Searching \ for '[EE] Driving large numbers of series white leds' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/displays.htm?key=leds
Search entire site for: 'Driving large numbers of series white leds'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Driving large numbers of series white leds'
2007\11\14@211049 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I need to make a large, flat panel filled with roughly 200-400 smd white
leds. I'm trying to get a thin, uniformly bright panel controlled by PWM
to achieve a good range of brightness, from completely off to on with
many (10bit at least) steps in between. I did consider
electroluminescent vinyl, but that stuff has a fairly short lifetime
before it burns out, and it's not all that bright anyway.

Researching this is looks like for large panels the usual way is to use
high voltage DC, often rectified directly from the 120VAC and use
massive strings of leds. For instance the NUD4011 low current led driver
is used to regulate current in such setups directly off of rectified AC.

I'm not much of a high voltage engineer, so I would rather keep to
something below 48V.

I'd also like to stick to using easilly available off the shelf
wall-warts, which on digikey seem to max out at about 60W 24V. Given
that constraint I can have at most something like 300-400 30ma 3.7V
leds, 33W-44W

For color matching and the like I would like to have every string
directly current regulated, no messing around with resistors. The leds I
would like to use have a quite forward voltage range and seem very
temperature dependent.


I recently did a test where I got 12 3.3V 20ma white leds in series and
tried using a lm317 in current regulating mode, 62ohm series resistor. I
also used a FET on the low side for PWM. Seemed to work great, although
I noticed that the circuit seemed to work best when I made sure there
was no capacitance anywhere. IE, on the oscilloscope at the start of PWM
the voltage going to the leds immediately dropped down to the right
level, with capacitance the waveform is increasingly rounded. I also
found that the current control seems to overshoot, causing the leds to
flash brightly, if I do stuff like touch the output with my hand. The
other annoying thing, is that the lm317 requires a minimum of 4ma of
load to work properly... which means I needed a resistor to ground,
dissipating quite a bit of watts!


I looked at a bunch of off-the-shelf LED drivers. While there are plenty
that will drive 10 or so LEDs in series, practically all of them are
either only available in non-hand-solderable packages and/or have
maximum input voltages less than 24V. (usually 10V or less) They also
tend to require a number of external parts, inductors and the like, and
given that I need 30-odd strings of 10-12 leds...

So, for the final design, I'm thinking use one 24V to 48V boost converter,
and using a bunch of simple linear regulators, either lm317's or
something dedicated to led driving. That said, the only such chip I
found for higher voltages was that NUD4011 one, and it looks like it's
basically impossible to get.


So, assuming this is a reasonable solution, is there a better solution
than the lm317 to limit current?

Any suggestions for the boost regulator? I've never designed a switch
mode regulator circuit before, something simple would be good.

If you got any better ideas, do tell!

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHO6qZ3bMhDbI9xWQRArLQAJ4yc6ZvvegGQgLVlDMuEq4DJ8FE2ACeJlz1
iPy1KVB1DolJlRH+1VJYnhg=
=irMq
-----END PGP SIGNATURE-----

2007\11\14@221231 by Spehro Pefhany

picon face
At 09:10 PM 11/14/2007, you wrote:
{Quote hidden}

A really simple current regulator suitable for this purpose would be a
BJT with base connected to 5V and a fixed emitter resistor. Eg. for
15mA, use about 300 ohms. You can switch the base between 5V and 0V
for the PWM. It will have a slight positive temperature coefficient,
(dI/dT ~= 0.04%/K) which may or may not compensate for LED brightness
variations with  temperature, but small percentage changes in current are
generally not very visible. The power dissipated in the resistor is
4.3 * Iled; the power dissipated in the transistor is Iled*
(Vsupply - Vled - 4.3V).

>Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spam_OUTspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com



2007\11\14@223309 by Marcel Duchamp

picon face
part 0 44 bytes
his is a multi-part message in MIME format.
part 1 631 bytes content-type:text/plain; charset=ISO-8859-1; format=flowed (decoded 7bit)

Peter Todd wrote:
>
> If you got any better ideas, do tell!
>

I don't know about better but here's one way to do it.  See attached gif.

U1 is whatever you are driving your pwm from. R2 is whatever is needed
to keep from overloading U1 when the current limiting fun begins.

