Need same color bi-directional LED 3mm

On Sunday, October 28, 2018 at 8:11:35 PM UTC-7, gnuarm.del...@gmail.com wrote:
> On Sunday, October 28, 2018 at 9:42:16 PM UTC-4, Jeff Liebermann wrote:

> O----+---|>|---+---|<|---+----O
>      |         |         |
>      +--/\/\/--+--/\/\/--+
> 
> Won't this do it? 

or the near-equivalent, using a silicon dual diode, common anode

> O----+---|>|---+---|<|---+----O
>      |         |         |
> O----+---|<|---+---|>|---+----O
> 

On Monday, 29 October 2018 04:48:25 UTC, Jeff Liebermann  wrote:
> On Sun, 28 Oct 2018 21:10:02 -0700, Jeff Liebermann <jeffl@cruzio.com>
> wrote:
> 
> >On Sun, 28 Oct 2018 20:11:31 -0700 (PDT),
> >gnuarm.deletethisbit@gmail.com wrote:
> >>O----+---|>|---+---|<|---+----O
> >>     |         |         |
> >>     +--/\/\/--+--/\/\/--+
> 
> >>Won't this do it? 
> >>Rick C. 
> 
> >I had thought of that but also noticed that the maximum LED reverse
> >voltage and current might be exceeded.  That's why I preferred the
> >back to back diode approach.
> >
> >Assuming 24VDC operating voltage and a 2.0V forward voltage drop on
> >the LED, that puts 22VDC reverse voltage on the unlit LED which is
> >rated at 5V reverse maximum.  I don't know exactly how the LED with
> >react when it enters the reverse breakdown region, but if it
> >avalanches and conducts somewhere above 5V, there may be a problem.  I
> >found some vague references suggesting that the reverse avalanche
> >voltage is about 20V or more, which might eliminate the problem.
> >
> >Looking at some typical specs,
> ><https://media.digikey.com/pdf/Data%20Sheets/Lite-On%20PDFs/LTL-42x1,42x2.pdf>
> >it seems (not sure) that it will zener at something above 5VDC and
> >will tolerate no more than 100 microamps.  At a voltage of
> >   24v - 2.0v - 5.6v = 16.4V    (the 5.6v is my guess)
> >the existing 5.2K resistor will produce:
> >   16.4V / 5200ohms = 3.2ma
> >which greatly exceeds the 100 microamp reverse current limit.  
> >
> >The maximum dissipation for green is 100mw.  With this arrangement,
> >the reverse polarized LED will dissipate:
> >   2.0V * 0.0032A = 6.4mw
> >which is much less than 100mw.  So, it just might survive reverse
> >conduction.  I don't know exactly what will happen, so it's probably
> >worth my time bench measuring the reverse breakdown voltage and
> >testing the idea.
> >
> >"What destroys a LED in the reverse direction?"
> ><https://electronics.stackexchange.com/questions/173283/what-destroys-a-led-in-the-reverse-direction>
> >  "In my experience the standard 3mm and 5mm red LEDs can block 
> >  12V easily so I have used them as reverse polarity protection 
> >  in 12V systems where current is about 10 mA DONT use them on 
> >  a 24 Volt system some LEDs died at approx. 30V"
> >
> >Thanks much for the suggestion.
> 
> Looks like it just might work.  I took a green 3mm LED, wired it in
> series with a 2.2K resistor, and connected it to an HP6235A variable
> voltage power supply.  I connected a DVM across the LED.  There was no
> indication of any reverse conduction or voltage limiting to 40V.  So,
> it just might work.  I don't have any of the 3 pin LED's in stock, so
> I'll order some immediately.  Meanwhile, I get to ponder over why the
> LED manufacturers specify a maximum of 5V reverse voltage when it
> seems to be much higher.  Any clues?
> 
> Thanks again.

5v is just a figure they're confident they'll all pass with no testing. They're not going to waste time/money testing a spec almost no-one is interested in ever using. Real breakdown v varies widely, some even do 100v.


NT

On Monday, 29 October 2018 05:39:46 UTC, Jeff Liebermann  wrote:
> On Sun, 28 Oct 2018 21:48:31 -0700, Jeff Liebermann <jeffl@cruzio.com>
> wrote:
> >Meanwhile, I get to ponder over why the
> >LED manufacturers specify a maximum of 5V reverse voltage when it
> >seems to be much higher.  Any clues?

why not make it even easier & just solder an smd green led between the existing led's pins. You might even get it shining up into the existing led perhaps. Any green glow should work, doesn't need an ideal distribution.


