing could cause different
impedence, the short, honest, answer is "I don't know". I'll do some
checking around

Well, not the low dose itself since we both know that it
takes less than 0.1 mg of Hg to run the lamp at any point in
time. However, perhaps a particular low dose technology is
trapping almost all of the Hg until the lamps are fully
warmed up - though I think this is unlikely. The OP would
not be seeing such a consistent power difference if the
problem was something as variable as Hg trapping.

Is it possible that Philips is selling non-compliant lamps?
I wonder if these lamps use electrode shields. If my memory
is correct, the shields raise the cathode fall a bit and
could be responsible for the higher power via a higher arc
voltage. But then the lamps would not meet ANSI specs and I
don't think that Philips would not do that.

We have Sylvania people and ex-GE people here but no one
from Philips as far as I know. We need to recruit some more
people :-) I also know we have at least one current GE
Lighting staff member who lurks but does not post.
--
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.

[Jeff Waymouth]

Philips T8 lamps do use cathode guards

Jeff Waymouth

Post by Victor Roberts
On Tue, 31 Jan 2006 06:47:50 -0500, Jeff Waymouth

Post by Jeff Waymouth
Vic,
To your quetion as to whethe low mercury dosing could cause different
impedence, the short, honest, answer is "I don't know". I'll do some
checking around

Well, not the low dose itself since we both know that it
takes less than 0.1 mg of Hg to run the lamp at any point in
time. However, perhaps a particular low dose technology is
trapping almost all of the Hg until the lamps are fully
warmed up - though I think this is unlikely. The OP would
not be seeing such a consistent power difference if the
problem was something as variable as Hg trapping.
Is it possible that Philips is selling non-compliant lamps?
I wonder if these lamps use electrode shields. If my memory
is correct, the shields raise the cathode fall a bit and
could be responsible for the higher power via a higher arc
voltage. But then the lamps would not meet ANSI specs and I
don't think that Philips would not do that.
We have Sylvania people and ex-GE people here but no one
from Philips as far as I know. We need to recruit some more
people :-) I also know we have at least one current GE
Lighting staff member who lurks but does not post.

[Victor Roberts]

On Wed, 01 Feb 2006 06:49:15 -0500, Jeff Waymouth

Post by Jeff Waymouth
Philips T8 lamps do use cathode guards

That should raise the voltage a bit. Also, I believe that
ANSI lighting conventions allow the reported power to be
rounded to the nearest watt for external publication, so two
lamps with the same power rating could differ by almost one
watt. I'll have to check the ANSI and IESNA rules for
reporting measured lamp power, if they even exist :-)
--
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.