Q1/R1 limit the current when it reaches Vbe/R1 or around 20 ma,
exclusive of temperature for the circuit shown.

Dissipation, etc. is left as an exercise to the student.

I'll leave the design of the clever switch mode power supply to others;
I'm out of town through the weekend for my daughters wedding.


part 2 5635 bytes content-type:image/gif; (decode)


part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2007\11\14@234626 by Rolf

face picon face
Did you see Olin's message a little bit ago. He pointed to this project
of his:

http://www.embedinc.com/pic/knli/index.htm

It appears very nice for your purposes because it will switch-mode up to
whatever voltage is required fro the required current.

You may want to inspect and adapt his project for your lower current
requirements....

Rolf

Peter Todd wrote:
{Quote hidden}

2007\11\14@235806 by Xiaofan Chen

face picon face
On 11/15/07, Peter Todd <.....peteKILLspamspam@spam@petertodd.ca> wrote:
> Researching this is looks like for large panels the usual way is to use
> high voltage DC, often rectified directly from the 120VAC and use
> massive strings of leds. For instance the NUD4011 low current led driver
> is used to regulate current in such setups directly off of rectified AC.
>
> I'm not much of a high voltage engineer, so I would rather keep to
> something below 48V.

High voltage may not be that difficult to deal with if you use Supertex
or some other parts.

http://www.supertex.com/feature_LED_general.html

Xiaofan

2007\11\15@055737 by cdb

flavicon
face
Check out  the Prema Led drivers  possibly the PR4402 might do what
you want.

http://www.prema.com

Or, perhaps playing with a 555 variant - both PWM and LED current
limiting at the same time.

Colin
--
cdb, colinspamKILLspambtech-online.co.uk on 15/11/2007

Web presence: http://www.btech-online.co.uk  

Hosted by:  http://www.1and1.co.uk/?k_id=7988359





--
Internal Virus Database is out-of-date.
Checked by AVG.
Version: 7.5.488 / Virus Database: 269.15.3/1082 - Release Date: 10/20/2007 2:59 PM


2007\11\16@055006 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, Nov 14, 2007 at 11:46:23PM -0500, Rolf wrote:
> Did you see Olin's message a little bit ago. He pointed to this project
> of his:
>
> http://www.embedinc.com/pic/knli/index.htm

That thing is really cool! I've actually got plans to make myself a nice
rugged caving headlamp with a Stenlight-like design. The mechanical
aspects of the project will likely be the most difficult actually, but
having a good switcher example is still very cool.

FWIW I'm very impressed by how clean and well documented the whole thing
is. Definetely a standard to aspire too.

> It appears very nice for your purposes because it will switch-mode up to
> whatever voltage is required fro the required current.
>
> You may want to inspect and adapt his project for your lower current
> requirements....

I think the nessesity of having the leds in 30 or so strings will be a
problem though. I'd have to have one switcher for each string, and
that'll be a lot of parts when all is said and done.

All the same, building one to better understand switching power supplies
is on my todo list now.

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHPXQw3bMhDbI9xWQRAtyLAKCXMXbF0VIpNFoKcgYnEtwM5R+PJQCfXPr/
ArLx8549xuEmhdRgpxYHihs=
=AwuL
-----END PGP SIGNATURE-----

2007\11\16@055007 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, Nov 15, 2007 at 12:58:03PM +0800, Xiaofan Chen wrote:
{Quote hidden}

I gotta admit I'm nixing high voltage not because of difficulty in the
circuit itself, those linear regulator based circuits look fairly
simple, but rather because I'm selling this stuff to other people and I
have no qualifications what-so-ever in anything safety related. If, say,
the plexiglass cover on the PCB falls off or cracks I'd rather
everything be touch safe.

That said, they make some very interesting chips, especially the fixed
current source ones like the CL25. One part for a temperature controlled
25ma current source could save me a lot of parts. The online store to
order them is reassuring too.

