Even with current technology, does LED not pull less for the same comfortable light levels? Any numbers for this?
This is an insightful paper to read:
http://cltc.ucdavis.edu/sites/defau...led-linear-retrofit-solutions-alcs-retail.pdf
When evaluating quality of light, glare and shadow formation matters. Many fixtures purposely sends light upward and opts to use parabolic mirror reflector. If these qualities did not matter, fixtures would not waste valuable lumens to massage light quality. Sometimes light that is highly directional and casts a very sharp shadow against the wall is desirable such as security lighting.
You could get a decent light level on a desk surface in middle of the warehouse with a projector beam headlight pointing at the desk from the ceiling. You'll pass the foot candle level on the desk plane, but you will have a strong shadow of anything between the ceiling and the desk and this is not a "high quality light". Granted this is an extreme example, but LED based solutions that claim to improve light level substantially with less lumen output have a tendency to lower quality of light in this manner.
Delivered lumen is all the lumens leaving the luminaire and how the delivered lumens form a particular luminous distribution map (Read that PDF) is optics issue, not LED vs competition. Fixture efficiency relates to lumens that leave the fixture vs lumens that consumed within the fixture.
For general purpose indoor area lighting, indirect,completely non-directional light is often the best quality light. It increases uniformity, reduces glare and you also get a low foot candles per lamp lumen.
-Light meters need to be used correctly. Often, light meter is used pointing straight up without bothering to take vertical foot candle.
One also needs to remember that light meter is not an image forming sensor and it is blind. It sees the world through frosted lens goggles. A large frosted light bulb and a tiny clear bulb with the same lumen output looks the same behind it. If you're looking at them with your eyes normally you can see the tiny bulb is a glare bomb.
I do not doubt this, but isn't it still less then what it is replacing, even at 14.3kw?
Compared to what? You have to think like you're car shopping.
Typical LED scenario
Before: Aging large sedan with a lot of problems.
Considering: Overpriced high mpg hybrid with mandatory for $60,000. Compulsively fully loaded(not available without unwanted bells and whistles...)
(why on the earth did you not consider $30-40K vehicle with features you want as option B and compare value with the first option????)
Standard HID requires cooling down before it can re-start. it takes 5-10 minutes to reach full output. Halls and gyms have a lot of lighting hours in between sessions and reservations. 50 minute session and 10 minute vacancy doesn't allow MH to power cycle. Equipment limitations forces the light to remain lit in between sessions and limits efficiency to 83% in this example. Neither LED nor fluorescent require the lights to remain on. So HID gym lights end up staying on all school hours unless there's about an hour or more of vacancy.
1.) The kWh savings from being able to turn on and off whenever is significant even if lm/W stayed the same.
2.) LEDs are incorrectly credited for savings from this. Facility decision makers are not made ware that *either* LEDs, T5, or T8 hi-bays can take advantage of this savings and convenience. If the only thing they have is a sales pitch on highly expensive LED and made to believe he needs to go LED to do this, he just doesn't know about anything else to compare against the highly expensive LED.
True, and then you replace them- when the lights get to dim. I know there were roadways/buildings/parking lots of the past where Westinghouse life guards were being driven decades to the point 1000 watt H36 bulbs where putting out as much light as 100 watt incandescent- but to me that is just bad maintenance. When a mercury vapor bulb hit 24,000 hours, you were supposed to replace it, even though it could technically go for 100,000.
So, you get my point. The cost difference between a 175 and 400W MH isn't much, but LED fixture prices rise proportionately with installed lumen capacity at this caliber of lighting. While probe MH can lose 40% between re-lamp cycles, fluorescent systems only lose 10% and when this feasible alternate can do it with 10% light loss, it's unreasonable to give special lenience for LEDs and allow 30% degradation before reaching end-of-life threshold for applications where 4'x2' fixtures can be used.
There's a big price to pay for allowing this much performance degradation.
1.) Requires at least an additional 43% lumen capacity, or quoting end of life at L90
2.) Adding installed system capacity of 43% plus active degradation compensation(which adds $ $ $) and planned kW growth for that lighting system.
