3 Watt LED Bike Light Experiments
May, 2008, Rev d.2
Michael Krabach


Synopsis of Experiments
Prototype 1 - Resistor Controlled Prototype
Prototype 2 - Triple Cree with large heat sink
Prototype 3 - Double Cree with Individual regulators
Prototype 4 - Single Cree and regulator
Prototype 5 - Single Cree and regulator
Prototype 6 - Triple Cree revised heat sink
Prototype 7 - Triple SSC in C Mag-lite Head
Prototype 8 - Triple Cree in D Mag-lite Head
Prototype 9 - Prefabricated 3 LED Mag-lite Head
Prototype 10 - Red flasher with Wide Optic Lens
Prototype 11 - Yellow flasher no Optics
Prototype 12 - Red flasher no Optics
Prototype 13 - Red flasher two LEDs
Prototype 14 - Triple Cree with 8 deg Optics
Prototype 15 - Red flasher two Cree LEDs
Prototype 16 - Red flasher two LEDs Clear Case
Prototype 17 - Auto 12 v clearance light
Light and Beam Measurements
Conclusions, Recommendations, and Further Speculations
Summary Table and Parts Sources

Prototype 12 Red Rear Multi-mode Flasher with no Optics

After the success of prototype 11 resulting in a weather sealed unit, I decided to install the components of prototype 10 in a VistaLite case. With only one AA battery used, the bottom battery compartment was open for the regulator installation. This made the unit easier to construct. The red LED (DX SKU-1776) star base is too large to fit in the center of the heat sink, with a screw being in the way. Also there must be room to insert the AA battery. This required trimming the LED star base. I used a grinding wheel very carefully. Tremendous heat is generated grinding the aluminum star. I monitored the heat with my fingers while holding the star against the wheel. When it got uncomfortable to hold, I stopped and let it cool. The process goes very slowly. The result is seen in the third photo below.

When the grinding of the star is finished, you must be very careful that small burrs from the grinding do not short the LED to the base. The fourth photo above shows the thin line of the circuit foil in on top of the insulating layer of the star. I used a fine file and razor to clean up the surface, then examined it with a magnifying glass for cleanliness. After trimming, the LED star fits in the center of the case. The LED star was bonded to the heat sink with Arctic Silver Thermal Adhesive as seen in the photo below.

The DX SKU-7882 regulator board is similar to the KAI SKU-3151 except that there is a small bridge that must be connected to work properly. The photo below shows the back of the circuit board with the bridge in the lower right corner of the board next to C1. A small tab of solder works well. On all the small tight soldering I use a 9 watt soldering pencil.

The trim pot on the regulator board must be adjusted as in prototype 10. In this case I just adjusted the trim pot for the minimum brightness, the voltage across the LED was 1.4v. This regulator has 19 modes (I counted only 17 modes) in three groups as in prototype 10. Using just the clicky switch was easy, just as it was in prototype 11. Practice is required to become familiar with changing groups and modes. It is not too hard to lose modes with a sloppy click when turning on the flasher. The modes and groups are as follows:

(Directions to configure modes and groups.)
(1) Low, Med, High, Fast strobe, SOS
(2) Low, Med, High
(3) Low, Med, High, Fast strobe, Police strobe, Med strobe, Slow strobe, Beacon strobe, SOS

The next problem was fitting the L2 optics within the case. I found that there was not enough clearance to fit the lens under the case cover, as seen below in the center photo. The only option was to bore a hole in the red case cover to allow the L2 optics to pass through. But this would upset the waterproofness of the case and interfere with changing the battery. When I examined the beam pattern between prototype 10 and 11, I decided, in the interest of simplicity, to leave off the L2 optic and diffuser cover. This resulted in a unit that looked exactly as the original VistaLite as seen in the last photo, and still worked quite well with a broad beam that was quite bright and attention getting.

This unit draws 62, 569, 820 ma on Lo, Med, and Hi for battery current on the steady modes. Across the LED it is 1.4v, 1.7v, 2.1v for Lo, Med, and Hi for steady modes. On police strobe, which uses the Hi level, it will last for about 6 hours with freshly charged NiMH batteries. When the (no load) battery voltage (not LED voltage) reachs about 1.1v the flashing will stop and the LED will stay on steady. The light is still bright enough to be useful for some time, ( not timed) but will slowly dim down in direct drive mode.