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHPW/G3bMhDbI9xWQRAjppAKCKsQURYlPdP3kFYUNeBk3iessuNQCfQVv1
orKw8ykFUIhSGqTqW7PDZJc=
=HP/1
-----END PGP SIGNATURE-----

2007\11\16@093307 by Timothy J. Weber

face picon face
Peter Todd wrote:
> On Wed, Nov 14, 2007 at 11:46:23PM -0500, Rolf wrote:
>> Did you see Olin's message a little bit ago. He pointed to this project
>> of his:
>>
>> http://www.embedinc.com/pic/knli/index.htm
>
> That thing is really cool! I've actually got plans to make myself a nice
> rugged caving headlamp with a Stenlight-like design. The mechanical
> aspects of the project will likely be the most difficult actually, but
> having a good switcher example is still very cool.
>
> FWIW I'm very impressed by how clean and well documented the whole thing
> is. Definetely a standard to aspire too.

A very good read!  I hadn't seen it before.

Some points worth repeating:

- Tupperware/Rubbermaid-type containers make good watertight enclosures
for one-off projects - I like 'em too (cf.
<http://6sys.no-ip.info/tjweber/gallery/dimbot/Open_4>).

- Just as wire coat hangers are a handy source of stiff metal, plastic
ones are a handy source of robust plastic rod!

- An interesting solution to the same power switch problem I was
studying with this project: <http://timothyweber.org/mancala> (also a
2xAA switcher for a portable device).  Olin's solution uses less weird
hardware and cleverer firmware.  I think I was obsessing about having
"no" current flowing while off; he's settled for "negligible" and made a
much more elegant solution in the process - fewer parts, more
flexibility in what's required for power-on.  I like it.

- Nice sleep/watchdog reset/jump back technique for low-power waiting in
the WAIT_WDT macro.

Typo in comment "N must not be from 1 to 255" twice I think.

Thanks for pointing out the project.
--
Timothy J. Weber
http://timothyweber.org

2007\11\16@100515 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, Nov 14, 2007 at 10:17:06PM -0500, Spehro Pefhany wrote:
{Quote hidden}

So essentially the base is set at a fixed voltage, I built it with 3.3V
actually, but anyway if current is zero, there is no voltage drop on the
resistor and the transistor is fully on, with 33ma flowing through the
transistor.

As the voltage current increases the voltage drop accross the resistor
increases, thereby lowering the effective voltage seen on the base
relative to the emitter. Once the voltage drop gets to 2.6V, the
transistor starts turning off, maintaining a level equilibrium as the
voltage continues to rise.

        V+
        |
       LED
        |
        C
3.3V---B<
        E
        |
        R
        |
       GND

'Cause I built one, and it seems to do all sorts of weird and wonderfull
things only somewhat related to current limiting... Not sure if I'm
understanding it correctly.

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHPa+S3bMhDbI9xWQRAtAqAJ9A7QsljQ1bSQrkGp4RBpZJRyeWhgCcDudB
njx4OWPB2Ja0AcGZn6aB38U=
=V/57
-----END PGP SIGNATURE-----

2007\11\16@100556 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, Nov 14, 2007 at 07:32:49PM -0800, Marcel Duchamp wrote:
{Quote hidden}

Well, gave it a try with a 2n3904, and the circuit works quite nicely.
Current regulation isn't world class by any means, but plenty adequate
for leds. The current limiting does seem to depend on R2 as well though,
10k vs. 100k changes the limited current by a few mA.

In actual use the current limiting kicks in when the voltage at the base
reaches 0.78 volts. Other references I found say Vbe is generally 0.7
volts. Finally on the 2n3904 datasheet, there are two graphs, Vbe
saturation and Vbe on, both of which are dependent on temperature and
collector current. Now my collector current, at 100k R2, should be way
less than 1ma, yet according to the datasheet's Vbe(on) graph even at
1ma Vbe is < 0.7V, (at 25C) and decreases as current goes down. This
isn't a device-to-device difference, I tried 2n3904's from different
mfgs and they both acted the same way.

What's going on there?

> I'll leave the design of the clever switch mode power supply to others;
> I'm out of town through the weekend for my daughters wedding.

Congrats!

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHPZlv3bMhDbI9xWQRAjAhAJ4iMAUZmyiRFCJVUMeS4LpY8byC+ACfbxDQ
MN2X6xIZxgnSyDTOznV1y9I=
=riIC
-----END PGP SIGNATURE-----

2007\11\16@102654 by olin piclist

face picon face
Timothy J. Weber wrote:
> Typo in comment "N must not be from 1 to 255" twice I think.

Thanks.  Fixed.