NT

On 2018-10-29, Jeff Liebermann <jeffl@cruzio.com> wrote:
> On Sun, 28 Oct 2018 17:47:57 -0700 (PDT),
> gnuarm.deletethisbit@gmail.com wrote:
>
>>On Sunday, October 28, 2018 at 6:08:11 PM UTC-4, Jeff Liebermann wrote:
>>> On Sun, 28 Oct 2018 13:26:14 -0700 (PDT),
>>> gnuarm.deletethisbit@gmail.com wrote:
>>> 
>>> >I gave you a link to a usable green/green bicolor LED.  What more do you want? 
>>> >Rick C. 
>  
>>> Perhaps free samples?  About 100 pcs will suffice.
>>> 
>>> I replied to your post with:
>>> <https://groups.google.com/forum/#!msg/sci.electronics.design/2NCYlHgUQy0/SiFwYL7gAQAJ>
>
>>Wrong post. Try this one. 
>>https://groups.google.com/d/msg/sci.electronics.design/2NCYlHgUQy0/6AC5FYjjAQAJ
>
> That points to:
><https://www.lc-led.com/View/itemNumber/726>
> That 3mm LED has 3 leads with a common anode.  How am I suppose to
> wire that for for back to back diode operation that will light up with
> either polarity?  There were plenty of 3mm green-green 3mm LED's with
> either common anode or common cathode available such as:
><https://www.ebay.com/itm/123452852566>
> I couldn't figure out how to wire them, so I ignored those.  I
> couldn't find any that had 2 LED's in series which could easily be
> wired back to back.
>
    3 wire LED for 2 wire operation

           //      //
  A  -+---|>|--+--|<|----+-[750R]-- B
      |        |         |
      +-[100R]-+-[100R]--+



-- 
  When I tried casting out nines I made a hash of it.

On 2018-10-29, Jeff Liebermann <jeffl@cruzio.com> wrote:
> On Sun, 28 Oct 2018 21:10:02 -0700, Jeff Liebermann <jeffl@cruzio.com>
> wrote:
>
>>On Sun, 28 Oct 2018 20:11:31 -0700 (PDT),
>>gnuarm.deletethisbit@gmail.com wrote:
>>>O----+---|>|---+---|<|---+----O
>>>     |         |         |
>>>     +--/\/\/--+--/\/\/--+
>
>>>Won't this do it? 
>>>Rick C. 
>
>>I had thought of that but also noticed that the maximum LED reverse
>>voltage and current might be exceeded.  That's why I preferred the
>>back to back diode approach.
>>
>>Assuming 24VDC operating voltage and a 2.0V forward voltage drop on
>>the LED, that puts 22VDC reverse voltage on the unlit LED which is
>>rated at 5V reverse maximum.  I don't know exactly how the LED with
>>react when it enters the reverse breakdown region, but if it
>>avalanches and conducts somewhere above 5V, there may be a problem.  I
>>found some vague references suggesting that the reverse avalanche
>>voltage is about 20V or more, which might eliminate the problem.
>>
>>Looking at some typical specs,
>><https://media.digikey.com/pdf/Data%20Sheets/Lite-On%20PDFs/LTL-42x1,42x2.pdf>
>>it seems (not sure) that it will zener at something above 5VDC and
>>will tolerate no more than 100 microamps.  At a voltage of
>>   24v - 2.0v - 5.6v = 16.4V    (the 5.6v is my guess)
>>the existing 5.2K resistor will produce:
>>   16.4V / 5200ohms = 3.2ma
>>which greatly exceeds the 100 microamp reverse current limit.  
>>
>>The maximum dissipation for green is 100mw.  With this arrangement,
>>the reverse polarized LED will dissipate:
>>   2.0V * 0.0032A = 6.4mw
>>which is much less than 100mw.  So, it just might survive reverse
>>conduction.  I don't know exactly what will happen, so it's probably
>>worth my time bench measuring the reverse breakdown voltage and
>>testing the idea.
>>
>>"What destroys a LED in the reverse direction?"
>><https://electronics.stackexchange.com/questions/173283/what-destroys-a-led-in-the-reverse-direction>
>>  "In my experience the standard 3mm and 5mm red LEDs can block 
>>  12V easily so I have used them as reverse polarity protection 
>>  in 12V systems where current is about 10 mA DONT use them on 
>>  a 24 Volt system some LEDs died at approx. 30V"
>>
>>Thanks much for the suggestion.
>
> Looks like it just might work.  I took a green 3mm LED, wired it in
> series with a 2.2K resistor, and connected it to an HP6235A variable
> voltage power supply.  I connected a DVM across the LED.  There was no
> indication of any reverse conduction or voltage limiting to 40V.  So,
> it just might work.  I don't have any of the 3 pin LED's in stock, so
> I'll order some immediately.  Meanwhile, I get to ponder over why the
> LED manufacturers specify a maximum of 5V reverse voltage when it
> seems to be much higher.  Any clues?

guaranteed by design... vs need to test each batch?

I've seen LEDs die from 9v reverse.


-- 
  When I tried casting out nines I made a hash of it.

