🚀 go-pugleaf

In article <anqolb$pj2@cleon.cc.gatech.edu>, byron@cc.gatech.edu says...
>
>In article <h6hvpu8hnhuag56b3rr5sqlompgmj984dm@4ax.com>,
>JoeBloe  <theguyattheendofthebar__nicepingeh?> wrote:
>-On 4 Oct 2002 12:48:17 -0400, byron@cc.gatech.edu (Byron A Jeff) Gave
>-us:
>-
>->In article <Qrhn9.27982$kd3.80464@news-server.bigpond.net.au>,
>->Grog <shem_an_na@SPAM_ME_NOThotmail.com> wrote:
>->-
>->-"John Woodgate" <jmw@jmwa.demon.contraspam.yuk> wrote in message
>->-news:B7hloTAPDZn9EwGz@jmwa.demon.co.uk...
>->-> I read in sci.electronics.design that Uwe Zimmermann <uwezi@kth.se>
>->-> wrote (in <3D9D5939.CD795C11@kth.se>) about 'Best way to power array of
>->-> LEDs?', on Fri, 4 Oct 2002:
>->-> >You then connect the drain of the transistor to the positive supply and
>->-> >the gate/source to the anode of the first LED in each chain. The 
cathode
>->-> >of the last LED is connected to the negative supply.
>->->
>->-> This is an *extremely* unreliable circuit, because it has no protection
>->-> against the large transients, of both polarities, that occur on vehicle
>->-> electrical supplies.
>->
>->That's an apples and gorillas discussion. Transient and spike surppression 
>->is a completely different issue from powering the LEDs. I didn't even 
attemp
>t
>->to address it.
>->
>->
>->->
>->-> Don't bother with adding transistors just to get rid of a few resistors
>->-> (or maybe add a few more, and MOVs and capacitors and fuses ...). KISS!
>->
>->But it isn't simple if the array isn't bright enough to do the job. It's
>->quite likely that the LEDs will need to be pulsed with higher current in
>->order to give brighter perceptual intensity. Once you start down that road
>->you're going to have to have transistors to modulate/switch power to the
>->LEDs.
>->
>->And frankly an extra 2N2222 and base resistor per string isn't exactly 
rocke
>t
>->science.
>->
>-
>-  Yeah... they're all of what 10 cents a piece?  Hehehe...
>-
>-  DC/DC converter from surplus store for five bucks, drop to five
>-volts, clean instead of 12, dirty.
>
>It really doesn't matter too very much since it'll be the same amount of
>current regardless of the voltage...
>
>-  Two copper rails.
>
>Why?
>
>-  A series resistor in each leg of each LED that gets BUSSED in parallel 
onto
> 
>- the copper rails.
>
>Again why when you can string multiple LEDs per string this saving both 
>components (resistors) and power loss (due to the lower voltage drop.
>
>-
>-  Ten bucks total... maybe.  On a bad day.
>
>I never heard price being an issue.
>
>Any you still haven't dealt with the most important issue, brightness! Here
>are the simple facts:
>
>* The LEDs at continuous max current probably won't be bright enough.
>* More current can be delivered only at a lower than 100 percent duty cycle.
>* This requires that the power to the LEDs to be switched.
>* Ergo you need a transistor. The current limiting is simply a bonus.
>
>LEDs are not about voltage. They don't care about voltage. I've run them at
>2.5V. I've run them at 120V AC. LEDs are current devices and have a brighter
>response to more current. So any LED supply is about current management.


You are absolutely right. This question about Voltage and LEDs just keeps 
coming up time and time again! You'd think that those posing it would take the 
trouble to learn the most elementary basics of what they are trying to work 
with.

To make the principles perfectly clear, LEDs emit photons in an approximate 
ratio of the amount of CURRENT (hear that?...**current!!) that passses through 
them. Voltage isn't even in the equation!! The voltage that appears across an 
LED is a RESULT of the current that you are pushing through it, not the other 
way around!. Like any diode, this voltage (called the forward voltage) tends 
to be relatively constant, and relatively independent of the current. But 
(watch my lips closely, now!) the value of this voltage is basically 
irrelivant! Forget about it! Don't even mention the "V" word when dealing with 
LEDs!

There! Got that off my chest.

Bob.