Hello all,

I’ve been lurking for several months, marveling at the collection of intelligence displayed here. I am also amazed at the friendly manner in which questions are answered, and lessons are taught. The willingness to teach, rather than shove opinionated theories down a poster’s throat – is simply not seen anywhere else. You should all be very proud of this forum.

My question involves the retrofit of el-cheapo Chinese LED flashlights (Harbor Freight), to work as headlights for my 5 year old grandson’s Gator. I have used these flashlights as-is for many weeks… but it has become a PITA to change batteries because the lights were left on. The flashlights are a bit difficult to reach, for turning off & on, as well. They are, however, the perfect size for mounting behind the previously fake headlight lens, and at $7.95 each… I can’t build and house a similar unit for anywhere near the price. As I have installed a pair of 12V 5AH SLA Batteries, in parallel, to operate his other “accessories”, controlled by a key switch he uses religiously to operate the drive system, I decided to try and operate the headlights from the same source.

Digging around the shop I found several NTE 1929 3-pin adjustable regulators, and the typical box full of assorted resistors (I swear I’m going to get more organized one day soon). I ASSumed that I remembered how to hookup a simple regulator… but obviously I don’t – or I wouldn’t be here asking. NTE offers no application data, and I don’t deal with component-level electronics enough to know when I can safely substitute “typical” ancillary components from other devices (ie: LM317), and use their application data. My degrees are Electrical… not Electronic.

Flashlights for headlights (2) – 32 White LEDs each, connected (I ASSume) in parallel
Powered by 3 AAA batteries in series (4.5V).
Operating voltage when running on the AAA batteries = 3.82 V (Off = 4.54 V)
Operating current when running on the AAA batteries = 340 mA (10.6 mA per LED ???)
* These things are BRIGHT for only 10mA - maybe the array is multi-level and they are NOT all in parallel? *

Regulator – NTE 1929 3-pin adjustable in a TO-220 case, 5 to 35V input, 1.2 to 33V @ 3A output, mounted on 2 back-to-back Heatsinks
Vref=1.2 to 1.3V, 1.25 typ. I Lmin=0 to 10mA, 3.5 typ. P D=15W
http://www.nteinc.com/specs/1900to1999/pdf/nte1929.pdf

Desired supply – 12V 10AH SLA Battery with the possibility of the lights being turned on while the battery is being charged (impatient 5 year old).
Charger output = 13.8V at 40% battery discharge state.

I am unsure whether I need voltage or current regulation with this setup. I do feel I’ll need some input filtering as the wire runs are in excess of 6 ft, crammed in small wireways with high current conductors, and the source supplies a water pump (to squirt ME with), a red strobe, a siren (to annoy everyone), and several auto-style control relays. In my previous attempts, I did see some noticeable oscillation in the regulator – so I feel output filtering is also in order.

I have searched this forum for similar applications, and can find nothing close enough to be comfortable using. I don’t have to use the NTE 1929, but since a dozen or so were just lying around… by golly they just aught to work! I have a reasonably complete test equipment inventory, and can check anything you wish. I’m just too rusty and dumb at this to complete the project without some serious hand-holding.


Sorry for the long post, and thanks in advance for your help

 

Did you know that I use one li-on cell rated at 3.7V to light up 50 white led's in parallel for about an hour.
Try it. No need for resistors or anything. bright white LED's and 3.7V lion's are the perfect combination.
But You will need a regulator if you using 12V SLA's.
U don't need filters for DC operated LED's thru batteries but Just use thick wires so that it will not drop too much voltage.
White led's are commonly used in parallel at 3V to 3.7V max. Using regulators is efficient than using resistor but is a bit expensive.

Rifaa

 

Did you know that I use one li-on cell rated at 3.7V to light up 50 white led's in parallel for about an hour.
Try it. No need for resistors or anything. bright white LED's and 3.7V lion's are the perfect combination.
But You will need a regulator if you using 12V SLA's.
U don't need filters for DC operated LED's thru batteries but Just use thick wires so that it will not drop too much voltage.
White led's are commonly used in parallel at 3V to 3.7V max. Using regulators is efficient than using resistor but is a bit expensive.