3.) Maintaining efficiency means the fixture would require scheduled cleaning or at the minimum wipe-down. So for cost analysis, it is important to compare the cost of LED cleaning only vs fluorescent remove old lamp, wipe down, then install new lamp.
Also, if the installed price of LED is significantly more than double that of the best T8 or T5 system, it wouldn't be totally unreasonable to consider installing double the quantity of fluorescent fixtures and wire them on controls side so that they only operate in alternating bank A or B pattern to spread the wear across two banks.
But my question to you is do you need full spectrum? For most applications it appears none full spectrum lights get the job done be it retail or office. FWIW before LEDs, florescent and coated MH did a good enough job in retail which also lacked full spectrum- and perhaps by a long shot.
CRI is not the "be all". It's the average of deviation distance for eight sample color chips, but these samples do not represent all the colors we can see and it also affects fluorescent secondary emission. Fabric and paper brighteners are deactivated under LED lighting. The missing spectrum between blue and green affects the way some cyan/torquoise printing ink and pigments appear.
The deviation can go either way. Too dull, or too vivid. Ok, so you have four 20" tires. If the actuals are right side at 20.4" and left side at 19.6", the average of deviation from center is 0.4". I think we can see the problem with that. A metric like CRI adds up each deviation and divide by the number of test points. The average of 0.4 and 0.4 is 0.4, or 2% of 20. If they're ALL 20.4", it is still 2% average deviation. This makes either of these examples look equal.
Adding another twist, ordinary light sources except for incandescent are poor in deep red rendition dulls the rendition of vibrant red, such as fruits, vegetables, red meat and blood. This falls outside the measurement scale of ordinary CRI. Broad spectrum fluorescent that extends output into deep red such as cool white deluxe also have phenomenal color rendition. You can not see it by looking at the light, or staring at the wall. Arrange some tomatoes, strawberries and red meat on the counter and switch lamp back and forth between standard 4100K LED or RE841 FL vs CWX and you will see it. Who cares? Those in the business where the appearance of such things influence buying decision certain do. After all, the same shipment looks more fresh and delicious under wide spectrum MH or CWX than general purpose LED/FL.
Also- consider this and tell me if you agree. In the 90s I remember most of the big super markets and box stores around me were lit with 400 watt coated probe start M59 MH lamps. These lamps had next to no lumen maintenance, color rendering and CRI got worse as they aged IMO (I remember despite the phosphor literally half would turn pink-eye pink the others dull green)- yet despite these lamps being awful by any of today's standards people got by and they did there job. Then came florescent high bays- F32T8 and then T5- they had superior lumen maintenance and still lasted 24,000 hours if not a bit more. But none the less they are still not full spectrum.
Yes, but extend that to 60-70,000 hours if they're using extra long life lamps. One of the popular applications for ceramic metal halides are produce department and anywhere else where you need broad spectrum light at higher kelvin that halogen lamps can not make without color filters. When 3,000K does the job such as boutique shops, halogen accent is their go to.
Now we have LEDs. Is it safe to say that most of the LED products are still better than old 400 watt MH lamps when all is said and done?
No. Plenty of substandard retrofit products. Retrofits have to operate like a chandelier hung inside an inverted metal bucket. The socket isn't much better than a chandelier chain for conducting heat away from the retrofit lamp. Metal halide lamps have a large outer bulb for ease of handling, but the actual lamp inside is quite small. There are special double ended HID lamps that look like halogen work light lamps. These require very careful handling but allows for better and smaller fixture with a better efficiency since it doesn't have to be built around the huge outer bulb. Both the LED elements and its transistorized ballast are heat sensitive. I don't mean that they can't handle "too hot to touch", but they can't operate at 180C coil temperature like magnetic ballast.
When you can have plenty of space in all directions and access to clearance on all six sides, it's easy to use LEDs with a train track like frame. LEDs mount on the bottom of ties and the space between each tie and generous open air clearance on top gives plenty of airflow through the fixture.
