Dimmable bulb brightness and power

Index, feed.

After having acquired a few Zigbee-controlled dimmable lightbulbs, I started wondering if the relationship between set brightness and power consumption was linear.

Measurements

I used a TP-Link HS110 smart plug to measure power. After sending Zigbee2MQTT a message to set a bulb to a given brightness, I waited three seconds before measuring power. Each of two RGB bulbs (Ikea 800lm, Philips Hue 1600lm) was tested on three color temperature settings (coolest, neutral, warmest) and 256 requested brightness values (0 .. 255).

Observations

Common to both bulbs:

• Power consumption is highest with the “neutral” color temperature setting. Presumably, so is perceivable brightness.

Ikea:

• There are a few outlier points at low brightness, where I measured 0W. Perhaps this is a limitation of the smart plug at low power levels, or perhaps the bulb exhibits irregular power consumption patterns.

• The highest measured consumption was 9.35W, a bit above the value rated by Ikea; 8.9W.

Philips:

• There are singularities. The measured consumptions drops as requested brightness increases, at different points for each color temperature. Since I have not tried reproducing this, I can’t tell whether it is a measurement issue (perhaps some other device on the same circuit turned on, causing a voltage drop and affecting the bulb or smart plug).

• The highest measured consumption was 12.42W, notably below the value rated by Philips; 16W.

Limitations

I have not measured light output, which would be a more relevant metric to compare to power consumption than the requested brightness value on an arbitrary scale.

www.kurokatta.org

David Haguenauer <www@kurokatta.org>
$Id: dimmable-bulb-brightness-vs-power.malblog,v 1.4 2024/06/17 18:54:41 homer Exp $