Maker Pro
Maker Pro

Led lights and voltage

If I connect 8 leds @ 1.6v vf onto a power supply producing 12vdc, I
won't have to use a resistor, right? Theory being that the 8 led's
will use .8 volts more than the ps can provide. They won't burn as
bright, but they won't burn up the ps, they won't destroy themselves,
and they won't require resistors?
 
If I connect 8 leds @ 1.6v vf onto a power supply producing 12vdc, I
won't have to use a resistor, right? Theory being that the 8 led's
will use .8 volts more than the ps can provide. They won't burn as
bright, but they won't burn up the ps, they won't destroy themselves,
and they won't require resistors?

Wrong. Why get rid of the resistors? Are they expensive or something?
LEDs "like" to be driven by a constant current source, the cheapest
approximation being a voltage source with a resistor.
 
  Under certain circumstances, even I leave out current limiting resistors.
There is enough inherent resistance on the supply side of whatever is driving
the resistor to 'internally' limit the LED current.

  There is nothing wrong with this.

  However, the OP was *way* too broad in his description of "a power supply
producing 12vdc" and "8 leds @ 1.6v vf".

  The supply is very nice round number, so implies either something thatis
regulated, or, at least has a lowish internal impedance.

  "1.6v Vf" sounds like something that came direct from the spec sheets. In
real life, it means little for this purpose.

The "1.6V" will be the voltage drop across the LED at a particular
current - often 20mA - and a particular temperature - usually 25C.

The forward voltage of a LED - as with any other diode - is roughly
proportional to the logarithm of the current, and decreases by about
2mV/C as the junction temperature rises. At high currents the
resistance through the LED can become significant.
 
H

Henry Kiefer

Jan 1, 1970
0
John said:
To elaborate, they behave somewhat like a zener with a broad knee.
Personally, I think a voltage regulated supply to LEDs is not a good
idea, it's not good design practice and would have to be pre-set to
batches of leds to ensure at least consistent brightness across the
product line.

Like the first reply asked, are resistors expensive or something?

A perfect LED with a perfect voltage source will either burn-out or
stays dark.

What it makes possible in reality to run a LED with a voltage source
without a resistor is just the combined internal resistance of the LED
and the voltage source. This acts as a small current source.

- Henry
 
R

redbelly

Jan 1, 1970
0
If I connect 8 leds @ 1.6v vf onto a power supply producing 12vdc, I
won't have to use a resistor, right? Theory being that the 8 led's
will use .8 volts more than the ps can provide. They won't burn as
bright, but they won't burn up the ps, they won't destroy themselves,
and they won't require resistors?

A power supply labeled "12vdc" could actually be 13V, or something
else. And whatever it is, can change appreciably when it is drawing
current, or has had a few minutes to warm up. And, diode current
changes quite dramatically with small changes in voltage.

There is a small chance you could get away with what you are
proposing, but it's not a very good idea. If you do, at least measure
the actual power supply voltage, at 20 mA load, and then measure the
actual LED current after installing them (and be willing to sacrifice
the LED's for the sake of experimentation.)

Regards,

Mark

p.s. They do get away without adding a resistor in coin-cell operated
LED flashlights. This is because the series resistance of the cells
provides a fair amount of current regulation.
 
J

Jan Panteltje

Jan 1, 1970
0
If I connect 8 leds @ 1.6v vf onto a power supply producing 12vdc, I
won't have to use a resistor, right? Theory being that the 8 led's
will use .8 volts more than the ps can provide. They won't burn as
bright, but they won't burn up the ps, they won't destroy themselves,
and they won't require resistors?

They wont burn at all, as 8 x 1.6 = 12.6, LEDs are diodes, current
will only flow once you equal or exceed the forward voltage drops.
 
J

Jan Panteltje

Jan 1, 1970
0
It's the art of knowing what's skipable..

Exactly, I tried it with taxes once, and that turned out very expensive.
OTOH one guy here declared the value of his house wrong,
comma sort of thing, and got 3 million Euro back from the taxes, automatic.
Years later they figured out his error, and asked it back.
He sued them, and the judge decided he did not have to pay it back.
(Tax office should have spotted the mistake).
Was in the paper last month.
Now I have this idea ;-)
 
S

Spehro Pefhany

Jan 1, 1970
0
That is for experts. They know to add a resistor.

- Henry

Not necessarily. See, for example, US 6313589


Best regards,
Spehro Pefhany
 
H

Henry Kiefer

Jan 1, 1970
0
Spehro said:
Not necessarily. See, for example, US 6313589

That patent is a temperature compensation circuit. Not bad idea.


- Henry
 
J

Jim Thompson

Jan 1, 1970
0
Not necessarily. See, for example, US 6313589


Best regards,
Spehro Pefhany

"Constant" light output.

...Jim Thompson
 
A

Alex

Jan 1, 1970
0
If I connect 8 leds @ 1.6v vf onto a power supply producing 12vdc, I
won't have to use a resistor, right? Theory being that the 8 led's
will use .8 volts more than the ps can provide. They won't burn as
bright, but they won't burn up the ps, they won't destroy themselves,
and they won't require resistors?

Ofcourse I agree with those that say its a current driven device..
Just funny .. I bought a lightchain with 480 diodes for less
than(equal 20$) a few days ago,as the supermarkets have turned down
the prices .
It works well conected to a transformer that is included in the box..
but I think the trafo will have an airgab ,to give a high R out from
trafo.
Also ,they use AC output,and LED's in anti parallel. Buying LED's
with socket this way is actually cheap.. and the transformer is worth
the price too.:)
Alex from Sonderborg in Denmark.
 
Top