LED lightbulb power consumption

Then archive this with it!
http://members.misty.com/don/ledx.html
More info on LEDS than you could ever want to know.

--
Gopher Greg
'77 CB750K Stock '78 CB750K AHRMA
'00 ZG1000 Stock '96 Ducati 900SS Former track bike
'01 GSXR750 Current race bike
**********pull 'mychain' to reply***********
("I've abandoned the idea of trying to appear a normal, pleasant person.
I had to accept myself as I was, even if no one else could accept me.
For the rest of my life I would continue to say precisely the wrong
thing, touch people in the raw and be generally unpopular. I had a
natural gift for it" W. F. Temple)

 

This thread is for the Google archives mainly, but I figured some of you
would be interested.

My LED bulbs to replace my dead incandescents came in. I haven't installed
them, but hooking them up to my refreshed "dead" battery from my debacle
earlier has piqued my interest. I hooked up the ammeter and checked out the
current. Wow. Low power galore! I love the targeted beam... perfect for
indicator lights. I also picked up some inverted lens "flood-style" LEDs
for the gauges... these should be good.

17.8mA @ 12.6V means 224mW. This is only 13% of the rated power (1.7W)
consumed by the incandescents! This means that big power savings are to be
had.

I'm happy... once exams and finals are done I can hook these babies up and
check out their performance.

 

Most of that power will probably be used to heat the dropping
resistor in series with the diode(s), depending on how they are wired.
A quick look shows UltraBrite white LEDs have a forward voltage of 3.5 V,
so with a standard 470-ohm dropping resistor to one diode, the resistor
accounts for ~70% of the power consumption.

To make them even brighter, use a constant-current source and
cool them with liquid nitrogen[1].

[1] J. Phys. E: Sci. Instr. 17, 555 (1984).

--
Ivan Reid, Electronic & Computer Engineering, ___ CMS Collaboration,
Brunel University. Ivan.Reid@[brunel.ac.uk|cern.ch] Room 40-1-B12, CERN
GSX600F, RG250WD "You Porsche. Me pass!" DoD #484 JKLO#003, 005
WP7# 3000 LC Unit #2368 (tinlc) UKMC#00009 BOTAFOT#16 UKRMMA#7 (Hon)
KotPT -- "for stupidity above and beyond the call of duty".

 

Not sure I'd call 1.7W a big power savings, but LEDs are also very
long lived, so that's a plus.

Be careful of LEDs that are supposed to replace a stock tail light or
brake light bulb. There are some that may be fine but there are also
some that are considerably less visible. Some of the cheap units
will use only a very few LEDs to keep the cost down.

I run a 52 LED truck taillight which I had to make my own bracket
for. Highly visible from the back, though I'd like more visibility
from the sides.

I'm not familiar with the inverted lens units.

 

Oops... I forgot to mention that I have 11 of these in my instrument panel,
plus my running lights in the tail light (16W) and the front turn signals
(16W). So overall it's a savings of around 32W.

Like you said, they last a long time. The LEDs were bought because I didn't
want to have to go back into the instrument panel when the lights failed one
by one.

I'll be putting one of the pair in to compare before-and-after so it's not
as much of a concern.

The taillight from the side? Mine aims straight back so it's not really
visible from the side. The integrated turn signals shield it.

The nice thing about drop-in replacements is that you can get bulbs like
these:
http://cgi.ebay.com/ebaymotors/1157...ewItemQQcategoryZ33713QQitemZ8056833696QQrdZ1

They're radial/axial shining ones that fill up your reflectors and go
sideways, too. They're physically big though and will just barely fit in my
lens.

They don't have a dome lens at the tip of the LED... they have an inverted
cone that goes towards the semiconductor. This shoots the light out mostly
to the side. These will go in my backlit gauges. Here are the ones I got;
look at the red picture:
http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&rd=1,1&item=8054126915

Compare them to these:
http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&rd=1,1&item=8053911705

Look at the white picture: these will go on my indicator lights.

 

Are you still running red lights now your'e off the pushbike?

 

Yup. My LEDs have a clear base with the resistor visible. Calculating from
the resistor code (green-brown-brown-gold) I have a 510 ohm resistor. Using
the formula P=i^2*R, I calculated the actual resistance to be 674 ohms. It
looks like there's another 164 ohms added in from the diode. So about 25%
of the power goes into powering the diode, and 75% goes to heat. Agrees
well with your numbers below.

My LED 2-AA flashlight has a thick, burly aluminum body that acts as a big
heatsink to uh, approximate liquid nitrogen, hehe.

 

Ah... yes... I have learned my lessons about running red lights. I'm not
totally off the bicycle these days. I organized some urban assault pushie
races through campus, and campus police just *love* it! It gives them
something to chase after, and the trick is not to look back if they yell at
you. One guy didn't heed this and got arrested on the last race. The cop
must have been so proud to rid the city of the scum known as
stop-sign-running bicycle riders. During these races, I just take it slow,
abide by the law, and watch the fireworks from afar, hehehe

I also refuse to ever get a DUI on a bicycle because it's as severe as a car
DUI.

 

Good to see you still have a sense of humor when it comes to the campus
cops. Enjoy it while you can. with your finals coming up, it's almost
time for 'the real world' for you. Just imagine a well paid job and more
motorcycles.......but only if you pass!

That is bollocks! We have an annual push bike pub crawl, and the only
time the cops get involved is if someone falls off. There is usually a
cop 'shadow' car to see things don't get too silly. I find phyisical
activity and beer don't mix too well, especially when you have to
balance on two wheels.

 

Ah, yes, 510-ohm 5%; I couldn't remember what was the next standard
value up after 470-ohm and didn't think to walk into the lab to check our
stock. <pause/> Next standard 20% value is 560, or 510 for the higher accuracy.
You've forgotten the diode forward voltage (which doesn't change much with
current in the operating region but probably increases with decreasing
temperature) in your calculation; (12.6-3.5)V/510-ohm = 17.84 mA
-- close enough for Government business!

Ah, I thought the callouses on your right hand were from something
different! :-;

--
Ivan Reid, Electronic & Computer Engineering, ___ CMS Collaboration,
Brunel University. Ivan.Reid@[brunel.ac.uk|cern.ch] Room 40-1-B12, CERN
GSX600F, RG250WD "You Porsche. Me pass!" DoD #484 JKLO#003, 005
WP7# 3000 LC Unit #2368 (tinlc) UKMC#00009 BOTAFOT#16 UKRMMA#7 (Hon)
KotPT -- "for stupidity above and beyond the call of duty".

 

Forward voltage... didn't know there was one. I googled it up and now I
know. I was thinking of the ideal diode. It looks like your calculated
17.84mA matches up almost exactly with my measured 17.8mA (limited by
multimeter res).

 

For more power savings, you could run 3 of them in series, with a smaller
resistor, 150 ohms or so.