SFT-90X-WS65-M-ENG
2025-03-08
- Type: single die, lateral
- Bin: J7
- Color group: BA (6500 K)
- CRI: 70
- Rated voltage: typ 3.1 V
- Max. Forward current: 20,000 mA
- Max. Peak current: — mA
- Viewing angle: 120 °
- Thermal resistance: 0.3 - 0.6 K/W
- Max. Temperature Tj: max. 150°C
Note: A preliminary data sheet is available, but as I do not know whether this may be passed on, it is not linked for the time being.
The SFT-90 tested here was purchased from Kaidomain in mid-February 2025. It was offered for sale publicly there for the first time. It is very likely that these are pre-production samples. This is the reason for showing the order code for ENG samples in this test. These may differ in performance and series distribution from the mass production samples!
The SFT-90 is a 7.00 x 7.00 mm LED without a silicone dome. As with the SFT-40 or SFT-70, the LED chip, which sits on a light gray substrate, is surrounded by a silver metal frame in which white silicone compound has been poured. This does not cover the 16 bonding wires, but provides additional robustness. The illuminated area is sharply defined. Due to the classic lateral design of the chip, no lateral light emission is to be expected as with most flip chip LEDs. In general, the design of the SFT-90 is very similar to that of earlier Luminus LEDs and is a logical development of the SFT-40 and -70.
The triangle in one corner of the frame indicates the position of the anode (+)
With a 7070 footprint, the SFT-90 is compatible with common boards (only for 6 V wiring). The triangular cut-out in the thermal pad indicates the side of the cathode (-).
The SFT-90 has a clearly defined, bluish-white illuminated surface. It only shimmers yellow-green at the immediate edge. This should not have any effect on the light image. There are no reflections from the frame or the bonding wires.
The illuminated area is 9.6 mm^2 in size.
- at 20,000 mA (official maximum current): 4956 lm @ 3.87 V
- Power at official maximum: 77.4 W
- Efficiency at 20,000 mA: 64.0 lm/W
- Maximum reached at 21.4 A, at this point 5030 lm @ 3.96 V
- Power at maximum 84.7 W
- Efficiency at maximum 59.4 lm/W
Data for 25 °C Tsp (at 85 °C the luminance values are around 13 % lower).
Here the SFT-90 is primarily compared with the SBT-90.2 (from Convoy, unfortunately with an unknown order code). The following statements refer to the latter, unless otherwise noted. The SBT-90.2 will be described in more detail later in a short review.
It is immediately noticeable that the performance of the SFT-90 is lower. The SBT-90.2 achieves a significantly higher maximum current, which may be due to the larger footprint (9090) in terms of heat dissipation. In general, it is not worth overcurrenting the SFT-90. The official maximum current of 20 A is already at the extreme end, which seems feasible with the current samples. It seems sensible not to exhaust the maximum permitted current, but to stay within a range of 16-18 amps.
The SBT-90.2 has 24 (2 * 12) bonding wires, the SFT-90 only 16 (2 * 8). Although those of the SFT-90 appear thicker, the resistance still seems to be higher. The Vf of the SFT-90 is significantly higher. This may simply be due to binning, but the difference is striking, especially as the data sheets indicate a slightly higher Vf of the SFT-90.
The luminance of the SFT-90 at maximum current possible is lower. However, the difference is of little relevance, especially as the luminous flux of the SBT-90.2 is higher, but reflections from the frame/glass cover reduce the luminance somewhat. Due to the ultimately higher maximum luminous flux, the luminance of the SBT-90.2 is ultimately a few percentage points higher.
The beam is perfectly fine. There is virtually no difference between SFT-90 and SBT-90.2, which is why a beamshot of an Emisar D1 with SBT-90.2 is shown here.
Depending on the reflector geometry and focus, a minimal yellowish corona may be present. This effect is more pronounced with the SBT-90.2 due to reflections on the outer frame on the package. In a quick test with a (too small) SMO reflector, there was no yellow ring around the spot on the SFT-90.
This means that the SFT-90 can be used in OP and SMO reflectors without any problems and produces a good beam.
The spectrum offers no major surprises and corresponds to a typical cool white LED with a low color rendering index.
As with many Luminus LEDs, the duv of 0.0071 is well outside the ANSI color space and therefore produces a visibly greenish tint.
However, the big advantage of the SFT-90 is the significantly lower tintshift with increasing operating current! This very annoying effect occurs mainly with SFT-40 and other Luminus LEDs and sometimes results in extreme green tints, especially at low operating currents. The SBT-90.2 also still has this effect, so replacing it with an SFT-90 may well be worthwhile for this reason in particular.
- Ra: 75
- R9: -25
- CCT: 7186 K
- duv: 0.0071
According to this test, the SFT-90 is definitely not an upgrade compared to the SBT-90.2. Compared to the latter, the performance is lower, the Vf is higher and there is hardly any overcurrent capability. The question also arises as to whether the SFT-90 can be used with the drivers already available for the SBT-90.2 with 22 A or more without frying it.
The SFT-90 offers the great advantage of a smaller footprint, which makes it suitable for use in a much larger number of lights, even with smaller reflectors. The price is also significantly lower, which is an equally big advantage for multi-emitter setups.
For users of existing SBT-90.2 lights, a swap with the SFT-90 is generally not worthwhile. The LED chip is identical, as are the performance and luminance (within the limits of measurement accuracy and series variation).
It is possible that performance will increase significantly in the future, especially as the SBT-90.2 may be EOL’d in the medium term and completely replaced by the SFT-90. In any case, more and more lights are likely to be equipped with the SFT-90 when mass production starts, if only because of the lower price.
Thank you for reading this test. 
