# Switching Off That Damned Warning Light!

## After we replaced some of the brake lights with LEDs, cancelling the action of the brake light warning lamp took quite some doing.

### by Julian Edgar

Click on pics to view larger images

In the past we covered the conversion of a centre high mount brake light from filament lamps to LEDs ["Hi-Po LED Brakelight Upgrades"]. The job was performed on my Lexus LS400, and in terms of the increase in light intensity and much faster turn-on time, it was (and has continued to be) a complete success. However, there was one downer: the reduced current draw of the LEDs meant that the brake light warning light on the dash was permanently illuminated. This warning light is designed to show when a brake light is broken, and it thought that two brakelights were broken all of the time!

There's no particularly easy way around this, other than fitting one or two extra brake lights somewhere so that the total electrical load remains the same. I didn't particularly want to do that, so I intended to use some high power resistors to dissipate the same amount of electrical power. That way, the brake warning light circuit would see nothing wrong - until a brake light actually did blow.

But rather than go off to an electronics shop and buy some of the resistors, I kept my eyes open. I pull apart lots of discarded consumer goods and salvage the parts, and I was waiting for just the right type of resistor to turn up. Both microwave ovens and photocopiers use high wattage ceramic resistors, but despite my best intentions, none of the right value could be found. So in the end, I did go to a shop.

Experimentation had shown that I needed to add a load of about 30 watts to the LED brake light if the brake warning light was to stay off - that is, the resistors would need to dissipate 30W. But what resistance should they be? Assuming a voltage of about 12V is available (that takes into account the voltage drop through the wiring), the new load would need to draw about 2.5 amps (30 / 12 = 2.5). The resistance can be worked out by dividing the voltage (12) by the current draw (2.5) - the answer is 4.8 ohms.

Easy, huh. "I'd like a 30 watt, 4.8 ohm resistor, please." Trouble is, nothing like that is readily available. Instead I used four 10-watt 5.6 ohm resistors. These were wired in a series/parallel arrangement - that is, each pair in parallel (giving 2.8 ohms resistance) with the two pairs wired in series (2.8 + 2.8 = 5.6 ohms). That way, the resistance stayed around 5 ohms but the power handling went up to 40 watts.

However, I didn't start off like that. Instead, I initially bought two 10-ohm, 10-watt resistors. (Wired in parallel that gave a 20 watt, 5 ohm combination.) I knew that I'd be working them fairly hard (remember, I needed to dissipate around 30 watts and this combo was good for only 20 watts) but I figured with a really good heatsink they'd be no problem. I used two 90 x 50mm heatsinks, each with multiple fins about 20mm long (these were salvaged from an old car sound amp). The two ceramic resistors were sandwiched between the heatsinks, with the whole assembly supported on a thick aluminium bracket, providing even more heatsinking as well as making the mounting easy.

I bolted the device in the boot, wired it in parallel with the LED brake light, started the engine and then wedged the brake pedal so that the brake lights were on continuously. The dash warning light stayed off (a good start) and I walked away. I figured that if the temp of the heatsink was (say) only warmish to touch after 15 minutes, that would be a good test.

But after just 10 minutes, the heatsink was bloody hot! Way hotter than I wanted something inside the boot. Sure, having the brakelights on for 10 minutes continuously isn't a very likely scenario, but in a long slowly moving traffic jam, it's possible. (The LS400 is an auto.)

### Start Again

There were obviously some major design problems - starting probably with the fact that I was over-rating the power of the resistors. Time to start again - and this time to do it properly. That was when I put together the four resistor, 40-watt series/parallel combination. But even with the extra power handling of the resistors, I still needed to dissipate all that heat.

I went to my box of salvaged heat sinks and found a beauty of a heatsink. From an old domestic stereo amplifier, it was 170 x 70mm, with fins 50mm high. It would fit neatly in a recess to one side of the spare wheel, under the boot floor. However, there was a problem. Heatsinks like this should be mounted so that the base of the heatsink is vertical, creating lots of paths for air to flow up past the fins. I had to mount it with the base horizontal, so the convective efficiency of the heatsink would be diminished. And I was buggared if I was going to build a third version!

What I needed was a fan... I took a look through my salvaged fans and found one that was ideal. A crossflow design (they're the type that use a long, cylindrical impeller), it was equipped with a brushless 24V DC motor. It had come out of an old printer. While nominally 24-volt, the fan worked fine (albeit more slowly) on 12V. The four high-power ceramic resistors were sandwiched between the heatsink and an aluminium plate on which the fan sat. Supporting the whole assembly was a long aluminium plate, which was bolted to two threaded studs already projecting into the boot space. The fan was wired to the brake light power feed (ie whenever the brakelights are on, so is the load and its fan) and then it was time for The Test.

This time, it worked beautifully. With vastly more heatsinking (and heat dissipation) capacity than the earlier design, the heatsink grows only warm to touch, even after 15 minutes.

My car might be the only Lexus in the world running around with a fan-cooled, resistive load in the boot (especially when hanging a light bulb in there would have done the same thing!), but I am happy that I have something durable and well-engineered.

Lot of trouble just to switch off a dash warning light, though...

 The Dollars and Sense If I'd been paying new prices, I would never have taken the approach that I did. But at the best estimate that I can make, the total cost of the components was less than AUD\$10. The resistors - the only parts bought especially for the project - cost \$6. The heatsink came from an amplifier I bought from a dump shop for \$1.50. Taking into account the other bits I salvaged from it, the heatsink probably cost about 50 cents. Buying it new would have set me back about \$20. The crossflow fan came from a printer bought by tender for \$6. Again taking into account the other bits I salvaged, the fan probably cost \$2. New, these fans are worth megabucks - eg \$120 is the list price from one supplier. The aluminium I bought by the kilogram from a scrap metal dealer - say \$1 for this much. The nuts and bolts I bought in a packet from a discount shop - say 10 cents for the ones used here. The wire came off a reel I bought at a garage sale - a few cents for this much. The cable ties and heatshrink I bought from discount electronics stores - another few cents. The pop rivets to hold the fan to the baseplate? A few more cents. As you can see, if you keep your eyes open and are prepared to dive inside goods other people throw away, there is a multitude of quality parts available for nearly nothing. When I salvaged the fan and the heatsink I had no idea that they'd end up in the boot of my car...