Multiple LED Torch with Single Driver Circuit

I'm building a dual LED torch - one LED for throw and one for flood. I was going to use two AK-47 drivers, but because both LEDs won't be on at the same time I didn't want to install two drivers. Then I thought of this novel way of replacing the second driver with a diode wired into the first driver. Surely this has been done before but I haven't bothered to look...

This only works with 7135 based drivers and it can't be used with the typical tail cap switch because these switch the battery negative circuit. This only works with switches in the battery positive circuit - like most side switch torches. Of course you will need a multi-position switch (see DX, KD, etc.) The switch I'm going to use works like this - OFF, LED1, OFF, LED2, OFF. (4 total positions)

Be careful of the current rating of the switch, and good luck trying to find that information...

You can add more LEDs by adding a diode and switch position for each one. If two positions are selected at the same time, both LEDs will light but they will split the current.

Also note that if the driver has mode memory that all the LEDs will share the same memory. (the driver doesn't know which LED is connected)

Very nice !

I have in mind something similar with a headlamp with two AA's with plenty of room for two emitters. One Osram warm as flooder and an XR-E with a narrow TIR optics . Also with two drivers in series , one boost and an AK-47.

The idea is putting the emitters in series and with a slide switch short one or the other.

Maybe this weekend I'll do it.

Nice. Where do you get your 7135 chips? do you salvage them from another driver, or is there somewhere to buy them?

I've been looking to nab some LR1108 chips, and the trail goes dead-cold at the chinese boarder. A hobby source of 7135's would give me a new starting point.

Devman, I didn't make the driver. And I don't know where to get the 7135's. The circle in the figure just represents one of any number of drivers that use 2, 3, 4... or 8 7135 chips. These drivers come from DX, KD or others.

Damn, sorry. Got my hopes up... the LR1108 is basically the same chip as the AM7135, only it's designed for 1000mA instead of 350mA

Only problem is I can't find anyone selling hobbyist amounts of it!

Devman, all the info I've found for LR1108 suggests that it is a voltage regulator not a current regulator (like the 7135). I don't think it will work the same way as a 7135. I did some quick searching and found www.crpowtech.com who make a PT4403 which appears to be the same as the 7135. (Don't know where to get these either...)

Catalyst Semiconductor (or ON Semi) make a CAT4101 which is rated up to 1 amp. It seems a bit bigger and more complicated, but this is a good thing. The extra power burned off at 1 amp needs a higher power package. Also it handles multiple series LEDs, has programmable current, allows PWMing, and is in stock at www.newark.com $1.92 each @ qty 1.

I'm not a big fan of linear current regulators. The efficiency is low with a freshly charged battery and their dropout voltage causes the current to drop off before the battery is fully discharged. YMMV. They are simple though... got to admire that!

Devman, I just noticed that you're in Canada. www.futureelectronics.com has the CAT4101TV-T75 for US$1.92 as well.

Hmm. I still think the damn thing is a current regulator. http://www.utc-ic.com/spec/LR1108.pdf says "The Guaranteed Output Current is 1A DC "

I only noticed it because it's available in what appears to be the same package as the AMC7135, making it an easy drop-in replacement.

(Then again, I did compsci, not EE... so I could be reading the data sheet upside down for all I know??)

Yeah its a voltage regulator all right. But at least now you can concentrate on other alternatives.

One problem with most any driver circuit is power dissipation, how do you get the heat out of it? Even a 90% efficient driver will dissipate 1 Watt when driving a 3A XML. And 1W is a lot of heat for a small driver. There is a driver that uses eight 7135's for 2.8A output, that's 2.5W dissipated with a fresh charge. Heatsinking the driver is a must.

Ahwell, thanks for humoring me. Seems I won't be chasing after the white stag.

And now back to our regularly scheduled thread. :P