What's a "bright" LED?

[Tomás Ó hÉilidhe-2]

Here's what I'm looking for:
    * A bi-colour two-lead LED which consists internally of two LED's in
parallel facing in opposite directions
    * The forward voltage of both colours must be less than 2.5 V
    * 20 mA should be enough to have it very lit
    * Extremely bright because I'm multiplexing them in a 30 stage
multiplexer and I want them to be visible in daylight

First thing I want to ask is:

Can someone suggest to me a minimal candela rating I should aim for? I
know 1 candela is the light given off a candle, but I haven't got much
of a clue how I should compare this to LED light.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Tomás Ó hÉilidhe wrote:
> Can someone suggest to me a minimal candela rating I should aim for? I
> know 1 candela is the light given off a candle, but I haven't got much
> of a clue how I should compare this to LED light.

There is no substitute for experience for this.  I guess this means you
*still* haven't done the LED tests I told you to do a week or two ago.  No,
I'm not just going to give you answers if you continue to refuse to do any
work on your own.  In this particular case any answer you get via email
won't be that useful anyway.  You're going to have to see this for yourself.
And yes, that means doing a little work on your own.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tomás Ó hÉilidhe-2]

Olin Lathrop wrote:
> Tomás Ó hÉilidhe wrote:
>  
>> Can someone suggest to me a minimal candela rating I should aim for? I
>> know 1 candela is the light given off a candle, but I haven't got much
>> of a clue how I should compare this to LED light.
>>    
>
> There is no substitute for experience for this.  I guess this means you
> *still* haven't done the LED tests I told you to do a week or two ago.


Sorry Daddy, no I haven't done the tests. If I thought they'd be
productive I'd do them.


> No,
> I'm not just going to give you answers if you continue to refuse to do any
> work on your own.


Maybe next time I want to pick a resistor value I should take out my
breadboard and test what current I'll get through it when I apply a
certain voltage.


> In this particular case any answer you get via email
> won't be that useful anyway.  You're going to have to see this for yourself.
> And yes, that means doing a little work on your own.

No have LED's. No can do tests.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Tomás Ó hÉilidhe wrote:
> Sorry Daddy, no I haven't done the tests. If I thought they'd be
> productive I'd do them.

You don't have the experience to know what will be productive and what not.
I suspect the real reason however is that you'd actually have to do some
work, which you clearly have a aversion to.

> No have LED's. No can do tests.

Of course you can.  LEDs are are both cheap and available.  Your project
relies on them heavily, so you really should get some and play around with
them.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tamas Rudnai]

Another subject to the old topic.... ehhh

> Maybe next time I want to pick a resistor value I should take out my
> breadboard and test what current I'll get through it when I apply a
> certain voltage.

No, this is not the same. With a LED you already know what current source
you need, it's in the datasheet of the LED. What you do not know is if the
LED is bright enough or not - bright enough for YOUR application for YOUR
eyes... Olin keeps telling you this, but you do not seems to understand: See
how the LED is working for you - different LEDs, different current sources,
different frequencies / duty cycles.

With a resistor you may can have experience of capacitivness and
inductiveness of that resistor for example. Or measure the resistance with
high frequency - you'll be surprised...

Tamas


On Sat, Jul 5, 2008 at 5:22 PM, Tomás Ó hÉilidhe <toe@...> wrote:

>
>
> Olin Lathrop wrote:
> > Tomás Ó hÉilidhe wrote:
> >
> >> Can someone suggest to me a minimal candela rating I should aim for? I
> >> know 1 candela is the light given off a candle, but I haven't got much
> >> of a clue how I should compare this to LED light.
> >>
> >
> > There is no substitute for experience for this.  I guess this means you
> > *still* haven't done the LED tests I told you to do a week or two ago.
>
>
> Sorry Daddy, no I haven't done the tests. If I thought they'd be
> productive I'd do them.
>
>
> > No,
> > I'm not just going to give you answers if you continue to refuse to do
> any
> > work on your own.
>
>
> Maybe next time I want to pick a resistor value I should take out my
> breadboard and test what current I'll get through it when I apply a
> certain voltage.
>
>
> > In this particular case any answer you get via email
> > won't be that useful anyway.  You're going to have to see this for
> yourself.
> > And yes, that means doing a little work on your own.
>
> No have LED's. No can do tests.
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Rudonix DoubleSaver
http://www.rudonix.com
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tamas Rudnai]

> LEDs are are both cheap and available.

