getting irradiance from source defined inphotometric units

Someone please speak up if my understanding is flawed, but ...

I would e.g. assume an efficacy of 100lm/W for my source, and if

ies2rad

would use 179 instead, the radiance values in my simulation would be

by

far too low...

If you're using IES files via ies2rad, the IES files are measured in
candela, and the physically measured values already account for
efficiency of the source from electrical power in - through to the
lighting power out. The factor of 179 is just a conversion from W/m2 to
lumens/m2, as a unit conversion, not an efficiency factor. The source
efficiency is already incorporated into the physical measurements
documented in the IES file.

I used white, incandescent, and fluorescent, and spread 200 lumens

over

a 1 meter polygon, and ended up with 63.7, 60.5, and 54.1 cd/m^2
respectively.

I had understood the factors lamp.tab are mainly intended to account for
lumen depreciation, and just happen to have some color offset included
as part of this. In Rob's tests the decreasing values roughly correspond
to sources with more lumens lost over their life. I think if you test
again with metal halide or mercury you'll find those a little lower
still.

If accurate results are the goal, I would typically use the approach of
specifying everything as white, with my own -m multiplier calculated
from lamp manufacturer maintained lumen data.

obtaining accurate CIExy data and lumen maintenance values for all the

new

lamps coming out seems a bit daunting too.

Regarding accuracy of factors in lamp.tab, the values for metal halide
and high pressure sodium seem higher than real world lumen depreciation,
and halogen is defined with a factor of 1, but should still have some
depreciation over time. Considering that these factors vary so much,
particularly between different types of metal halide, this could be
difficult to keep accurate as part of the distribution.

···

____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses

Thanks Chris,

You are right-on, the reductions in luminance I was seeing were directly attributable to the lumen depreciation factors used for each light source. The color shifts are not having a net impact on the photometric values. I looked at five different light sources with lampcolor, converted the rgb radiance to luminance values, and the differences all correlated to the lamp lumen depreciation being applied. Phew.

I will continue using white as a default light source color (and remind myself once again that that lumen depreciation is being applied if I do stray from white!).

- Rob

···

________________________________
From: Christopher Rush <[email protected]>
Reply-To: Radiance general discussion <[email protected]>
Date: Thu, 8 Oct 2009 12:33:32 -0600
To: Radiance general discussion <[email protected]>
Subject: RE: [Radiance-general] getting irradiance from source defined inphotometric units

Someone please speak up if my understanding is flawed, but ...

I would e.g. assume an efficacy of 100lm/W for my source, and if

ies2rad

would use 179 instead, the radiance values in my simulation would be

by

far too low...

If you're using IES files via ies2rad, the IES files are measured in
candela, and the physically measured values already account for
efficiency of the source from electrical power in - through to the
lighting power out. The factor of 179 is just a conversion from W/m2 to
lumens/m2, as a unit conversion, not an efficiency factor. The source
efficiency is already incorporated into the physical measurements
documented in the IES file.

I used white, incandescent, and fluorescent, and spread 200 lumens

over

a 1 meter polygon, and ended up with 63.7, 60.5, and 54.1 cd/m^2
respectively.

I had understood the factors lamp.tab are mainly intended to account for
lumen depreciation, and just happen to have some color offset included
as part of this. In Rob's tests the decreasing values roughly correspond
to sources with more lumens lost over their life. I think if you test
again with metal halide or mercury you'll find those a little lower
still.

If accurate results are the goal, I would typically use the approach of
specifying everything as white, with my own -m multiplier calculated
from lamp manufacturer maintained lumen data.

obtaining accurate CIExy data and lumen maintenance values for all the

new

lamps coming out seems a bit daunting too.

Regarding accuracy of factors in lamp.tab, the values for metal halide
and high pressure sodium seem higher than real world lumen depreciation,
and halogen is defined with a factor of 1, but should still have some
depreciation over time. Considering that these factors vary so much,
particularly between different types of metal halide, this could be
difficult to keep accurate as part of the distribution.

____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

Hi Thomas,

Two words: "my bad". You are absolutely correct; if left to its own devices, ies2rad will scan the header lines for a lamp type and look for a match in the lamp.tab. Sorry about that! I guess I forgot about that option because it seemed so random I never understood why anyone would want to leave it up to such a potentially unreliable method, and have defaulted to using the -t option to explicitly specify my lamp types. Indeed, Greg notes in the manpage, regarding the whole [LAMP] header:

"This method of assigning colors to fixtures is less than perfect, and the IES would do well to include explicit spectral information somehow in their specification."

In any event, I apologize for the mis-information. If you ask me, I feel like this is one more reason to always use the -t option.

···

________________________________
From: Thomas Bleicher <[email protected]>
Reply-To: Radiance general discussion <[email protected]>
Date: Mon, 12 Oct 2009 02:36:48 -0600
To: Radiance general discussion <[email protected]>
Subject: Re: [Radiance-general] getting irradiance from source defined inphotometric units

Hi Rob.

On Sun, Oct 11, 2009 at 7:15 PM, Guglielmetti, Robert <[email protected]> wrote:

Hi Lars,

I'm a little hazy on a couple of your questions. When you say "used a WHITE keyword in your IES", do you mean using the "-t white" option in your ies2rad command, or actually adding a keyword to the ies file? Because ies2rad ignores the keywords in the ies file

That's not quite correct. ies2rad looks at the [LAMP] entry and
tries to match an entry in lamp.tab to the string found. Mostly
that doesn't work because there is no standard in naming lamp
sources. The few matches I remember were accidental.

It's better to specify the lamp type on the command line even
if you're not actually interested in the light color. I even thought
about patching ies2rad to make the WHITE type the default and
skip the lookup altogether. That would garantee a predictable
conversion.

Regards,
Thomas

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general