Rotating Light Function

It is a one-off prototype for a magazine article. My readers have to be able to get the parts on Sunday afternoon in East Undershirt Wyoming from Rat Shack floor sweepings.

Jim

much

there

You can do it piece-wise linear using diodes and resistors in the feedback of an opamp. In this way the gain of the amplifier changes with the voltage and most nolinear curves can be approximated. A similar method driving an led directly with no opamps could use a couple of zener diodes each with a series resistor plus a resitor without a zener all connected in parallel then the combination in series with the led. When the voltage starts up and gets past a couple of volts, the led lights with the current from the straight resistor. As the voltage increases to the value of the first zener, the zener turns on and a new higher current lights the led brighter than it would be without the zener and it's resistor because there are now two parallel current circuits to the led. Similarly, a third zener, which may not be necessary, will come in at a still higher voltage and the led will get even more current because there are now three parallel current paths. The result will be an accentuated peak in brightness like a beacon sweeping by as the driving triangle voltage peaks. I think this is what you are asking for.

Don't give up on the microprocessor solution, there are 8 pin PIC's available for about a buck. Furthermore, everything can be done in there including the ramp up and ramp down without a triangle wave required. It would probably be cheaper as a total solution, require less parts and be less prone to copying and fiddling with. Bob

Suppose it was an LED you were trying to light and you fed that to one side of the LED and had a network on the other side that subtracted a voltage level from the triangle wave (and the LED would have a log-like response). For example, a resistor to ground, and a PNP emitter follower or two with additional inflection points of brightness and series resistors. I think we're still in the 2-bit range.

Problem is that the LED (or an incandescent lamp) are going to have some kind of nonlinear response with voltage/current and that will have to be taken into account.

If you're making toy-like quantities, your brute-force microcontroller idea will probably be cheaper, more repeatable and more reliable.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

You need to light levels. A low level most of the time and a fast peak level.

Feed the LED with a suitable resistor for the low level. Configure a

555 timer with a short output pulse (about 50 ms) and a slow cycle (about .5 second). Feed the LED with output thru a smaller resistor.

Finally, experinment with large capacitors right accross the LED to make the peak a bit of a slope up and down.

--
Luhan Monat (luhanis 'at' yahoo 'dot' com)
"The future is not what it used to be..."
http://members.cox.net/berniekm

I tried your scheme using an LM324 and a handful of parts. Not a very convincing appearance, IMO.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

You can get a tiny11 micro from digikey for $0.54. Not too far off your two bits, and Atmel advertises these for under a quarter in 1k quantities. (two bits is a quarter isn't it?)

But then you probably don't want to mess with getting a programmer, and all the development tool software. These are not terribly expensive, but are definatly more of a pain in the ass than a LM324, if you can figure out the circuit. The piece wise linear diode thing sounds like a good idea, or you could possibly do something with a op-amp and transister to get a log amp.

Ethan Petersen

light, much

peaked,

The

inverse

single

program it

single

the

when there

I suppose a motor, torch bulb and milk bottle top, all pointing at an ORP12 to drive the LED is out of the question? :)

Seriously, why not use a 324 to produce the triangle wave, plus a section of it to act as comparator to hit the LEDs with max i during the triangle peak?

Or... put the led on a motor.

I doubt youll get any kind of decent simulation since the flash off those lights is fairly intense, and leds are nothing like that, unless youre using an ultrabright.

NT

Just make a tiny lazy susan with a hole in the middle, and a reflector on top at 45 degrees.

Cheers! Rich

RST Engineering \\(jw\\) wrote in sci.electronics.design:

If there is a convenient clock signal around, build a little something with 8 output bits (say) exactly one of which is high at any time. A shift register or a counter/demux combination would do. Connect the outputs to the summing point of a driver op amp through appropriate resistors for different brightness during each clock cycle.

The advantage over simpler solutions that have been suggested is that the brightness at each stage is simple to determine an change. If it fits in the environment it may be worth consideration.

Anno

Do it digitally with table lookup.

