Fiber View

Fiberglass Illumination Test Fixture

You may have seen at one point or another someone design an PCB with LEDs on the backside and a void in the copper layer on the opposite side. The effect is that the light from the LED will travel through the fiberglass material of the PCB and you can see it glowing from the other side. This got me thinking, just how far can the light from an LED travel through a PCB? So to do this, I designed a very simple test fixture.

• Fixture Specs:

  • Input voltage - Designed for +5V

  • Number of LEDs: 10

    • 5 flat-edge side-mount

    • 5 rounded side-mount

  • 10 current limiting resistors (2 of each):

    • 442 ohms (5mA)

    • 191 ohms (10mA)

    • 120 ohms (15mA)

    • 86.6 ohms (20mA)

    • 68 ohms (25mA)

The fixture was designed to the constraints of 100x100mm so that it would qualify for PCBWays' 5$ service. in this space is actually two boards, one that hosts the LEDs with series resistors and places to solder power & ground connections, and the second PCB is a ruler of sorts. The side with the LEDs has a white solder mask cover it, which I thought might help the light reflect a bit better than other colors. The ruler side is only fiberglass with no solder mask and only a minimal amount of copper to act as markings to take measurements from. One side of the ruler is flat which mates with flat side mount LEDs, and the other has notches cut out to mate up with rounded side mount LEDs. I wanted to see if one was better than the other.

I was pretty surprised by the results, to be honest. As expected, there was a direct correlation between the LED current and how far the light from it went into the PCB. The photos are a bit exaggerated as to how

What is the use for all of this? Well I have a specific application in mind. There is a pretty obvious hint on the PCB what I intend to use this effect in, but that’s all I’m going to say about it :).