********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2007\11\16@125939 by Spehro Pefhany

picon face
Quoting Peter Todd <EraseMEpetespam_OUTspamTakeThisOuTpetertodd.ca>:

{Quote hidden}

It forms a current sink, but the voltage at the collector has to be
more than the base (approximately, it can be a bit less in fact, more
like 2.6V as you observed since Vce saturates at some tens of mV) for
it to work properly. If the condition you are mentioning can exist you
might want to add a base resistor of Re * 10 or so to keep the current
out of the PIC within reason.

The main difference between using Vbe and the above type of circuit is
that the unit-to-unit variation and temperature change in Vbe doesn't
impact the above type of circuit nearly as much, since the 0.6-0.7V is
swamped by the 3.3V voltage, so it's 5x better in performance. The
downside is that you waste a bit of power and drop a bit of voltage in
the resistor. If you want the best of both worlds you have to put more
circuitry in there, such as an op-amp to use a very low value resistor
(eg. 100mV drop), but then you might have to worry about response times
with the PWM. The transistor won't have any problems with the PWM
frequency and the circuit is very simple. Short-circuit proof too,
like any constant-current circuit.

Say, didn't someone around here mention something about a beer sometime,
Peter?

Best regards,
Spehro Pefhany
--
"it's the network..."                          "The Journey is the reward"
s...spamspam_OUTinterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com


2007\11\17@151430 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, Nov 16, 2007 at 09:33:02AM -0500, Timothy J. Weber wrote:
> - An interesting solution to the same power switch problem I was
> studying with this project: <http://timothyweber.org/mancala> (also a
> 2xAA switcher for a portable device).  Olin's solution uses less weird
> hardware and cleverer firmware.  I think I was obsessing about having
> "no" current flowing while off; he's settled for "negligible" and made a
> much more elegant solution in the process - fewer parts, more
> flexibility in what's required for power-on.  I like it.

I gotta agree with him too. I've designed two devices now that are fully
sealed after manufacture, and need a one time activation to do their
thing. After asking about no power latch circuits here on piclist, I
settled with just having a pic in sleep mode waiting for that first
button press. Both times I implemented the turn on with a reed switch
and the turn on logic was that the reed switch had to be held closed by
a magnet for 10s before ativation happened. The only problem I found was
reed switches need the magnet fairly close to them to actually activate,
for a device embedded in epoxy or acrylic it's tough to position them
close enough.

I'm thinking about ditching that setup for something a little more
involved, but without magnet, namely mechanical tilt switches. Rotate
the unit in a defined series of actions with timing. Problem is, as far
as I can tell buying mechanical tilt switches is damn near impossible
now. All digikey has is either very large ones for more industrial uses,
or smaller photo sensor/hall effect based ones with current consumptions
of at least 10uA, if not 20ma.


Hmm... Just occured to me though... If you don't want your sensor to
drain your battery, why not use a sensor that generates electricity for
you? As in... solar. Looking at digikey I just noticed they have tiny
solar cells in 8 and 16 pin SOIC packages for sensing and (according to
the datasheet) trickle charging. With a 50uA short circuit current that
big lead acid might take a few decades, but it would be enough to
bootstrap a uC. That, or use one with one of the above 10uA tilt
sensors! It'd make sure the device wasn't accidentally turned on during
shipping!


In any case, our Randy Glenn has one of those devices.. You turned it on
yet? :)

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHP0b83bMhDbI9xWQRAgQ8AJsE9+zvmk2ArXfVuW+vvjGlor3L5ACgkFj+
e5Wp0iqqasz9p7uK2Wp9kII=
=svJq
-----END PGP SIGNATURE-----

2007\11\17@151430 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, Nov 16, 2007 at 12:59:34PM -0500, Spehro Pefhany wrote:
{Quote hidden}

Ahh, good point, forgot about that. Could even be causing some of the
weirder stuff I was seeing as the amount of current that would be
flowing would end up being a big fraction of the total power
consumption.

I think the voltage at the base vs. the collector thing might have been
doing it. I noticed that after reaching a bit of a "plateau" in the
limiting, current started actually going *down* as I applied more
voltage to the leds. Then again, at one point it also looked like I was
getting 10mA of current to flow through a 1n4001 in the reverse
direction.

> The main difference between using Vbe and the above type of circuit is

Could you draw a schematic of what using Vbe looks like then? I don't
think I understand the difference between what I built, and what you
were describing.