--
Many thanks,

Don Lancaster
Synergetics   3860 West First Street  Box 809  Thatcher, AZ 85552
voice: (928)428-4073 email: don@tinaja.com

Please visit my GURU\'s LAIR web site at http://www.tinaja.com

For the original 'electronic' route, I'd suggest the following: Triangle wave generator. Signal#1 Comparator off this, generating a 'pulse' on the peak 15% of the triangle wave. Filter the triangle to generate a sinusoid, and have this with perhaps 4* the peak level of the triangle wave. Signal#2. Then have a MOSFET switch, operated off the square wave, to switch between Signal#1, and Signal#2. Feed the output of this into the LED, with a current limiting resistor. This will give an off period when the triangle is below the LED forward voltage (the period when the beam is completely aimed away), then a smooth rise in intensity (as more of the light becomes visible, but still unfocussed), to the comparator switch point, corresponding to the edge of the 'beam', then a smooth rise and fall (corresponding to the sinusoidal waveform, and similar to crossing the beam from the reflector), then the sharp drop to the lower dim level, and a slow fade to nothing. A PIC with a PWM, really would be as cheap.

Best Wishes

Version 4 SHEET 1 880 680 WIRE -64 304 -192 304 WIRE 32 336 32 288 WIRE 48 224 -64 224 WIRE 48 288 32 288 WIRE 48 288 48 224 WIRE 96 448 80 448 WIRE 96 448 96 208 WIRE 112 336 112 176 WIRE 144 192 144 144 WIRE 144 208 96 208 WIRE 144 208 144 192 WIRE 144 224 144 208 WIRE 144 304 -64 304 WIRE 160 176 112 176 WIRE 160 176 160 160 WIRE 192 160 160 160 WIRE 192 160 192 112 WIRE 208 160 192 160 WIRE 208 192 144 192 WIRE 240 64 144 64 WIRE 240 144 240 64 WIRE 240 304 144 304 WIRE 240 304 240 208 WIRE 288 112 192 112 WIRE 288 176 272 176 WIRE 288 176 288 160 WIRE 288 192 288 176 WIRE 304 160 288 160 WIRE 304 192 288 192 WIRE 304 304 240 304 WIRE 384 160 384 96 WIRE 384 352 80 352 WIRE 384 352 384 192 WIRE 400 96 384 96 WIRE 400 96 400 0 WIRE 400 160 384 160 WIRE 400 192 384 192 WIRE 432 64 240 64 WIRE 432 144 432 64 WIRE 432 304 304 304 WIRE 432 304 432 208 WIRE 480 96 400 96 WIRE 480 176 464 176 WIRE 480 336 112 336 WIRE 480 336 480 256 WIRE 544 64 432 64 WIRE 544 144 544 64 WIRE 544 304 432 304 WIRE 544 304 544 224 WIRE 624 0 400 0 WIRE 624 224 624 0 WIRE 624 304 544 304 FLAG 304 304 0 SYMBOL Opamps\\\\LT1014 240 112 R0 SYMATTR InstName U2 SYMBOL voltage 544 128 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 12 SYMBOL res 128 48 R0 SYMATTR InstName R1 SYMATTR Value 1k SYMBOL res 128 208 R0 SYMATTR InstName R2 SYMATTR Value 1k SYMBOL pjf 32 352 R0 SYMATTR InstName J1 SYMBOL voltage -64 208 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V2 SYMATTR Value PULSE(0 12 0 .1u .1u 400m 600m) SYMBOL res 608 208 R0 SYMATTR InstName R3 SYMATTR Value 4.7meg SYMBOL cap 272 112 R0 SYMATTR InstName C1 SYMATTR Value .47=B5 SYMBOL res 464 80 R0 SYMATTR InstName R4 SYMATTR Value 10k SYMBOL res 400 144 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R5 SYMATTR Value 10k SYMBOL res 464 160 R0 SYMATTR InstName R6 SYMATTR Value 82k SYMBOL res 400 176 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R7 SYMATTR Value 100k SYMBOL res 16 320 R0 SYMATTR InstName R8 SYMATTR Value 1meg SYMBOL Opamps\\\\LT1014 432 112 R0 SYMATTR InstName U1 TEXT 48 32 Left 0 !.tran 5 Look at the output of the integrator.