LED house lighting calculations

[Esoteric]

Okay so as I am setting up the house lights for this install, they want 20 fc at the floor, so I did all the calculations. It is a 14,000 sq/ft room, at 20 lumens per sq/ft that is 282,400 lumens. The fixtures in question are 5400 lumens each. That means, giving a 10% leeway looking at appx 60 units. My question is, I didn't take into account ceiling height in my calculations, and I can't find anywhere how I should take this into account. I used the inverse square rule in college, but that was not using lumens. Do you have any insight on what I am forgetting?

Mike

 

[ajb]

Ceiling height doesn't really fit with your method of calculation. When you converted from required foot-candles to lumens, the inverse-square factor is implicit because you already know the area you're illuminating. That's what inverse-square is really doing, factoring in the increased area of the base of a cone of light as the height of that cone increases.

The mistake I think you might be making is that you're not accounting for the *angle* of that cone. Your method gives you the required number of fixtures to acheive an *average* of 20 fc over the illuminated area (since you're throwing 282,400lm at 14,000 sf), but does nothing to guarantee that those lumen are evenly distributed across the floor. You need to look at beam angles + ceiling height for that.

 

[epimetheus]

I'm interested in this calculation you're doing, but unfortunately I've never really done this type of calculation before. Do you have the photometrics of the fixture you're using? From reading definitions in the Wiki, it seems that the lumen output from a fixture drops as the throw distance increases, proportionally to the inverse square law. The 5400 lumens per fixture that you quote needs to be qualified with the distance from the source that is measured at.

Using a 26 degree S4-750 as an example; it has a field angle of 25 degrees and outputs 196fc at 30' with a field diameter of 13.4'. Since one foot candle equals one lumen per square foot, that's 27641 lumens over a field area of 141 square feet.

Is that example correct? It doesn't seem like it is. My apologies if I'm only adding confusion here.

 

[tjrobb]

FWIW, the commercial calculations are somewhat complex. I'll just note the "basics".

First, find the (room) cavity ratio: ( 5*ceiling height*(length+width) ) divided by the area.
After you have the RCR, find the manufacturers specs on the coefficient of utilization (COU) for the fixture (it'll be in the specs). Then FC = ( COU*number of lamps in each fixture*number of fixtures*lumen depreciation*ballast factor ) divided by area of the room. Lumen depreciation is dirt, lens obstructions, etc.; ballast factor is lumens lost with the ballast(s).

Yeah, the calculation bites. If you prefer I have a spreadsheet I built when I got tired of doing it by hand. You will still need the spec sheets to find the COU, though.

 

[Esoteric]

Man, I totally forgot that from my commercial lighting class. Is there any way to do it without the COU? It is not in the lit for the units.

Mike

 

[dvgmacdonald]

If you've got an ies photometric file on the fixture and CAD of the room you can use software to do it. This also helps determine the best layout for the fixtures. Rule of thumb, wall to 1st fixture on either side should be roughly 50% of the difference between fixtures.

Some of the manufacturers have tools on their sites to do simple layouts, too. I know Cooper used to have one that would let you choose a fixture, put in basic room dimensions & give it some performance specs & it would output a number of fixtures and a layout.

Re: COU, if you can't get it from the manufacturer you can find another fixture with similar beam spread & distribution (not always easy, but it can be done) and use the COU from its datasheet. If it's just not available I'd say you should calculate using at least 150% the number of fixtures that your initial lumen calculation suggested. It's fairly rare to get a COU over 90% (though the fact that it's a theatre means it's probably fairly high as our spaces tend to have lots of floor area & relatively little height, plus your LEDs are going to dim over time to about 70% at 50,000 hours (that's probably a long time, but even so, I'd use a maximum of 85%), and you'll see some dirt depreciation (even if you're cleaning these fixtures quarterly you can probably say you get a value around 95%. These relatively rosy calculations give me about 70% of the rated lumens getting to the floor, or a requirement of 1.38 fixtures to give you the rated lumens of 1 fixture. That's a relatively optimistic calculation, too.

