Author Topic: Voltage controlled LED drivers.  (Read 6537 times)

0 Members and 1 Guest are viewing this topic.

Rembrant

  • Newbie
  • *
  • Posts: 5
Voltage controlled LED drivers.
« on: May 02, 2007, 03:42:47 AM »
 You guys are amazing. I simply interred into a discussion that it seems you can't handle.

First, I have nothing to hide. my name is luke lucent, I live in Florida. I attended Pine Forrest High where I graduated(1999) with an B average. I later, attended the University Of West Florida where I graduated(2004) with a B average. The real kicker is? I actually paid for it my self, with a real job. Second, I am not so arrogant as to claim I am the MASTER of anything. Third, I never was rude or disrespectful to anyone. In fact, I didn't just burst on to the scene proclaiming anything. Perhaps it is true that some of you have become over emotional and broke out the instant blinders.

  I'll get right to the point. You are teaching the wrong information about LEDs. A current controlled LED driver will simply not do the job when powering an LED. It has a specification for a reason. The manufacturer of said product actually knows more about it than you all. Fourth,(I count pretty good too) The Point of this master class is what? to prove that you can power an LED with a current driver. You sure can, but efficiency and product life will suffer when you begin to drive the device with more voltage than it was rated for. Thats why it has a rating at all.


The truth about the LED is that if powered by a regulated source at the manufacturers voltage specification it will draw a predictable current and produce a predictable amount of light. If it doesn't, it is defective. It is true that if the voltage is steady the current is steady.


The benefits of an all parallel device are evident. Easy replacement, being the most notable and the fact that if one goes out, you don't lose a whole string. So now we have got all of the LEDs powered Correctly and we have maximized the use and regulation. All without an overcomplicated circuit that only regulates the current and forgets about the voltage.


Now what is left? I would suppose that we want our light to be efficient as well. Why not design a PWM circuit that will power the voltage regulated Driver and regain some of the power lost from a linear voltage conversion? Ever heard of a class D amplifier or maybe a computer power supply.


 To you all I am an idiot. So you can ready up on such dumb ideas from other "Posers" here www.rebelwolf.com/essn/ESSN-Jun2005.pdf

and here www.elecdesign.com/Articles/ArticleID/9254/9254.html

Or some other interesting things about PWM and LEDs that only Posers can get to by searching here http://www.google.com/search?q=LED+PWM+NASA

hell even some of us posers are directly controlling LEDs with PWM and making them brighter. Certainly, everyone at NASA is a huge poser.

« Last Edit: May 02, 2007, 03:42:47 AM by (unknown) »

BruceDownunder

  • Full Member
  • ***
  • Posts: 159
  • Country: au
Re: Voltage controlled LED drivers.
« Reply #1 on: May 02, 2007, 05:11:23 AM »
Rembrant,,,


I sort of got some interest in the middle of your posting ,,, the first paragraph must have been when you were finishing off your, "mrs Palmer and her 5 daughters act" ,,


anyhow ,most of us here are fairly stable ,,so keep in touch ,,you might make the grade..


fair dinkum ,,you are a first class wanker.


(I'll surely be banned for this,at 62 who gives a $#|+)


Sorry Admin--I've read this post for an hour and I'm furious


bruce

« Last Edit: May 02, 2007, 05:11:23 AM by BruceDownunder »

ruddycrazy

  • Hero Member
  • *****
  • Posts: 517
Re: Voltage controlled LED drivers.
« Reply #2 on: May 02, 2007, 05:38:19 AM »
OK Rembrant I think what Bruce said is the sentiment of the majority of us now just to shoot your theory on led's being voltage controlled heres a link to a Dallas application note where they state


 'LEDs are current-driven devices in which the light output depends directly on the forward current passing through them.'


http://www.maxim-ic.com/appnotes.cfm/appnote_number/1804/CMP/WP-37


Now do you think that a company that makes led's would lie to their customers


I think NOT


So if you can't be nice to the members of this great forum do us all a favour and dont bother coming back here


 Just my 2.2 cents worth and Admin's if your going to ban Bruce for fessing up what we all think that will be a dark day on feildlines.


Bryan


.

« Last Edit: May 02, 2007, 05:38:19 AM by ruddycrazy »

dinges

  • Hero Member
  • *****
  • Posts: 1294
  • Country: nl
Re: Voltage controlled LED drivers.
« Reply #3 on: May 02, 2007, 06:37:40 AM »
"The truth about the LED is that if powered by a regulated source at the manufacturers voltage specification [...] "


Perhaps you've missed Amanda's explanation that there IS no set voltage; it varies, not only per batch but also by temperature. And that current varies a lot as supply voltage varies a little.


"It is true that if the voltage is steady the current is steady.". Yep. But it's a big 'if'. If temperature of the environment varies, Vf varies a little and If will vary a lot more. As the LED heats up its Vf and thus If (in a voltage source) will vary. In a current source, If will stay the same no matter the temperature, Vf of the LED, etc.