Yes, just go to Maplin and pick up a lucky bag for LEDs, you'll have enough
for playing for a while.

Tamas



On Sat, Jul 5, 2008 at 5:55 PM, Olin Lathrop <olin_piclist@...>
wrote:

> Tomás Ó hÉilidhe wrote:
> > Sorry Daddy, no I haven't done the tests. If I thought they'd be
> > productive I'd do them.
>
> You don't have the experience to know what will be productive and what not.
> I suspect the real reason however is that you'd actually have to do some
> work, which you clearly have a aversion to.
>
> > No have LED's. No can do tests.
>
> Of course you can.  LEDs are are both cheap and available.  Your project
> relies on them heavily, so you really should get some and play around with
> them.
>
>
> ********************************************************************
> Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
> [phone].  Gold level PIC consultants since 2000.
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Rudonix DoubleSaver
http://www.rudonix.com
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tomás Ó hÉilidhe-2]

Tamas Rudnai wrote:

>> Maybe next time I want to pick a resistor value I should take out my
>> breadboard and test what current I'll get through it when I apply a
>> certain voltage.
>>    
>
> No, this is not the same. With a LED you already know what current source
> you need, it's in the datasheet of the LED. What you do not know is if the
> LED is bright enough or not - bright enough for YOUR application for YOUR
> eyes... Olin keeps telling you this, but you do not seems to understand: See
> how the LED is working for you - different LEDs, different current sources,
> different frequencies / duty cycles.
>  

This applied to when the voltage source was very close to the forward
voltage of the diode. Since my source voltage is well above the diodes'
voltage drops, I don't need to worry about them being dim; I just pick
the right resistor and I'm good to go.

Theory will have to suffice for now because I haven't got the LED's to
test out practically.

Of course, if I was working with a lower voltage voltage source, there
might be some merit in the experiment.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Gerhard Fiedler]

Tomás Ó hÉilidhe wrote:

> Theory will have to suffice for now because I haven't got the LED's to
> test out practically.

I'm the first one to look into datasheets and calculate something before
trying it out -- but there are some things (especially when you don't know
what the number in a datasheet mean for you) where you /have/ to do some
experimenting and figure that out for yourself. It wouldn't help you if I
told you that 1 cd is good for me... you may want a different viewing
angle, or want it brighter or be happy with less bright, or whatever...

In what end of the world are you that LEDs are difficult to get by? You
don't need bicolor LEDs; you just need some LEDs for which you have
datasheets.

Gerhard

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tamas Rudnai]

> In what end of the world are you that LEDs are difficult to get by? You
> don't need bicolor LEDs; you just need some LEDs for which you have
> datasheets.

I believe he's from Ireland too (same as me). Actually there is only one
shop so far I could find here in Dublin where you just go in and buy some
components: Maplin. Unfortunately Maplin is UK based and many times they do
not have in stock all of those components that is on the catalog. I'd say
too many times... Sometimes hard to get 100nF... However, a pack of LED is
easy, as they have that lucky bag which contains many LEDs different
colours, shapes and sizes. Also Radionics delivers in 2 days in my
experience. Fairchild and Digikey also good source here. So yes, end-of-the
world (next stage is America if you go a bit further to the west) but even
here he could have get those LEDs very easy.


Tamas
PS: Maplin is open at Sunday too...



