Lm317
Do a search at national semiconductors site for the LM317. In the data sheet they so how to use it as a constant current source. I think it's rated at 40v thou I don't think it would matter because in their fore mentioned schematic it is not connected to ground(regulating current not voltage). By the way, I use this set up to charge nicads...Works good.
Do a search at national semiconductors site for the LM317. In the data sheet they so how to use it as a constant current source. I think it's rated at 40v thou I don't think it would matter because in their fore mentioned schematic it is not connected to ground(regulating current not voltage). By the way, I use this set up to charge nicads...Works good.
Linear Tech has an application note for "Current Sources for Laser Diodes" -- you will find schematics, the entire shooting match for CCS using both linear and switched regulators.
If it is a design where "metrics" are important you will probably want to consider using the Linear LT1533 slew controlled switching chip with post regulation.
by the way, Linear now sells their stuff, fewer than 500 pieces, direct to the DIY'r -- and prices are about 40% less than Digikey -- shipping is via FedX.
If it is a design where "metrics" are important you will probably want to consider using the Linear LT1533 slew controlled switching chip with post regulation.
by the way, Linear now sells their stuff, fewer than 500 pieces, direct to the DIY'r -- and prices are about 40% less than Digikey -- shipping is via FedX.
Sch3mat1c said:Switching regulator, not linear. I guess for efficiency, but 40V supplies usually have plenty of current to spare. And what LEDs take 700mA???
Tim
Luxeon emitters., even the Agilent super-brights can be pushed to 100ma.
depends upon how many candela you want.
millwood said:unrelated questions: do we know if LEDs wear out? assuming that they are working within their spec.
I personally haven't experienced any worn LEDs but others have.
Agilent has some data on "life cycle" on their website -- it's obviously a function of current, but also the heat exposure of the diode -- we are talking tens of thousands of hours at 20 ma, and a few thousand hours at 40ma.
Hmmm
Ya, my bad....I was focused on the word "LED" not "700ma". I gotta agree with Sch3mat1c......700ma is alot for a LED, unless your planning on putting a bunch in parallel. Since you got so much voltage why don't you hook em' up in series. You could daisy chain quite a few LED's on 40v.
Ya, my bad....I was focused on the word "LED" not "700ma". I gotta agree with Sch3mat1c......700ma is alot for a LED, unless your planning on putting a bunch in parallel. Since you got so much voltage why don't you hook em' up in series. You could daisy chain quite a few LED's on 40v.
And by the way...
And another thing, a linear regulator would be just fine because "40V supplies have plenty current to spare". I mean.. designing a 21watt constant current (SWITCHING!) supply for a LED is not worth it, and why constant current anyway? If a constant voltage is put across a LED, a constant current would flow thru it. nuff said.
And another thing, a linear regulator would be just fine because "40V supplies have plenty current to spare". I mean.. designing a 21watt constant current (SWITCHING!) supply for a LED is not worth it, and why constant current anyway? If a constant voltage is put across a LED, a constant current would flow thru it. nuff said.
Re: Hmmm
Just 1 LED.
http://www.lumileds.com/
nine76chris said:unless your planning on putting a bunch in parallel.
Just 1 LED.
http://www.lumileds.com/
Re: And by the way...
This particular LED's forward voltage varies from 3.0V to 4.5V. I put 3 in series (9V to 13.5V)
nine76chris said:If a constant voltage is put across a LED, a constant current would flow thru it. nuff said.
This particular LED's forward voltage varies from 3.0V to 4.5V. I put 3 in series (9V to 13.5V)
Re: And by the way...
it isn't correct to assume that a constant voltage impressed upon a LED will result in constant light intensity. firstly, not all VF's are equal, secondly, the current in a led is also a function of the LED temperature (didn't I point this out a few days ago?).
you can check out the Agilent or Allegro Microsystems websites for reasons to the above.
a 21 watt constant current/constant voltage supply isn't like splitting atoms -- there are even PCB layouts on some of the manufacturer's websites -- and you can have both constant current and voltage control at the same time.
nine76chris said:And another thing, a linear regulator would be just fine because "40V supplies have plenty current to spare". I mean.. designing a 21watt constant current (SWITCHING!) supply for a LED is not worth it, and why constant current anyway? If a constant voltage is put across a LED, a constant current would flow thru it. nuff said.
it isn't correct to assume that a constant voltage impressed upon a LED will result in constant light intensity. firstly, not all VF's are equal, secondly, the current in a led is also a function of the LED temperature (didn't I point this out a few days ago?).
you can check out the Agilent or Allegro Microsystems websites for reasons to the above.
a 21 watt constant current/constant voltage supply isn't like splitting atoms -- there are even PCB layouts on some of the manufacturer's websites -- and you can have both constant current and voltage control at the same time.
Re: Hmmm
paralleling LED's is an invitation to disaster.
nine76chris said:Ya, my bad....I was focused on the word "LED" not "700ma". I gotta agree with Sch3mat1c......700ma is alot for a LED, unless your planning on putting a bunch in parallel. Since you got so much voltage why don't you hook em' up in series. You could daisy chain quite a few LED's on 40v.
paralleling LED's is an invitation to disaster.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Design & Build
- Parts
- switching regulator as a current source