{Quote hidden}

We did you know. I'm pretty busy this week, but drop me a line offlist
about the week after.

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHP0ub3bMhDbI9xWQRAmkkAJ4wA5/rDBGjUmgpwYkpZkXPY1HnTgCfU7Ez
qRo8RSUqm1wt6OE16JSHDEQ=
=KFbd
-----END PGP SIGNATURE-----

2007\11\17@153026 by Carl Denk

flavicon
face
The October 2007 issue of NUTS & VOLTS magazine has an article on tilt
switches. The tilt switches (non-mercury, uses ball) are available from
Electronix Express.

Peter Todd wrote:
{Quote hidden}

2007\11\17@163725 by wouter van ooijen

face picon face
> Problem is, as far as I can tell buying mechanical
> tilt switches is damn near impossible now.

http://www.goldmine-elec-products.com/prodinfo.asp?number=G16165

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu



2007\11\17@164013 by Timothy J. Weber

face picon face
Peter Todd wrote:
> Hmm... Just occured to me though... If you don't want your sensor to
> drain your battery, why not use a sensor that generates electricity for
> you? As in... solar. Looking at digikey I just noticed they have tiny
> solar cells in 8 and 16 pin SOIC packages for sensing and (according to
> the datasheet) trickle charging. With a 50uA short circuit current that
> big lead acid might take a few decades, but it would be enough to
> bootstrap a uC. That, or use one with one of the above 10uA tilt
> sensors! It'd make sure the device wasn't accidentally turned on during
> shipping!

I like it!  Low-pass the input and look for peaks (or troughs) in output
to indicate the shadow of a hand passing over the unit.  I want one that
you activate by waving your hand over it to the rhythm of the chorus to
"Yellow Submarine."  :)
--
Timothy J. Weber
http://timothyweber.org

2007\11\17@165332 by Jinx

face picon face
> Problem is, as far as I can tell buying mechanical tilt switches is
> damn near impossible now

Peter, I use the RBS 040200 ball switch in a product. Even at the
initial development stage they were cheap ($1.72) enough from RS
(361-5093). This RoHS alternative is a 1/4 the price of RS mercury
tilt switches

My pdf says they're made by Active Switch And Sensor Ltd. This
pdf doesn't

26kB

http://www.comus-intl.com/products/RBS040200.pdf


2007\11\17@180816 by Peter Todd

picon face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sat, Nov 17, 2007 at 04:40:06PM -0500, Timothy J. Weber wrote:
{Quote hidden}

I'd better be carefull there... I don't want to make a product that only
those skilled in Dance Dance Revolution can use...


Definetely spoke too soon on those tilt switches though, seems like
everyone but digikey has them. Guess which place I checked?

- --
http://petertodd.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHP3Rc3bMhDbI9xWQRAhxNAKCfTRATueLaaC7pn/FiLp9vOzDKtACcCGFl
Mc25oCV/+t3KVIR6Q2t+7GA=
=Fndu
-----END PGP SIGNATURE-----

2007\11\17@200947 by Timothy J. Weber

face picon face
Peter Todd wrote:
>> I like it!  Low-pass the input and look for peaks (or troughs) in output
>> to indicate the shadow of a hand passing over the unit.  I want one that
>> you activate by waving your hand over it to the rhythm of the chorus to
>> "Yellow Submarine."  :)
>
> I'd better be carefull there... I don't want to make a product that only
> those skilled in Dance Dance Revolution can use...

What, you think that wouldn't be a selling point??  :)

> Definetely spoke too soon on those tilt switches though, seems like
> everyone but digikey has them. Guess which place I checked?

Yeah, usually they're the comprehensive one.
--
Timothy J. Weber
http://timothyweber.org

2007\11\18@140718 by Spehro Pefhany

picon face
At 03:14 PM 11/17/2007, you wrote:

>Could you draw a schematic of what using Vbe looks like then? I don't
>think I understand the difference between what I built, and what you
>were describing.

I was referring to circuits such as the one Marcel suggested, which has
one Vbe drop across the 30R resistor, which might allow you to get one
more white LED in series because it has a couple of volts less minimum
voltage drop.


>We did you know. I'm pretty busy this week, but drop me a line offlist
>about the week after.

Sounds good.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
@spam@speffKILLspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com



More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...