Sauce lightwand help with driver circuit

I have taken the cicuit out of my Sauce Light Wand and I would like to use it to drive a string of 15 sets of LED's. What I am doing is building 15 frosted tubes with an RGB LED set in each that will be driven from the Sauce circuit through power transistors or ?.

I am "always" the only camper in the non-electric area with lights, but I am tired of steady burning LED's. I want to have a sring of 15 home-made Sauce Light Wands that are all controlled from the same circuit.

Can a transistor smoothly shift the brightness of a string of LED's like the Sauce does? Also do I need to use NPN or PNP and what would the circuit look like? I plan on running the Sauce on the AA, but powering the transistor and LED strings from the 12V RV house battery. This way I have all of the modes of the Sauce, but in a string of 15 tubes (my awning is 15 feet long).

Any ideas would be great.


Thanks,

Rob

You have a neat idea. In order to properly advise you - you'd have to post some details about the Sauce's circuit.

I'm suspecting that the circuit is just an IC (microcontroller) with three of the pins wired directly to one LED color each through a resistor and then to the positive side of the battery. If this is the case, the LEDs brightness is controlled by pulsing the leds on and off faster than you can see, with the brightness controlled by how long they are on compared to how long they are off. This is called pulse width modulation. You can perfectly preserve this type of operation with the right circuit.



If this describes your Sauce and you can run off Sauce power, then instead wire each of the three IC pins as follows. The connections below will preserve the intended operation of the light by not inverting the LED's lighting. This assumes you will power your new string from the same voltage that the sauce light runs on (you probably want to upgrade from AA's to D's):
<UL TYPE=SQUARE><LI>Break the connection to the LED or resistor
<LI>wire IC pin to a 220ohm to 1K ohm resistor. A smaller value will keep the transistor saturated and prevent it from overheating.
<LI>connect resistor to base of a good quality PNP transistor that can handle 800mA minimum. You want a high HFE value if you can get it (100+).
<LI>connect emitter of transistor to the positive battery terminal.
<LI>connect collector to LED anode (non flat side of package)
<LI>connect cathode (flat side) to a resistor and the minus side of the battery. The resistor determines how much current you get through the LED. You can add as many resistors and LEDs as you like until you hit the transistors limits.
[/list]

Your string will consist of four wires - one from the minus side of the battery, and one wire from each of three collectors. At each LED site on the string you put a LED + resistor to the minus wire.

If you really want to power the LED's from 12-15V, it's harder. You'd have to connect the pin to two cascaded NPN transistors to prevent inverting the drive signal. The final NPN transistor would connect to the LED's and the +12-15V. In this case, the string wiring is also tougher - you'll have to use 4-7 reds in series with a resistor to +12-15V, maybe 3-5 greens in series, and perhaps only two blues in series -- due to the different voltages each LED color drops (consumes). (reds are ~1.5V, lime greens ~2.2V, blues 3.5-4.5V depending on chemistry. If you get newer true greens that are InGaN you can only put 2-3 in series as they use 3-4V each).


You'll have one of circuit like the above for each color. Run the sauce from it's battery or an LM317 regulator set for the right voltage off your 12-15V battery.

The circuit will generate and radiate a LOT of radio interference. Expect AM radios, CB's, etc to be affected.

FYI the green LED in the Sauce *is* a InGaN device. Treat them the same as you would the blue, unless you want to intentionally use yellowish-green GaP or GaAlP based LEDs in your add-on string.

I think it would be much simpler and more
reliable just to use a potentiometer in
each series string. Alternately, use
a higher rated reostat to control several
strings in parallel. Of course
a standard potentiometer driving the
base of several small transistors
through 5k resistors would
also accomplish the feat of a dc lamp
dimmer, is that all this ic chip does?

--Al

Thanks for all the input php_44? I think what you have drawn is a darlington pair? I have some basic alalog knowledge, but not much. Just curious as to why this setup vs just using a power transistor as a current amplifyer or switch?

I am not sure if the Sauce circuit is pulsing the LED's or if it is acutally varying the current to them. I will have some time next week to do some testing. The LED's actually vary in intensity in order to generate the range of color that you see. The device is very clever and the amazing thing is that it runs off a single AA battery!

I fould that if you hold the button in for 10 seconds or so, it disables the auto shutoff feature. The hidden modes are all located within the strobe mode. Very cool. My kids love the thing an it has been running on the same AA battery for two weeks of pretty heavy use.

I also like the idea of using a voltage regulator to replace the AA battery in my light array. Again thanks.

I was also wondering how a triac or FET would work in this situation. And is there anything I can do about the stray RF?

Rob

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Robl:
And is there anything I can do about the stray RF?<HR></BLOCKQUOTE>

1) If it is a "current controled" system rather than a PWM system, there will essentially be no RF.

2) If it is a PWM, you could perhaps build a low-pass filter on the base of the transistors (if you run the transistors in linear mode instead of saturated mode).

3) Use shielded cable.

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Robl:
Thanks for all the input php_44? I think what you have drawn is a darlington pair? I have some basic alalog knowledge, but not much. Just curious as to why this setup vs just using a power transistor as a current amplifyer or switch?
<HR></BLOCKQUOTE>
You're welcome. It's not a darlington pair. A darlington pair is formed when the two collectors are connected together, and the emitter of the first transistor goes to the base of the second. In the circuit I posted, the first transistor is inverting the signal from the IC, so that the second transistor (which inverts again) will have an output that matches the IC. This circuit will only work if the IC's output is PWM as I explained previously - if the output is a variable analog voltage (which I highly doubt) the circuit will not work. If the output is truly analog - you'd be better off with op-amps.

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Robl:

The device is very clever and the amazing thing is that it runs off a single AA battery!
<HR></BLOCKQUOTE>
Because there is a blue LED in the light - there must be a boost type of power supply in your Sauce light. The supply raises the 1.5V to probably at least 4-5V to power the LEDs. It would therefore be simplest to simply power your new device with either a good alkaline "D" cell, or an LM317 set for 1.5V output. Alternatively, you could attempt to find out what voltage the Sauce light really runs on by measuring the voltage supplied to the IC, then disabling and dismantling the boost supply and using an LM317 to replace the boost supply.


<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Robl:

I was also wondering how a triac or FET would work in this situation. And is there anything I can do about the stray RF?
<HR></BLOCKQUOTE>
You can get some ferrite beads or cores (like the ones on cheap printer cables) and pass your string's wires through them. A triac would not work - they work best on AC - with DC they don't shut off. You can substitute an "N" channel enhancement power Mosfet for the last transistor at greatly increased cost (unless you have some) and no real other benefit. Choose a device with a gate threshold of 5V or so.

Let us know if you tackle this and how it comes out!!