On Sun, 28 Oct 2018 23:00:38 -0700 (PDT), tabbypurr@gmail.com wrote:

>On Monday, 29 October 2018 05:39:46 UTC, Jeff Liebermann  wrote:
>> On Sun, 28 Oct 2018 21:48:31 -0700, Jeff Liebermann <jeffl@cruzio.com>
>> wrote:
>> >Meanwhile, I get to ponder over why the
>> >LED manufacturers specify a maximum of 5V reverse voltage when it
>> >seems to be much higher.  Any clues?
>
>why not make it even easier & just solder an smd green led between 
>the existing led's pins. You might even get it shining up into 
>the existing led perhaps. Any green glow should work, doesn't need 
>an ideal distribution.
>NT

You already suggested that and I already replied that I like the idea
and that it will probably work.  See:
<https://groups.google.com/d/msg/sci.electronics.design/2NCYlHgUQy0/VISQS3YsAgAJ>
I would have tried it by now, except I can't find any 0805 green LEDs
in my mess.  I'll need to order some.

-- 
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

On 29/10/2018 02:44, Jeff Liebermann wrote:
> On Sun, 28 Oct 2018 19:23:57 -0700 (PDT), tabbypurr@gmail.com wrote:
> 
>> How you might make a back to back 3 leg green work: Wires to
>> the outer legs. An smd diode across each half, sandwiched between
>> its legs. If you solder the inner lead to the smds first the
>> outers should not be hard to do. That just might be a solution.
> 
> Good idea.  I'll try it later tonite if I can find some green SMD
> LEDs.  I may need to scrap some boards for parts.
> 
> The center to center lead spacing on a 3mm LED is 2.54mm.  The wire
> spacing on a 0805 SMD LED is 2.0mm, so I think I can make it fit and
> work.  It would be nice if I could get the light emitting "top" of the
> SMD LED directly against the bottom of the existing 3mm LED for
> maximum light transmission.  If it's not bright enough, one SMD LED to
> light up the existing 3mm LED and another in parallel pointed the
> opposite direction.  This may actually work.
> 
> My preliminary tentative thanks and undying gratitude if it works.
> If it hasn't already been mentioned, you can use two SM right angle LEDs 
in inverse parallel.  Loads at 3mm x 1mm and smaller.  This one's 2.1mm 
x 0.7mm, picked at random...

https://uk.farnell.com/kingbright/kpja-2107cgck/led-2-1x0-7mm-side-view-green/dp/2290368?st=right%20angle%20led

Cheers
-- 
Clive

Kingbright makes a few 5mm two-lead Green/Green LEDs:
https://www.kingbright.com/content/listitem/psearch/245
(at bottom)
https://www.newark.com/kingbright/l-57ggd/led-green-t-1-3-4-5mm-20mcd-568nm/dp/46W4629?CMP=AFC-OP
Unfortunately it's UK stock, with a $20 freight charge.

12V version (L57GGD12V, with dropping resistor):
https://www.digikey.com/products/en?keywords=L57GGD12V

Neither distributor page mentions that it's a bidirectional device, but 
both datasheets show it.

--S
-- 
sceptre@sdf.org
SDF Public Access UNIX System - http://sdf.org

Also Lumex SSL-LX5093GGD:
https://www.lumex.com/C060207P04.html#start
(see Bipolar section)

Only stock is at Heilind:
https://estore.heilind.com/SSL%2DLX5093GGD/LMXSSL%2DLX5093GGD.html

-- 
sceptre@sdf.org
SDF Public Access UNIX System - http://sdf.org

On Tue, 30 Oct 2018 02:05:23 +0000 (UTC), sceptre@sdf.lonestar.org
wrote:

>Kingbright makes a few 5mm two-lead Green/Green LEDs:
>https://www.kingbright.com/content/listitem/psearch/245
>(at bottom)
>https://www.newark.com/kingbright/l-57ggd/led-green-t-1-3-4-5mm-20mcd-568nm/dp/46W4629?CMP=AFC-OP
>Unfortunately it's UK stock, with a $20 freight charge.
>
>12V version (L57GGD12V, with dropping resistor):
>https://www.digikey.com/products/en?keywords=L57GGD12V

English version of Kingbright web page:
<https://www.kingbright.com/content/listitem/psearch/245/1>

Having the exact right part would sure make my life easier.  However,
all the above are 5mm parts and I need 3mm for it to fit inside the
relay.  Looks like Kingbright makes one that's bi-color green/green in
3mm:
<https://www.kingbright.com/attachments/file/psearch/000/00/00/L-937GGD(Ver.11B).pdf>
The 60 degree viewing angle will be much better than the 20 or 30
degree LED's I've been buying.

However, all the stocking distributors are too far away:
<https://octopart.com/l-937ggd-kingbright-12142861>
It does list one jobber in the USA, but they want me to ask for a
quote, which usually means sky high prices, large minimum quantities,
or both.

>Neither distributor page mentions that it's a bidirectional device, but 
>both datasheets show it.

Oh yes... I'm familiar with the problem.  The device is called a
"bi-color" LED, which translates to a "two color" LED.  But, what does
one do when both colors are the same?  I have no idea, and apparently
neither do the manufacturers and distributors. 

Thanks much.  I'll dig through the Kingbright web pile and see if I
can find anything else that might work.  It looks like I have an extra
month to deal with the LED problem because the mill in question just
destroyed some spindle bearings, a drive belt, a nylon gear, etc.

-- 
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558