On Sat, Jul 5, 2008 at 10:08 PM, Gerhard Fiedler <lists@...>
wrote:

> Tomás Ó hÉilidhe wrote:
>
> > Theory will have to suffice for now because I haven't got the LED's to
> > test out practically.
>
> I'm the first one to look into datasheets and calculate something before
> trying it out -- but there are some things (especially when you don't know
> what the number in a datasheet mean for you) where you /have/ to do some
> experimenting and figure that out for yourself. It wouldn't help you if I
> told you that 1 cd is good for me... you may want a different viewing
> angle, or want it brighter or be happy with less bright, or whatever...
>
> In what end of the world are you that LEDs are difficult to get by? You
> don't need bicolor LEDs; you just need some LEDs for which you have
> datasheets.
>
> Gerhard
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Rudonix DoubleSaver
http://www.rudonix.com
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tomás Ó hÉilidhe-2]

Tamas Rudnai wrote:
> So yes, end-of-the
> world (next stage is America if you go a bit further to the west) but even
> here he could have get those LEDs very easy.

Try searching for two-pin LED's that have a certain minimal candela
rating. Impossible on Farnell's website. Might try Radionics, but I
think I've tried them already.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tamas Rudnai]

http://ie.farnell.com/jsp/search/browse.jsp?N=500008+1000125&Ntk=gensearch_001&Ntt=led&Ntx=


On Sat, Jul 5, 2008 at 10:50 PM, Tomás Ó hÉilidhe <toe@...> wrote:

>
>
> Tamas Rudnai wrote:
> > So yes, end-of-the
> > world (next stage is America if you go a bit further to the west) but
> even
> > here he could have get those LEDs very easy.
>
> Try searching for two-pin LED's that have a certain minimal candela
> rating. Impossible on Farnell's website. Might try Radionics, but I
> think I've tried them already.
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Rudonix DoubleSaver
http://www.rudonix.com
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[William "Chops" Westfield]

On Jul 5, 2008, at 10:40 AM, Tamas Rudnai wrote:

>> LEDs are are both cheap and available.
>
> Yes, just go to Maplin and pick up a lucky bag for LEDs, you'll  
> have enough
> for playing for a while.

Um.  Useless, unfortunately.  Find an LED you like, and THEN what?  
It looks just like all the other LEDs, you don't know who made it or  
what its part number is, and in general, even if you've decided that  
it's the BEST LED IN THE WHOLE WORLD, you can't count on being able  
to buy any more of them!

Frankly, for high brightness, you're in trouble.  For some reason it  
seems to be quite uncommon for manufacturers to put together their  
brightest LED chips in this configuration.  High brightness 4-lead  
RGB leds are pretty common, and three-lead two-color aren't TOO hard  
to find, but the two-lead bipolars are rare...

BillW



--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tomás Ó hÉilidhe-2]

Tamas Rudnai wrote:
> http://ie.farnell.com/jsp/search/browse.jsp?N=500008+1000125&Ntk=gensearch_001&Ntt=led&Ntx=
>

Congratulations you can search for an LED.

Now try searching for a bi-polar, bi-colour, bi-lead LED that has one
dome and that's above 50 mcd.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Tomás Ó hÉilidhe wrote:
> Now try searching for a bi-polar, bi-colour, bi-lead LED that has one
> dome and that's above 50 mcd.

First, the exercise was to find out how bright is bright enough, so
individual LEDs are just fine, even easier to use.  Since you don't know how
bright you need, you don't know whether 50mcd is enough or not.

Second, two-lead bicolor LEDs being so hard to find should tell you it's
probably a bad idea to rely on them in your design.  There are lots of
alternatives, like using separate LEDs, common anode or common cathode, or
even RGB LEDs.  Some of these solutions simplify the drive requirements and
even get around the problem of using color to convey information, which is
not a good idea as has already been discussed.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tomás Ó hÉilidhe-2]

Olin Lathrop wrote:
> Since you don't know how
> bright you need, you don't know whether 50mcd is enough or not.
>  

I was hoping somebody with experience could tell me. If someone hadn't
got a clue about guitar strings and asked me what ones to get for a
beginner, I'd be able to tell them to get 8's or 9's. In similar vein, I
was hoping someone could cut it with a simple "You'd want about half a
candela at least".