Rifaa

The Li-On cells have an internal resistance, which you are depending on for this application. All batteries have this internal resistance. If you use a voltage regulator the internal resistance is much lower, it is how they are designed, so the odds of doing damage to your LEDs go way up. The only good thing I could say about the design I think you're talking about is the cascade failure I described in my article LEDs, 555s, Flashers, and Light Chasers wouldn't happen, but that's about it. A good LED costs around 50¢ to 85¢, a resistor around 3¢. To me the economics just makes sense.

 

The Li-On cells have an internal resistance, which you are depending on for this application. All batteries have this internal resistance. If you use a voltage regulator the internal resistance is much lower, it is how they are designed, so the odds of doing damage to your LEDs go way up. The only good thing I could say about the design I think you're talking about is the cascade failure I described in my article LEDs, 555s, Flashers, and Light Chasers wouldn't happen, but that's about it. A good LED costs around 50¢ to 85¢, a resistor around 3¢. To me the economics just makes sense.

Sheeeesh!! I am stupid.
Batteries have internal Resistance, I knew that, but it never occured to me to apply it. I am being ignorant..this is not good.
Still I did not say any thing wrong . Did I ?
Bill..you are a life saver. I'll tell you about it later on

Rifaa

 

Don't sweat it, brain farts are part of living, just like the other end.

Which is to say, I've had my share.

 

AAA alkaline cells that are made in China and are shipped on a boat for months have an internal resistance that will prevent damage to parallel LEDs without an additional resistor.
But a name brand battery made locally and recently will have a very low internal resistance that might instantly blow up some paralleled LEDs.

Your LEDs need to have their current limited, not their voltage.

 

Which brings us back to that 3¢ resistor. A simple 4.5V regulator, and a whole bunch of resistors (one per LED).

 

The NTE1929 seems to match the LM350 3A positive adjustable regulator quite closely.
LM350 datasheet: http://www.national.com/ds/LM/LM350.pdf

As AudioGuru mentioned, the design of the LED lamp is depending upon the internal resistance of the batteries to limit the current through the LEDs.

One item that hasn't been mentioned thus far is the very real possibility of damage to vision by looking at high-intensity LED light for more than a moment or two. Children tend to stare at things in fascination. It would be a real tragedy if your youngsters' vision were permanently damaged by gazing at those headlights for too long.

How bright is too bright, and how long is too long to look, is not within my realm of expertise.

The safest thing to do would be to replace the LED lamps with incandescent side marker lamps from a junked automobile. These were designed to run on 12v anyway, so connections would be easy and straightforward; just hook them up after the "ignition switch", and optionally add an on/off switch.

 

Sorry, I guess the post was too long to actually be read.

Sarge,
Thanks for the eyeball safety concerns, but - as stated - the LEDs are behind a fake factory lens, hence diffusing the light. Current drain on the battery was why I did not want to use bright incadescent lamps. I do appreciate the cross-reference to an application data sheet... guess I'll have to take from there.

 

You addressed my safety concerns, so we can proceed. As our Forum holds safety issues as priority over all else, I think you can understand why that needed to be cleared up first.

If you just want to limit the current to 340mA:
R1 = Vref/DesiredCurrent
where:
DesiredCurrent = 340mA
Vref = 1.2v to 1.3v, nominally 1.25v
Conversely:
Iout = Vref/R1

So, if Vref is the nominal 1.25v, R1 needs to be about 3.67 Ohms. The larger R1 is, the less current will be supplied to the LEDs.
Power dissipated in R1:
P=EI (Power in Watts = Voltage x Current)
P = 1.25v x .34A = 0.425 Watts. Double it for reliability; you'd need at least a 1W resistor.

This is using the LM350 as a current regulator:
1) Input connected to +12v
2) Out connected to ADJ using the 3.67 Ohm 1W resistor
3) +Supply to the LEDs connected to the ADJ terminal.
4) Other side of LEDs connected to ground. (-12v)

Note that most of the power will be dissipated across the regulator itself. You'll need a heatsink.

 

How about hooking the two head lights in series to take some strain off of regulator. I made some single LED flash lights with 3 AA, & 3 N's, no risistors, no failures yet. Just checked one of the AA's, " best if installed by MAR 2003 ", it's getting a little dim. On the other hand just made up a 9 LED lamp for 12V, 3 in series with resistors.