Use Fatman to test LEDS w/ constant current?

smitty244

Newly Enlightened
Joined
Nov 28, 2004
Messages
115
Location
Toronto, Ontario
Hi, I have a few Luxeons that I want to test to determine their respective Vf. I don't have a constant current power supply at all, and it is a little expensive to get one from what I see. The fatman converter board doesn't need heatsinking and can be set to various current settings, so would it be possible to just use that to provide a constant current and then take the Vf reading from across the LED with a multimeter?
 

MrAl

Flashlight Enthusiast
Joined
Sep 9, 2001
Messages
3,144
Location
New Jersey
Hi there,

Someone else may be able to answer your question about the
fatman, but you should also know that building a constant
current supply isnt hard at all. If you've ever soldered
parts together you can build your own with a wall wart
and a LM317 ic chip. Let me know if you're interested
and i'll post some details.

Take care,
Al
 

wasBlinded

Flashlight Enthusiast
Joined
Dec 14, 2004
Messages
1,222
Location
Oklahoma
Anything that can be set to give constant current at 350 mA (for Lux Is) and 700 mA (for Lux IIIs) will work. Ideally you'll have two DMMs, one measuring current to be sure you are where you want to be, and the other measuring the Vf.
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
smitty,

not to barge in, (I'm sure Al won't mind), here's the basic circuit:

317current.PNG


12V in isn't required, but the '317 in this configuration needs about 3V of headroom, so you'll need more than 6.5V input.

You'll want the '317 in a TO-220 case, and you'll want to bolt it to a heat sink, since it'll be dissipating a bit of power.
 

modamag

Flashlight Enthusiast
Joined
Aug 16, 2004
Messages
2,101
Location
Bay Area, CA
If you're using the LM317 make sure you're using the 1.5A version. Otherwise when you pump up the volume, it pops (no problem) then exposes your Lux the full voltage and that will cost you $10-$20.
 

jtr1962

Flashaholic
Joined
Nov 22, 2003
Messages
7,505
Location
Flushing, NY
Here's another simple circuit you could try:

LED_Constant_Current_Circuit.GIF


For best control use a ten-turn potentiometer and make sure the 1V is derived from a voltage regulated source. Also, use a 1% or better resistor for R1 with a rating of at least 2 watts. You can measure the voltage across R1 with a DMM to give you the current in amps. +V of 5V or so should be enough, and you can use that same 5V to power the opamp. For Q1 use a logic level n-channel MOSFET such as the IRLZ34. If +V is a regulated 5V source then it won't need to be put on a heat sink. However, if you're using a 12V supply for everything then put Q1 on a fairly substantial heat sink such as one from an old Pentium processor.

This circuit as shown uses parts readily available. Ten turn pots can be had for $5 or you can use a regular pot if you don't care about very fine adjustments. The value of the pot isn't critical. Everything else will cost well under $5 and chances are most people already have most of these parts in their toolbox. Use a breadboard to wire everything. Connect the unused op-amp (-) terminals to ground and the unused (+) terminals to the corresponding outputs. This will hold the other three outputs on the op-amp to ground.

Note that this is a linear circuit, meaning that the voltage difference between the LED+R1 and the power supply is taken up by the MOSFET. I have made constant current switching regulators which I used to pump 7 amps through thermoelectric modules but these are quite a bit more complex. Also, the selection of inductors can make a huge difference in efficiency. I got efficiencies of better than 96%. I also wound my own inductors on some toroidal cores I purchased. Although I have no plans to do so, I could use even larger cores, reduce the switching frequency to cut my transients losses, use a MOSFET with lower Ron, and probably approach 99% efficiency. However, a some point the diminishing returns aren't worth the hassle. I personally consider making a switching regulator just for LED testing to be gross overkill.
 

MrAl

Flashlight Enthusiast
Joined
Sep 9, 2001
Messages
3,144
Location
New Jersey
Hello again,

smitty:
evan's circuit looks good. That's just about what i was
going to post also...thanks evan!

Take care,
Al
 

andrewwynn

Flashlight Enthusiast
Joined
Apr 28, 2004
Messages
3,763
Location
Racine, WI USA
if you already have a fatman.. use the fatman :-D.. why re-invent the wheel.. i use my drivers as 'test beds' all the time.

-awr
 

smitty244

Newly Enlightened
Joined
Nov 28, 2004
Messages
115
Location
Toronto, Ontario
Thanks for all the help! I think I'll just stick with the fatman for now with the method mentioned above, and when I'm feeling daring I'll try to build my own. Thanks again!
 
Top