LED test / review - Cree CLP6B-WKW (5000 - 6500 K, low-power 3-channel emitter in PLCC6 format)

LED Test / review EN

Cree CLP6B-WKW 5000-6500 K


Here is a short test of a small PLCC6 emitter that may be suitable for certain applications with 3 separately controllable LED chips.

This emitter was purchased from Mouser in mid-2017.

Technical data

Tj 25 °C, If 50 mA
  • Type: multi-die
  • Bin: D0-F0 (11.2 - 22.4 cd)
  • Color group: W4. W5 (5000-6500 K)
  • CRI: unknown
  • Rated voltage: typ. 3.8 V (max 4.4 V) V
  • Max. Forward current: 3 x 50 mA
  • Max. Peak current: 3 x 100 mA
  • Viewing angle: unknown
  • Thermal resistance: 3 x 160 K/W
  • Max. Temperature Tj: max. 110°C

Datasheet can be downloaded here: Datasheet (newest version, Cree)

First appearance

Due to the nature of a low-power LED, it is constructed differently than the representatives otherwise tested here. The LED consists of a white heat-resistant housing, in which the luminous surface and the LED chips are embedded within a circular yellow surface.

The connecting legs protrude from the side of the housing. There is no heat dissipation via a dedicated thermal pad, which limits the performance considerably. Heat can only be dissipated via the connection legs here.

To control each of the 3 LED chips separately, the LED has 3 connections each (anode/cathode).

Due to the asymmetric shape (dimensions 6.00 x 5.00 x 2.50 mm) and the very unusual footprint, this LED is not compatible with common LED boards or accessories. Nevertheless, using this LED in flashlights apart from work lights or similar is hardly advisable anyway, the application areas of such emitters are rather in the area of display backlighting or general lighting tasks.

As a disadvantage, I see the relatively large dimensions, which can be an exclusion criterion in certain space-critical applications.


Technically, this LED is a multi-die emitter, whereby here the luminous surfaces are more reminiscent of COB. Due to the strong color differences and the non-homogeneous luminous surface, the result in reflectors is corresponding, with strong yellow-blue colors in the light pattern on the whitewall.

In any case, an optical system is required which mixes the light completely.

The luminous area is 3 x 0.26 mm² in size. Please note that an exact determination of the luminous area size is only possible to a limited extent due to lateral radiation or shining through of the yellow phosphor.

It is noticeable here that no bonding wires are visible on or directly next to the luminous surface. This is usually the case with LEDs of this type. However, it is unlikely that the LED chip is flip-chip, especially considering the low price.

Power and overcurrent capabilities

Within official parameters, as far as known:

  • at 150 (3 x 50) mA (official maximum current): 42.0 lm @ 3.34 V
  • Power at official maximum: 0.50 W
  • Efficiency at 150 (3 x 50) mA: 83.8 lm/W

It is interesting how the Vf specified by Cree in the data sheet is significantly undercut at the official maximum current.


  • Maximum reached at 0.80 A, at this point 115.1 lm @ 4.74 V
  • Power at maximum 3.80 W
  • Sweet spot at about A ( lm @ V)
  • Power at sweet spot W
  • Efficiency at maximum 30.3 lm/W
  • Efficiency in the sweet spot lm/W

The CLP6B is impressively powerful (measured by its design and the virtually non-existent heat dissipation). The absolute luminous flux maximum is reached at 0.8 A. This is accompanied by an extremely high Vf. Nearly 4.8 V are by far the highest I have ever measured for a white LED.

Nevertheless, the performance is remarkable because of the extremely high Vf and the poor heat dissipation, especially if you consider the low Tj of only 110 °C (normally 150 °C).

Although operation under these extreme currents is not recommended, it shows how conservatively these LEDs are specified in the data sheets in some places. Due to the frequent mounting on non-heat conducting PCBs (in my case the LED was partially soldered on a DTP-MCPCB, which is definitely the best case to assume) and the lack of heat dissipation, I explicitly do not recommend overcurrenting, and if so, then only by a few dozens of mA.


Light quality and use in optics

This LED is only intended for optics which mix the light completely (by means of diffusers or textured lenses). In reflectors, this LED produces a light pattern which has blue-yellow colors and is very similar to this one (in the picture above there is a CXA1304 COB in SMO reflector shown).

Color and light quality

The tint is surprisingly pleasant. With a duv of well below 0, the light appears slightly reddish despite the nominally high color temperature, which is quite usable despite the low color rendering. It should be noted here that the tint can vary greatly depending on the batch and color group, especially since this emitter is only available in very coarse color groups, at least at wholesale.

With an R9 of -20, the rendering of the color red is above the value of about -40 expected for 70 CRI LEDs.

As the operating current increases, the duv drops below the BBL, with the effect becoming less pronounced as the current continues to increase. Only at currents above the possible maximum does the light turn bluish, indicating that the permissible Tj has been exceeded.

  • Ra: 70

  • R9: -20

  • CCT: 6337 K

  • duv: -0,0024


For a PLCC6 LED in a plastic package, the performance is surprisingly good, especially when the extremely high Vf and the missing thermal pad are taken into account. This LED is definitely not suitable for applications where high luminous flux is required, nor should it be used in reflectors. However, if an LED for general lighting tasks with separately controllable luminous areas and low luminous flux is desired, this model may be worth a closer look.


  • High luminous flux (according to the circumstances)

  • 3 LED chips separately controllable


  • no thermal pad, limited heat dissipation

  • Extremely high Vf, but only when overcurrent is applied

  • bulky dimensions (especially height)

  • no High-CRI variants available


Thank you for reading the test. :slight_smile:

Greetings, Dominik

2 Thanks