5.0 × 5.0 mm 6 pins top-emitting

HD108

5050 · smd · HD108 · Dongguan Shiji Lighting

SPI RGB 4.5–5.5 V active

Also sold as HD108 5050, 16-bit DotStar

At a glance

Colour
RGB16 bits per channel
Data
SPIclock + data, timing-tolerant
Supply
4.5–5.5 Vno backup data line
Size
5.0 × 5.0 × — mm6 pins · top
Max current
—per-channel not recorded
PWM
—no reset period
Library support
FastLED WLED Adafruit

16 bits per channel rather than 8. The practical difference is at the bottom of a fade, where 8-bit parts visibly step and then jump to black.

Current per channel and the maximum are left blank: no datasheet for this part is held, and the earlier 20 mA / 60 mA figures were a design rule of thumb rather than a sheet’s rating. A blank makes you check; a plausible wrong number does not. The 120 degree viewing angle and 1.7 mm height were unsourced and are blank.

Variants

One row per orderable SKU. Everything not listed is inherited from the part above.

Order codeFaceLensWhite PinsSiliconStatusFirst seenDatasheet
HD108 white clear — 6 — active unknown —

Protocol — HD108

This is an SPI part: the clock is explicit, so there is no fixed bit period. The frame layout is shown below.

HD108 SPI frameHD108 clocked frame: start frame, one 32-bit LED frame per pixel with brightness and colour bytes, and an end frame, plus data and clock waveforms1 · Frame stream (data clocked on CLK)Start frame64 × 0LED 164 bitsLED 264 bits…LED N64 bitsEnd frameclock ≥ N/2 edgeseach LED forwards what it does not keep on DO / CO2 · One LED frame: 64 bits, MSB first1gain 1gain 2gain 3ch 1 16ch 2 16ch 3 161 marker bit, three 5-bit brightness (gain) fields, then 16-bit channels3 · DATA and CLK, first 8 bits of an LED frameDATACLK110101001gain 1, bits 4..0gain 2…data changes while CLK is low,LED samples on the rising edgeno fixed bit period: clock rateis set by the host
Frame layout: HD108-style 64-bit LED frames convention, see the protocol notes for end-frame differences

Compatibility

  • protocol compatible APA102

    Same SPI framing concept, but each pixel carries 16 bits per channel, so an APA102 driver will not produce correct output without explicit HD108 support.

Dating evidence

Nobody announces these parts, so this records what we can actually cite rather than a release date. Library support proves only that the part existed by that date.

  • 2025-10-30 first library support FastLED — first commit introducing HD108 ↗ day

Similar parts

Same protocol or package first, then the same data scheme and colour channels.

Was this page useful?

0 people found this useful