LED Current Limmiter

Status
Not open for further replies.

Frosty_47

New Member
Hello Folks,

I want to use a 2.8 Amp 10 Watt LED (Digikey part number: 897-1021-1-ND) to make a bright flashlight that retails for over $200. I need your help in designing an efficient and adjustable current limiter. Perhaps some form of PWM control to limit the current ?

Thanks
 
How are you going to heat sink it?

Does it matter if you use a proprietary IC or do you have to use common components?

This site contains a few ideas on the discrete component front.
2-transistor Black Regulator

What sort of batteries are you using?

I'd recommend four D cells and if you're going with rechargeable then you need real D cells which will have a capacity of at least 8500mAh, not AA cells with extra padding.
 
You'll need a constant current driver. Most manufacturers sell constant current ICs, which could make a 2.8A limiter with only a handful of support components. Have a look at what Maxim and Allegro have to offer.

A
 
can I assume that the P channel FET can be replaced by any FET of similar characteristics ? I have plenty of IRF9540 parts laying around, how is the current regulated ? is that diagram for a fixed current or can it be varied to suit different leds ?
 
Last edited:
so what parts control the current ? I presume that there is a resistor whos value controls the current setting ?
 
Yes, the resistor controls the current setting but it's not very practical to vary a resistor which is carrying several amps because it would require an old fashioned rheostat.
 
well I was thinking for permanent seting of current but making it usable for various currents, so if i do not require 2.8 amps (as my leds do not) I can make that one time change and run my leds off it too. which value is it ? I guess the 0.33 ohm one ?
 
The voltage across the resistor is equal to Vbe, roughly 0.7V. The current can be calculated using Ohm's law.

You could use a rotary switch with different resistor values.

There are other similar methods using a comparator described in the thread I linked to. These are better because a pot. can be used to adjust a reference voltage which can provide a greater range of adjustment.
 
yea I'd probably go with the comparator method, microchip did an ap note on using a pic to make a small current regulated switch mode supply for leds and had the decency to provide the initial diagram using a comparator rather than a pic, the advantage of the pic was known frequency and various programmed light levels etc.
 
yea I'd probably go with the comparator method,
That's probably the best solution.


Except they made a mistake with the inverting and on-inverting inputs, the thread explains all.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…