 

[Esoteric]

Okay, somewhere my calculations got jacked. The unit has a COU of .95 and when you take into account .05 for obstruction, .05 lamp dep, .05 for ballast factor, I get....

fc=(cou*lamps*fixtures*dep*ballast)/room area gives me...

20=(.95*1*x*.05*.05)/14000
Solve for x and I end up with x = 135000000+ fixtures. That can not be right.

Mike

 

[Esoteric]

I found this in an old notebook however and it seems a little more reasonable...

fc = (lamps*lumens*COU*Light Loss)/area
23=(x*5476*.95*.85)/14000
solve for x = 72 units

Mike

 

[epimetheus]

Awesome info in this thread folks! Mike, I would imagine that your lamp dep., obstr. and ballast factors should be .95 instead of .05. Also, since fc=lumen/sq. ft., should x= number of fixtures times lumen output per fixture?

Edit: My comments were in reference to post #7.

 

[Esoteric]

You might be right. If I did .95 instead of .05 with the way the formula is worded that would mean I would be losing 95% of the light at the point of illumination? These units retain 95% of their light at the point of illumination.

Mike

 

[tjrobb]

Nope, the .95 means that for every 100 lumens in the lamp(s), only 5 are "eaten" by the fixture (wasted).

 

[dvgmacdonald]

Nope, the .95 means that for every 100 lumens in the lamp(s), only 5 are "eaten" by the fixture (wasted).

Quite right, I was using those percentages as the light that got to the surface (12% light loss = 88% in the calc).

One thing to note, with LEDs you must take the photometrics from the entire fixture, not the "lamp". Because it's not just optics, but heat too, that effect LED output, the whole package is important to the photometry. Also, the IES has some standards on test conditions for LED fixtures. If you test them as soon as you turn them on in your 20 degree F garage you'll get really different numbers than you will after they've been burning for 24 hours in the desert summer. Make sure you're getting photometry that complies with the standards (LM-79). Otherwise, it's sort of meaningless, at least in terms of total luminous output (the distribution should stay roughly the same).

Also, LM-80 data on lumen maintenance should become available very soon (at least for high quality architectural lighting products). Take a look at these results, too, as the 50K or 100K hour life to 70% lumen output that we hear about so often may well not turn out to be the case.

Don't get me wrong, I'm a big believer in LED lighting, I just think it's important that people understand how to separate the truths from the myths and the wheat from the chaff.

 

[Esoteric]

I agree 100%. These are guaranteed to 70% at 50,000 hours.

Mike

 

[renegadeblack]

While I can't add any speculations to the calculations, PLEASE oh PLEASE makes sure that your LEDs and dimmers match up... it's just plain awful when they dont.

 

[Esoteric]

I am sorry? LEDs and dimmers match up? I don't understand.

Mike

 

[dvgmacdonald]

I think he's trying to get to the point that most, if not all, LED fixtures cannot simply be hooked into a stage dimmer & expected to work. Many require reverse phase dimmers, others require that you put a PWM dimmer between the driver & the fixture, others have a built in DMX interface, but that will require a bit more wiring than simply swapping the fixtures. In any case, make sure that you actually have equipment that can dim your new fixtures.

 

[Esoteric]

I think he's trying to get to the point that most, if not all, LED fixtures cannot simply be hooked into a stage dimmer & expected to work. Many require reverse phase dimmers, others require that you put a PWM dimmer between the driver & the fixture, others have a built in DMX interface, but that will require a bit more wiring than simply swapping the fixtures. In any case, make sure that you actually have equipment that can dim your new fixtures.

Oh, yes of course. All my LED products use 120v fixed power with DMX control. We will be pulling out the dimmers they were using for houselights and replacing them with traditional 120V circuit breakers.

Mike

 

[theatre4jc]

Old thread I know, but it's got some great info. Mike, what fixture were you using? I'm putting together a proposal and am looking at various LED house lights, trying to find fixtures in the 3-3500K range, already am looking at Affineon fixtures but want to see if there are others out there that might be a little more budget friendly.

 

[Esoteric]

My "buget" suggestion to them was the Chauvet Colorado1 VW. Nice, bright fixture. IP65 rated. DMX control. 2 year warranty. Drop me a line at [email protected] and i can run you through what we talked about.

Mike