LED test / short review - Fireflylite FFL351A 1800 K 95 CRI - beautiful warm light with perfect beam!

LED test / short review


Fireflylite FFL351A (1800 K 95 CRI)


2024-05-25


Link to original test: Klick

  • Bin: unknown
  • Color group: unknown (1800 K)
  • CRI: 95

Note: There is no official datasheet available!


The emitter was sent free of charge by Fireflylite in mid-May as compensation for a very long delay in delivery. Thanks to Fireflylite for this service!

In contrast to the previously tested FFL351A in 3700 K, the top surface of the 1800 K version is orange. The illuminated area is almost invisible due to the coloring and the thick phosphor. In contrast to most modern flip chip LEDs, the phosphor appears more “segmented”, similar to the surface of stars such as the sun (called granules here), especially at very low operating currents.

As with most modern LEDs and the FFL351A with 3700 K, light is also emitted to the side. Otherwise there are no differences in design and handling to the 3700 K version.





  • at 3,000 mA (official maximum current): 560 lm @ 3.22 V
  • Power at official maximum: 9.7 W
  • Efficiency at 3,000 mA: 57.9 lm/W
  • Maximum reached at 9.0 A, at this point 1024 lm @ 3.99 V
  • Power at maximum 35.9 W
  • Efficiency at maximum 28.5 lm/W



Data for 25 °C Tsp (at 85 °C the luminance values are around 13 % lower).


Due to its very warm CCT combined with high color rendering, the 1800 K version is very inefficient, especially at currents close to the maximum possible. At around 29 lm/W, its efficiency is lower than that of an overdriven halogen incandescent light. The FFL351A 1800 K should not be driven with more than 5 A, particularly because the increase in luminous flux beyond this is only marginal anyway. Nevertheless, the maximum achievable luminous flux of over 1000 lm at 1800 K color temperature is impressive, especially for an LED without silicone dome.

The FFL505A has a considerably smaller illuminated area and a high red component in the spectrum, which is why it is hardly more efficient than the 1800 K version despite its larger footprint. This is a general problem with LEDs with a very small luminous area - the heat dissipation area from the LED chip is simply limited by its surface area and as soon as the maximum heat density for it’s small thermal dissipation area is reached, a further increase in performance is impossible.

The higher Vf and the changed characteristic curve compared to the 3500 K variant are striking. It is possible that a different LED chip is used here, or that the series dispersion is clearly pronounced.



The tint is great and very warm. With 1890 K and a negative duv, the light simply looks good. The high color rendering index makes blue and green colors in particular stand out much better. In comparison, the light of the HL2X 1800 K appears much more orange - more like PC Amber - and blue colors in particular are covered by some kind of “grey haze”, which is also consistent with the lower color rendering index of the HL2X, which is only at around 80.

For all those who are looking for an LED for living room or ambient lighting and prefer very warm CCT, the FFL351A 1800 K is close to absolute perfection. Especially as the beam looks perfect in SMO reflectors, as with the colder version, and there are no annoying artifacts, this emitter is definitely worth another look.


  • Ra: 92
  • R9: 86
  • CCT: 1890 K
  • duv: -0.0031

This LED should currently occupy a special position. With XP footprint and therefore maximum compatibility even in existing solutions, quite high output for this warm CCT and high CRI and excellent light in reflectors and lenses alike, this LED is particularly suitable for living room and mood lighting.

Unfortunately, as with many Chinese emitters, there is no information other than that given on the Fireflylite website, and there is also no official data sheet available. Furthermore, this emitter is currently only available in China.


Thank you for reading this short test. :slight_smile:

Greetings, Dominik


v1.0.0
11 Thanks

I’m not sure if this segmentation is normal or should be expected. The one other time I’ve seen such segmentation is a damaged SFT40 from djozz’s overcurrent test.

505A seems inefficient compared to 351A, the output of 505A 3500K kind of tracks the 351A at 1800K…

Edit: just realized the difference is in the Vf, maybe it makes 505A a tad bit more efficient, but I expect more from the jump in footprint and color temp.

Like I wrote: this whole efficiency topic also depends on the size of the LES (LED chip). We compare a LED with a significant smaller LED chip (FFL505A), which is also round, with a larger one (FFL351A), which is also square and has therefore much bigger heat dissipation area.

The smaller the LED chip, the smaller the area over which heat can be dissipated. With LED chips, heat can only be dissipated via the underside. At some point, a limit is reached beyond which no further heat can be dissipated over a defined area. This can be seen in the 3535 footprint, for example; due to the limited size of the thermal pad, LEDs in this size class can only achieve 7-10 A, anything above this is only feasible with a (strong) reduction in Vf - which significantly increases efficiency. The maximum possible output remains unchanged, however, and is around 50-60 W. The LED with the highest tested output on an 3535 footprint is the Luxeon MZ, at 58.5 watts. However, it should be noted that this is actually a 4040 LED, so the output on a 4040 board is likely to be somewhat higher. In addition, there are four individual LED chips on the substrate, which also further increases the heat dissipation surface. And - not to forget - the Vf of these quad-die emitters with 4P config is really low so the achievable current can be really high.

The 5050 footprint allows outputs of up to around 70-80 watts, although the design of the LED again plays a key role here. The XHP50 LEDs can exploit this output, whereas an SFT-40 is limited by its single (smaller) LED chip and only achieves a maximum of 56 watts. Also the (usually) higher Vf of these single-chip emitters plays a certain role here.

LEDs with very large LED chips or multi-die emitters always have an advantage here. In this respect, LEDs such as the 16-die Sanan models could exploit the maximum physically possible with their footprint.

2 Thanks

The pictures of the LED in my test are always taken before the testing. These emitters came right from the reel with this segmentation. I also swapped two emitters with these 1800 K FFL351A and they had the same segmentation on the LES as the sample shown here.

2 Thanks

Thank you Dominik! Reading your test is a pleasure

1 Thank