[Podagrower]

I'm looking for ideas to solve a problem. I have an installation where the measured foot candles are lower than the computer models. 3 computer models using 2 different programs show approximately 40% more foot candles than 3 different foot candle meters are showing.

So far, 2 different electrical contractors and the electrical engineer have looked at the install and found no issues. Measured voltage at panels and fixtures are well within tolerance for the ballasts, voltage drop is almost non existent, the building is served by it's own POCO transformer, there are probably 200 fluorescent fixtures on the 277 system, and maybe 50 computers on the 120 side, each circuit is a dedicated neutral, so I can't imagine a harmonic issue.

I'm looking for ideas on what the problem is. I think we have eliminated: voltage, neutrals, temperatures, surface reflectivity, and mounting heights as possible causes. I'm willing to entertain just about any theory at this point.

 

[wildleg]

were the bulbs burnt in properly ?

are the fixtures connected to an energy management or dimming controller ?

are the ballasts dimmable (are they set to max) ? (it's easy to have the settings wrong in some dimming panels)

 

[99cents]

You might go back and do an old fashioned zonal cavity calc to verify the software results. What software are you using?

 

[Podagrower]

were the bulbs burnt in properly ?

are the fixtures connected to an energy management or dimming controller ?

are the ballasts dimmable (are they set to max) ? (it's easy to have the settings wrong in some dimming panels)

All but one office area were burnt in extremely well. The office area is controlled by occur sensors, and those bulbs show significant blackening, but changing them to new bulbs did not result in any significant change.

Most of the fixtures are controlled by a lighting control panel, but it has no dimming capabilities, and we bypassed it completely (used an extension cord to connect a group of fixtures to a 120v circuit with no change in levels)

There are a few dimmable fixtures, but the trim pots are all the way up, bypassing the dimmer produces no change.

 

[Podagrower]

You might go back and do an old fashioned zonal cavity calc to verify the software results. What software are you using?

I don't remember the software name the EE and lighting supplier used, but the Architect was satisfied with it, the lighting supplier also used a second program and got the same results.

Our "control" is a 10x14 room with a single 3 lamp F32T8 parabolic mounted at 10'. Computer spits out 56, the EE and lighting supplier's brains spit out 50, but we have 25. (Rough numbers from memory). Wall reflectivity set at 30 in the computer models, closer to 50 in reality, white acoustic ceiling, dark carpet on the floor.

 

[99cents]

I'm betting on a design issue, not an electrical one. The old adage about software - garbage in, garbage out - might be at play here. Input values might be overstated, photometry might even be off. If you supplied and installed fixtures and lamps according to spec, let the lighting agent and engineer fight it out.

 

[daveEM]

^^ What 99 says. The meters are correct and I might have used two. Three and you're really checking things out.

 

[dspiffy]

I'm betting on a design issue, not an electrical one. The old adage about software - garbage in, garbage out - might be at play here. Input values might be overstated, photometry might even be off. If you supplied and installed fixtures and lamps according to spec, let the lighting agent and engineer fight it out.

If it ends up being anything other thing this, I'd really love to hear.

 

[wildleg]

unless the lenses you got are totally screwing the light, I agree with ^ those guys, cause it sounds like you've covered all the other bases

(BTW, do the fixtures have good reflectors ?)

lenses and reflectors can make a huge difference.

 

[Podagrower]

Parabolic lenses, no special reflector behind the bulbs, which I know would help. So far 2 of the existing fixtures have been removed, 1 sent to the manufacturer, 1 taken to the lighting distributor, both produced the expected light levels when installed in other buildings. The EE and lighting supplier have been unable to provide a solution in 6 months, and I'd like to have a resolution.

 

[dspiffy]

So when you measure one fixture directly, it measures lower output in this building than in the other two buildings?

 

[wildleg]

Parabolic lenses, no special reflector behind the bulbs, which I know would help. So far 2 of the existing fixtures have been removed, 1 sent to the manufacturer, 1 taken to the lighting distributor, both produced the expected light levels when installed in other buildings. The EE and lighting supplier have been unable to provide a solution in 6 months, and I'd like to have a resolution.

wow. you done did your homework. Now I'm a little confused - are you saying it is (a) definitely a design issue or are you saying (b) they actually aren't putting out the same amount of light in the installed building ? I'm seeing a, but if it's b then it shouldn't be too hard to find the culprit (?)

 

[wcord]

Did the other buildings have the same color floor and office furniture?
Not that the difference should account for such a large decrease in output
Grasping at straws on the design side.

 

[99cents]

This has to be a design issue. Any half a$$ed lighting designer should be using maintained lumens, not initial, and allowing for dirt depreciation. A new, clean fixture with new lamps should be giving you higher readings than your software. Your software should be giving you your lumens a year or two from now, not today.

I hope this wasn't a design build because this is going to take some money to rectify.

 

[99cents]

Parabolic lenses, no special reflector behind the bulbs, which I know would help. So far 2 of the existing fixtures have been removed, 1 sent to the manufacturer, 1 taken to the lighting distributor, both produced the expected light levels when installed in other buildings. The EE and lighting supplier have been unable to provide a solution in 6 months, and I'd like to have a resolution.

Something's not adding up here. If this is a deep cell parabolic your light levels are highest directly below the fixture and fade out quickly towards the perimeter. The only thing I can suggest is to isolate one fixture and attempt to compare your readings to the lab photometry.

 

[Podagrower]

So when you measure one fixture directly, it measures lower output in this building than in the other two buildings?

wow. you done did your homework. Now I'm a little confused - are you saying it is (a) definitely a design issue or are you saying (b) they actually aren't putting out the same amount of light in the installed building ? I'm seeing a, but if it's b then it shouldn't be too hard to find the culprit (?)

That is what the manufacturer and designer are saying. Nobody else was at their facilities to witness the testing. I really didn't want to think that they would both lie, but I'm really starting to question their results now.

 

[Podagrower]

Did the other buildings have the same color floor and office furniture?
Not that the difference should account for such a large decrease in output
Grasping at straws on the design side.

I do not know about the manufacturer's space, I assume it to be a "clean" room. The designer's office space has lighter colored walls and flooring, but it would be a heck of a straw grasp to double the foot candles.

 

[Podagrower]

This has to be a design issue. Any half a$$ed lighting designer should be using maintained lumens, not initial, and allowing for dirt depreciation. A new, clean fixture with new lamps should be giving you higher readings than your software. Your software should be giving you your lumens a year or two from now, not today.

I hope this wasn't a design build because this is going to take some money to rectify.

No, not design build, thankfully.

 

[Podagrower]

Oddly enough, I started playing around with some free foot candle calculators online. It helps a lot. I can't get the foot candles with a computer that they have, either in their design or their test. The software I'm using kicks out a number much closer to what we are actually seeing in the field.

 

[99cents]

I appreciate a guy who likes to walk away from a job well done but is this really your problem? An engineer who depends on a lighting supplier to do his layouts for him is going to find himself in deep chit sooner or later. If you installed spec fixtures according to the drawings, isn't your job done?