> Second, two-lead bicolor LEDs being so hard to find should tell you it's
> probably a bad idea to rely on them in your design.  There are lots of
> alternatives, like using separate LEDs, common anode or common cathode, or
> even RGB LEDs.

Separate LED's wouldn't really do the trick for the game I have in mind.

I don't have enough pins on the 887 to work with 3-pin LED's.


> Some of these solutions simplify the drive requirements and
> even get around the problem of using color to convey information, which is
> not a good idea as has already been discussed.

The Connect4 game is well established as using colour to indicate the
players' chips. Of course a pattern could work instead of colour, but I
haven't seen any pattern LED's.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[William "Chops" Westfield]

On Jul 5, 2008, at 9:22 AM, Tomás Ó hÉilidhe wrote:

>>> Can someone suggest to me a minimal candela rating

Ok.  1Cd (1000mCd) is sort of the beginning of the range that truely  
qualifies as "ultrabright."  That's for narrow viewing-angle LEDs;  
for your app you may want to consider wide-angle LEDs, even  
"diffused" domes, which will cut the Cd rating WAY down.
If you have a datasheet that talks about "lumens" rather than  
Candela, that's a better rating of overall "brightness."
The way I look at it, Cd is more relevant of you're shining the LED  
on something and want to know how bright the "spot" is, while Lumens  
is more relevant if you're looking AT the LED (NOT shining the LED in  
your eye!) and what to know how bright it looks.

BillW


--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jinx-4]

> Try searching for two-pin LED's that have a certain minimal candela
> rating. Impossible on Farnell's website. Might try Radionics, but I
> think I've tried them already.

What happened with these people I posted the other day ? Blue and
yellow superbright bi-colours

http://www.lc-led.com/Products/department/27

http://www.lc-led.com/View/itemNumber/411

http://www.lc-led.com/products/n500tby4d.html



--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Tamas Rudnai]

 > In similar vein, I was hoping someone could cut it with a simple "You'd
want about half a
> candela at least".

In similar vein what would you say if I'd ask you which string shall I buy
for my guitar - you don't know what do I have, do you? I can't play and
never ever get a guitar in my hand but I would even telling you certain
parameters of that string. And then if you tell me what string shall I buy I
will keep saying it is not good, does not sound good to me or broke off the
head machine.

Actually I found couple in Farnell (just have to use the right searching
phrase) but as you answered arrogant I just do not feel showing the link for
you. Other thing is that you still do not know about 50mcd - that's quite
high in my opinion, and it also indicated by the fact that you could not
find it easily. Once I almost got blind looking at the led directly - for an
indicator of your game board you'd probably enough much less - but back to
Olin's standard sentence: Try those leds out, see how bright are they...

Good luck,
Tamas



On Sun, Jul 6, 2008 at 12:24 AM, Tomás Ó hÉilidhe <toe@...> wrote:

>
>
> Olin Lathrop wrote:
> > Since you don't know how
> > bright you need, you don't know whether 50mcd is enough or not.
> >
>
> I was hoping somebody with experience could tell me. If someone hadn't
> got a clue about guitar strings and asked me what ones to get for a
> beginner, I'd be able to tell them to get 8's or 9's. In similar vein, I
> was hoping someone could cut it with a simple "You'd want about half a
> candela at least".
>
> > Second, two-lead bicolor LEDs being so hard to find should tell you it's
> > probably a bad idea to rely on them in your design.  There are lots of
> > alternatives, like using separate LEDs, common anode or common cathode,
> or
> > even RGB LEDs.
>
> Separate LED's wouldn't really do the trick for the game I have in mind.
>
> I don't have enough pins on the 887 to work with 3-pin LED's.
>
>
> > Some of these solutions simplify the drive requirements and
> > even get around the problem of using color to convey information, which
> is
> > not a good idea as has already been discussed.
>
> The Connect4 game is well established as using colour to indicate the
> players' chips. Of course a pattern could work instead of colour, but I
> haven't seen any pattern LED's.
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Rudonix DoubleSaver
http://www.rudonix.com
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist