Nichia sent me hundreds of several different LED types for my past city garden light project prototyping. I found some very interesting “BLF-worthy” LEDs. One of them is currently the only medium power Nichia LED which is OEM rated up to 2,4 watt/LED. For 3030 medium power LED, it’s like an XHP50 to XML. It’s Nichia’s most powerful medium power LED. There are two variants available: NF2W757G-V3 (3V) and NF2W757GR-V3 (6V)
I’m only interested in NF2W757G*R*-V3 (GRV3) because it’s OEM rated up to 2,4watt, 2nd most efficient in Nichia MP LED, and available all the way from 2000K up to 6500K in R70, R8000, and R9050.
If you opt for 3V version, forget NF2W757G-V3, better go with NF2W757G-V3F1 (lower voltage, brighter). There’s NF2W757H with even higher efficiency butt… unfortunately, only a tad higher than GV3 at below 35mA. Above 35mA, GV3, GRV3, and GV3F1 smoked it by significant lead.
Prepared some test victims:
The biggest board (VS35SP36) is from my early attempt, you can see some solder balls and slightly misaligned LED. I don’t think that would cause measurable output difference. I don’t want to kill another 36pcs (only few left) nor doing rework on them. Because another rework will just put more stress on the LEDs, possibly reducing the output. They’re all NF2W757GRT-V3, P20, sm305, R8000.
VS35SP36 wired in 36P (6V)
Total loss water cooling. Had to shielded from the light, that was blinding!
My RD DPS5015 couldn’t go higher than 15,09 A at below wattage. I only have 180watt DC PSU at 19,65V max voltage. Wiring the module to 3S12P would require close to 21V input.
The LED module maxed out just a tad lower at 14,4A.
Here’s the result:
It’s actually unfair to compare the tiny 10mm non DTP copper board to 21mm ceramic non DTP board but that’s all I had. You might think GRV3 over current is not that great or perhaps why the 10mm board performed somewhat bad. The output of 10mm started to look less than adequate at 200mA. Remember this is a 6V LED, that means it outputs roughly twice the 3V LED. So if you use 3V LED, 10mm copper non DTP should perform equally good to ceramic up to 500mA (in single LED application).
The voltage swing is quite high from 5,68V at 50mA up to 9,01V at 850mA
I included the plot of ceramic MCPCB ouput prediction using data from VR21P4 above multiplied by 36. In real world we can expect about 90-95% of the plotted output. RD’s DPS5015 couldn’t get steady reading below 50mA (pulsating output). The efficiency should be higher than 240lm/watt at 30mA total current.
4 LEDs damaged. Still unsure if it’s test related or simply burnt saw dust. My workshop was rather dusty during the test. Saw the smoke started at 10,8A but the output was climbing steadily up to 14,4A. At 8800lm, even a slight oily surface can start the burn.
Conclusions:
- See what can we get from 36pcs $0,1 LED + $1 MCPCB… Of course any high power LED such as XPL2, 319B, etc…will outperform these LED. But we have to spend $$$ to get the same or higher efficiency than this. At $4,6/module it’s the cheapest 9000lm CRI80 I’ve ever seen.
- To get the most out of these GRV3, a nanoceramic MCPCB is needed. But at lower power level this VS35SP36 is more than capable to give 5000lm of efficient lighting. In general lighting it would be hard to design a fixture with passive cooling for more than 50 watt of LED power. In my opinion, 2000lm per module with this GRV3 is the sweet spot for efficiency and heat management. 180lm/watt is very good.
[Clemence]