There's little doubt in my mind why others, with more electronics knowledge than I have (unlike you, I'm just self-educated on the topic) don't bother to answer you. Your statement is so wrong as to be looking like a troll post.


In fact, the way you behave in here (when newbies start throwing tantrums, I immediately assign the 'troll' label to them; call me a bigot), I think you are.


By all means, power your LEDs by a voltage source. But if you start exhibiting incorrect information in here, expect to be criticized. 'Truth' is written with a capital 'T' in here.


'Why not design a PWM circuit [...]'


I'd say, put your money where your mouth is. Amanda has given us some simple circuits; why don't you give us a much better one, a PWM driver for the LEDs?

« Last Edit: May 02, 2007, 06:37:40 AM by dinges »
“Research is what I'm doing when I don't know what I'm doing.” (W. von Braun)

ghurd

  • Super Hero Member Plus
  • *******
  • Posts: 8059
Re: Voltage controlled LED drivers.
« Reply #4 on: May 02, 2007, 06:39:04 AM »
I go through a LOT of LEDs.


"It has a specification for a reason. The manufacturer of said product actually knows more about it than you all."

LED data sheets are like toilet paper, only intended to be used once, then thrown away.


"The truth about the LED is that if powered by a regulated source at the manufacturers voltage specification it will draw a predictable current and produce a predictable amount of light. If it doesn't, it is defective."


Most of my data sheets for white LEDs say 3.6V and 20ma. One says 3.6V and 15ma.

If we find an LED that meets both those numbers at the same time (or even close), it goes straight into the trash.  It probably doesn't happen 1 in 5,000.


Maybe they are sending me 99.98% defective parts. Thank goodness.  

I would not want LEDs meeting the data sheet specifications.


Don't let us stop you.  By all means, build a white LED bulb with 50 parallel LEDs being fed 3.6V, then run it 24/7 for a few weeks.

G-

« Last Edit: May 02, 2007, 06:39:04 AM by ghurd »
www.ghurd.info<<<-----Information on my Controller

TomW

  • Super Hero Member Plus
  • *******
  • Posts: 5130
  • Country: us
Re: Voltage controlled LED drivers.
« Reply #5 on: May 02, 2007, 07:54:17 AM »


 Just my 2.2 cents worth and Admin's if your going to ban Bruce for fessing up what we all think that will be a dark day on feildlines.


Bryan




Never happen...


jeeze, it is always the young pups fresh from the mint who know it all.


I cannot condone profanity, but that word is common on the media here so that bit is a grey area unless it gets too prolific.


Call it as you see it, guys. but remember its a family type board.


Cheers,


TomW

« Last Edit: May 02, 2007, 07:54:17 AM by TomW »

Bruce S

  • Administrator
  • Super Hero Member Plus
  • *****
  • Posts: 5370
  • Country: us
  • USA
Re: Voltage controlled LED drivers.
« Reply #6 on: May 02, 2007, 08:03:14 AM »
rembrant;

  I haven't had time to read what others are going to say and if Admin wants to kick me off for this so be it...

Go jump into traffic, I could careless what you think, come to St. Louis and I'll show you how to give respect where it's do. Amanda did this on her own time so go screw yourself.


Bruce S

« Last Edit: May 02, 2007, 08:03:14 AM by Bruce S »
A kind word often goes unsaid BUT never goes unheard

finnsawyer

  • Hero Member
  • *****
  • Posts: 1565
Re: Voltage controlled LED drivers.
« Reply #7 on: May 02, 2007, 08:39:45 AM »
Which came first, the chicken or the egg?  Which is more important for an Led, voltage or current?  SamoaPower gave us a set of representative curves.  It is clear from the linearity of the current versus voltage curve in the region of interest that the dependence of current on voltage may be represented by an exponential function from which I obtained the relationship I = 1x10^(-7)x 2.718^(3.22xV).  You are invited to try your hand at fitting the curve.  The importance here is that we may differentiate the current with respect to voltage and evaluate it at three volts.  This gives us dI/dV @ 3 volts = 1x10^(-7)x3.22X2.718^(9.66) mhos.  Or simply .005 amps per volt.  That is, 5 milliamps per volt.  It is quite feasible to build a voltage regulator that will keep the voltage stable down to the millivolt range, which would allow an Led string to happily perk along indefinitely.


Of course the regulator should also be capable of allowing for the change in Vf due to temperature, but there should be ways to do that as long as the 2 millivolt per degree drop is representative of all the Leds.  For a string of three we would only need to consider a total of six millivolts per degree even though the regulator may power several strings.  This brings us to the nub of the issue.  Is it better to build several current regulated supplies for several strings of Leds or one perhaps more complicated voltage regulated supply that can by itself power several Led strings, and which will continue to work adequately even after one or more 'normal' string failures.  I suspect that for a manufacturer looking to minimize his overall costs, the latter approach would look attractive.

« Last Edit: May 02, 2007, 08:39:45 AM by finnsawyer »