So 12 is better than HPS but let's talk theory a bit because I don't want to end up with so much light I'm harming the plants. 4 CXB3070s in a 2x2 area (same as 16 in a 4x4) at 1.4A. That's about 28000 lumens in the 2x2. I realize it's overkill but is it too much?
For a comparison in a 4x4, 112,000 lumens versus 140,000 from a 1000w HPS. You can see why I'm wanting to do what I have in mind. The light quality on the HPS might not be as great but it's still 140,000 lumens coming down.
If that light level is unlikely to harm the plants, the second question is whether there would be any benefit over say 12 emitters rather than 16. I think for many growers maximizing their yield in a given area (typically smallish area) is more important than saving some electricity.
The other issue would be heat, but here again the CXBs would be emitting less heat than the HPS. Of course HPS could be enclosed in a vented hood or cool tube. While similar could be done with LED if heat was an issue it would be a DIY job and nothing as simple as a cool tube.
To the point, I assume you too are an evil genius, so why 12 instead of 16?
Ok lets talk theory. I order to do that...forget about lumens. Lumens are bass on the spectrum and the luminosity function. If the spectrums are not 100% the same...lumens are worthless. And even then are better ways of measuring ouput.
Lumens are for humans. Plants use photons. About 8-10 per carbon molecule fixated to f we really get down to the details.
My goal was to make a 1:1 swap with a 1000hps that will actually save you money. There has been no doubt in anyone's mind for years that watt for watt led's are better. That just cost a lot, and no way of making it back over time. 12 does that based on the PPF(photosynthetic PHOTON flux) output and the PPFD( photosynthetic photon flux density/DISTRIBUTION) footprint.
PPF is how much his possibly emitted and the PPFD is how many are reaching the canopy. Both are important when designing. PPF for easier calculation purposes can be expressed in radiant watts...or PARwatts.
A CXB3070 fixture at 1400 producing close to 50% light(48+%) and pulling around 600w comes out to ~290-300PARwatts. And is emitting it in the correct direction...at the canopy with little or no need to redirect via reflectors lenses or anything. And will also maintain that efficiency and output indefinitely with the correct cooling system...which isn't hard to do at all.
A 1000w hps brand new out the gates is hitting ~36% efficient and producing 360 PARwatts. But is doing it in 360*and requiring at least 50% of it to be redirected back towards the canopy. That can not be done without a price...photos being that price. Reflectors are not truely very effective in the grand scheme. 98% reflectors do not reflect 98% or even close to it, closer to 85 at best. Also the bulb itself is blocking upwards of another 10% of that already penalized reflected light. The grand total after all losses has been tested to only emitt out 70-80% of the total PPF..and still has some straying off the canopy(but we'll let that slide for this). So we are looking at ~250-290 PARwatts available to hit the canopy.
That is for a brand new bulb. Check the arc rod in a hps bulb after 1, maybe 2 months use...bent from the heat...the heat that's degrading the bulb very fast. Their lm70 will shows you in perfect conditions that they drop below 90% their ouput very quickly. And below 85 by 3 grows. While the led is still at 100...actually about a month into led they get over 100% for awhile. Phosphorus likes a little use.
As detrimental as that sounds for hps...hps still kick ass as we know with that output. But we can clearly see their weaknesses and is exactly how we can capitalize on it using led's. And how I have come to what I have come to.
16 would work just fine I'm sure. Would not expect any kind of problems. Cost is up and savings are down...that is personal choice. Spread could be up. All depends on design.
Edit:
Oh ya...lets not forget about he spectral benifits of leds. Distributing all that quanta over a broad range, minimizing any over saturation and also improving plants physiological processes.