Re: Any innovative ideas for this LED driving problem?



On Thu, 29 Nov 2007 09:21:21 -0800 (PST),
ferrari.secret.santa@xxxxxxxxx wrote:

I am sure many of you are familiar with the newer higher-power SMT
LED's that are out. They generally require larger copper pads on the
PCB to dissipate heat.

I am working on an application where I need a lot of brightness, and
very small size. I also need to have my design be dual color (for
example red and green). My solution was to put the LEDs in strings of
three, side by side but oppositely biased to one another. So that if V
+ is on the left and GND on the right, its red, if V+ is on the right
and GND on the left, its green. Since red and green would never be on
together, this allows me to use the same copper for power dissipation
on both strings.

I was originally using an H-bridge to swap the polarity. But the
design has turned into 6 strings of 3 for red and 6 stings of 3 for
green. Six h-bridges... I am trying to keep the board as small as
possible and I am running out of room doing this with discrete
components.

Of course they make motor drivers that are H-bridges built in, but
they are expensive and overkill for what I need. If I had my wish
come true, it would be for a small chip with 5 or so pins... power and
ground in, a control line, and to outputs. If control is low, output
1 is V+ and output 2 is ground, and if control is high, output 1 is
ground and output 2 is V+. Does such a beast exist?

If not, is there any other suggestions on how to tackle this? I could
do away with the h-bridge idea and just parallel the strings of LEDs,
mass the cathodes (since that's where heat dissipation is needed) and
seperate the anodes and control via transistors switched from the
microcontroller, but I am not sure if that is the best way.

As a non-professional designer, does anyone have alternate suggestions
or something I am missing or perhaps a more elegant solution?

Thank you!

How much voltage? How much current?

John

.



Relevant Pages

  • Re: Any innovative ideas for this LED driving problem?
    ... PCB to dissipate heat. ... this allows me to use the same copper for power dissipation ... I was originally using an H-bridge to swap the polarity. ... ground in, a control line, and to outputs. ...
    (sci.electronics.design)
  • Re: Any innovative ideas for this LED driving problem?
    ... PCB to dissipate heat. ... this allows me to use the same copper for power dissipation ... on both strings. ... ground in, a control line, and to outputs. ...
    (sci.electronics.design)
  • Re: Bass speaker cone
    ... Les Cargill wrote: ... dissipate heat. ... I can't back this up with precise statistics, but I'd be willing to bet that most speaker cabinets' power handling limit is determined by physical displacement capability, not power dissipation. ... If a speaker's not "stuck" at past Xmax when it fails, then it failed due to heat dissipation. ...
    (alt.guitar.bass)
  • Re: Bass speaker cone
    ... dissipate heat. ... I can't back this up with precise statistics, but I'd be willing to bet that most speaker cabinets' power handling limit is determined by physical displacement capability, not power dissipation. ... If a speaker's not "stuck" at past Xmax when it fails, then it failed due to heat dissipation. ...
    (alt.guitar.bass)
  • Re: Re: Beco 815af
    ... reduce the sensitivity until it does. ... start adjusting the fine control. ... there start adjusting the dissipation factor control to minimise even ...
    (sci.electronics.repair)