LED test / review - ChangDa SFH43 3000 K 90 CRI - 16-die LED with low Vf and yellow tint

LED test / review EN


ChangDa SFH43 3000 K 90 CRI


2024-09-15


  • Type: multi die, flip chip
  • Bin: unknown
  • Color group: unknown
  • CRI: 90
  • Rated voltage: 3 V
  • Max. Forward current: 40,000 mA
  • Max. Peak current: — mA
  • Viewing angle: — °
  • Thermal resistance: — K/W
  • Max. Temperature Tj: max. —°C

Note: there is no official datasheet available!


The LED tested here was purchased from Kaidomain in August 2024. It is listed there as “SFH73 16x core 3 V 40 A”.

There is not much information about these LEDs. After some information given by BLF user @yoelpez, it turned out that these LEDs are manufactured by a small Chinese company called “ChangDa”. They buy various LED chips from Sanan and use them to manufacture their own LEDs in different sizes and CCT. So apart from the chips, there is no connection to Sanan!

The nomenclature of these ChangDa LEDs is relatively easy to understand from the information available:

1st and 2nd digit:

SF (ChangDa LED)

3rd digit: Number of LED chips

  • S = 1 *
  • Q = 4 *
  • N = 9 *
  • H = 16 *
  • P = 25 *
  • Y = 64 *

4th and 5th digit:

Size (side length) of a single LED chip in mil: 35, 40, 43, 55, 60, 75

Based on this nomenclature and the products listed by the manufacturer in the Taobao store, it is clear that the naming of the LED by Kaidomain is most likely incorrect. A single LED chip of the LED tested here corresponds to 43 mil, and only one SFH43 is listed in the Taobao store. According to the current state of knowledge, ChangDa does not use a 73 mil chip at all. In addition, the LED with 16 individual 73 mil chips in its 7070 footprint would simply be far too small.

This is why the LED is called “SFH43” in this test.

The design of the SFH43 is kept simple. A dark gray substrate is covered with a white silicone potting compound. The transparent protective layer is on top of this. It should be noted that this silicone layer is rubbery and literally “attracts” dirt and dust. The orange phosphor layer, under which the 16 LED chips are located, is embedded in this. All the chips are connected in parallel to ensure a forward voltage of around 3 V.

The SFH43 is 7.05 x 7.05 mm in size. The transparent silicone layer protrudes 0.05 mm from the substrate. It is interchangeable with other 7070 LEDs thanks to the familiar 7070 footprint, provided the low supply voltage is not a problem.

It can be seen how the phosphor was not applied evenly. Outside the center, the layer is obviously thinner, as a colder CCT is visible there. In addition, the gaps between the individual LED chips are clearly visible, especially at the edges. There may still be problems in applying the phosphor evenly during the production process. The uneven boundary of the phosphor also indicates this.

The LES is 32.3 mm² in size. Due to the light emitted from the side and the gaps between the individual LED chips, it is almost impossible to determine the exact illuminated area.






  • Maximum reached at 37 A, at this point 5599 lm @ 3.66 V
  • Power at maximum 135.5 W
  • Efficiency at maximum 41.6 lm/W




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


In the absence of an official data sheet and operating parameters, as well as a lack of comparison options, it is difficult to evaluate the performance. The maximum current of 40 A (!) specified by Kaidomain is not reached even with excellent cooling. The thermal resistance rather than the 7070 footprint is probably the limiting factor here, as this only reaches its limits at around 200 W. It is likely that the specified 40 A does not represent the official maximum current, although it is unclear to what extent this information is actually available from the manufacturer.

The Vf is extremely low for an LED in the 3 V class. This is due to the 16 chips connected in parallel, which are only supplied with 2.3 A each at 37 A and have a forward voltage of around 3.66 V each. In general, the parallel connection of many smaller LED chips is a good way to significantly increase the efficiency of an LED, which is also reflected in the power diagram.

The luminance is very low. This is due to the side-emitting LED chips and the very large illuminated area. A cool white SFH43 should have around 40-50% higher luminous flux and luminance.



The beam is perfectly usable, but artifacts can occur in the beam, especially with SMO reflectors, which is caused by the unevenly applied phosphor. The rings visible on the beamshot could be caused by inaccurate focus; this must be checked carefully before installation in a light, depending on the available optics. An OP reflector is therefore recommended for a good beam.



The tint is very yellow. This is clearly recognizable and the duv is correspondingly high.

The color rendering value is high, although the light quality is subjectively much lower than with comparable emitters. Colors appear washed out and less saturated. If good light quality is required at 3000 K, emitters such as the Nichia 519A / 719A / B35AM or Luminus SFT-40 / SFT-70 are clearly preferable.


  • Ra: 93
  • R9: 64
  • CCT: 3403 K
  • duv: 0.0058

Thanks to the 16 individual LED chips, the Vf is very low and the efficiency is high. The beam is particularly useful in OP reflectors, but the tint is very yellow, as with the SFN40 tested earlier, and is significantly less pleasant, especially in comparison with Nichia emitters.


Thank you for reading this test. :slight_smile:

Greetings, Dominik


v1.0.2
19 Thanks

Nice test, i was always wondering if chinese SFH leds are any good. Pls if you get a chance do SFH55 it is a 50A led. Crazy bright output though

6 Thanks

That’s surprisingly good efficiency for high CRI next to SFT40 even at lower currents. That tint is really bad though. I wonder if FFL could get their hands on these emitters and put a really good phosphor mix on these.

1 Thank

Thanks for your investigation! Resellers like KD do sometimes change their names to maintain information asymmetry.

And even though many emitters use SanAn chips, their obtained chip quality or binning is not transparent, so it makes more sense to just use their own trade/brand name than “SanAn”.

I am pretty sure, that KD just mistakenly switched naming of SFH43 and SFH73, so far all other LED names on Kaidomain are correct.

The only thing that surprises me is that ChangDa (which I tested earlier) doesn’t list SFN40 on their Taobao page. I have not found any further information (apart from KD listings and my test as well as dubious links).

SFN40 is indeed ChangDa’s emitter, but it may have been discontinued. I have only seen a few discussions in the CN flashlight community, and it may have only been mass-produced and retailed for a short period of time.

2 Thanks

It’s not one of the ones for sale but it appears in multiple pictures in the listing:


I have some ChangDa SFN43 5500K heading to me.
I’m hoping the tint is pleasant.

Thanks for testing all these LEDs for us!

1 Thank

That is interesting. thanks for the insight to this mysterious domestic market.

Tbh I wouldn’t be surprised if the tint is more on the green side. I am pretty sure that ChangDa just buys some phosphor mixtures and put them on the LED, without thinking too much about it…

2 Thanks

BTW, I heard that S = SMD and F = flat-top.

1 Thank

If you take this meaning, it seems that ChangDa is at least considering offering LEDs with domes - why else would they include this as a differentiation in the product name?

Do you know whether ChangDa has offered or currently offers domed LEDs?

As far as I know, they have not released a dome emitter, it may just be a reserved name.

1 Thank

Awesome! That will be a hell of a led to test!

BTW, just like you suspected on reddit. I searched and found that both the SFH43 and SFH74 on KD are actually SFH43. The early versions came in larger packages, but later they were changed to smaller 7070 packages.

2 Thanks

Thanks for your research to get more clarity onto these Chinese emitters :+1:

1 Thank