any way to get two LEDs to blend into one another?

Michael · Jul 24, 2003

Hi - I'd like to have two LEDs (a white and a blue) throb in a sort of
strange way. Let's say the white one is at full brightness, I'd like the
blue to be completely dark. The white would slowly start fading, and the
blue would start lighting up. At the point where the blue was fully lit,
the white would be dark. And so on and so on. How best would one setup this
circuit? Thanks!

Mario Trams · Jul 24, 2003

Michael said:
Hi - I'd like to have two LEDs (a white and a blue) throb in a sort of
strange way. Let's say the white one is at full brightness, I'd like the
blue to be completely dark. The white would slowly start fading, and the
blue would start lighting up. At the point where the blue was fully lit,
the white would be dark. And so on and so on. How best would one setup
this circuit? Thanks!

You could do that easily with a micro controller and control LED
brightness via pulse width modulation.

Alternatively you might use an OpAmp circuitry that produces a
waveform like that:

/\ /\
/ \ / \
/ \ / \
/ \ / \
/ \ / \
/ \/ \

(Of course, with the frequence stretched to your requirements)
You would control one LED with the signal above, and the other
LED with the inverted wave form.

Recently I had to produce a circuitry producing the waveform
above. If you want, I can send a picture of the schematics by
Email (I'm too lazy to produce some ASCII-art here...)

Regards,
Mario

Sir Charles W. Shults III · Jul 24, 2003

Yes! There is a simple way to do this.
You will use a pair of 555 chips, but a little differently than you normally
would. Place the two LEDs in anti-parallel- anode to cathode, cathode to anode.
Now, when current flows one way, only one will light, and vice versa.
Now make two 555 astables, but set them to run very close to each other in
frequency. Connect your LED pair from the output of one to the output of the
other- a funny thing will happen. The two square waves will "beat" against each
other and make the light from each fade in opposite directions.
So it will slowly go from white to blue, and back again. If there is not
enough drive current, you can make a dual transistor driver to hand that easily
enough.

Cheers!

Chip Shults
My robotics, space and CGI web page - http://home.cfl.rr.com/aichip

Spehro Pefhany · Jul 24, 2003

Hi - I'd like to have two LEDs (a white and a blue) throb in a sort of
strange way. Let's say the white one is at full brightness, I'd like the
blue to be completely dark. The white would slowly start fading, and the
blue would start lighting up. At the point where the blue was fully lit,
the white would be dark. And so on and so on. How best would one setup this
circuit? Thanks!

A couple of LMC555's, plus one resistor and capactor each, to make two
quasi-sawtooth waves, one operating at the "throb" frequency, and one
at a few kHz (PWM frequency). Resistor from pin 3 to 6,2, capacitor to
ground.

And an LM393 dual comparator, each comparator output sinking current
from one of the LEDs, and a common resistor from the LED anodes to
Vcc. Each comparator input goes to one timing capacitor, one is
reversed from the other. 8 parts total, plus the LEDs.

Of course a micro could be used too, but that would involve
programming.

Best regards,
Spehro Pefhany

hvac200 · Jul 24, 2003

Go here http://ourworld.compuserve.com/homepages/Bill_Bowden/ and
look for Fading Red Eyes in the led section.

Michael · Jul 24, 2003

http://ourworld.compuserve.com/homepages/Bill_Bowden/

I must admit I'm not following your logic. I can't see how I would use that
circuit for this.

Michael · Jul 24, 2003

Mario Trams said:
You could do that easily with a micro controller and control LED
brightness via pulse width modulation.

Alternatively you might use an OpAmp circuitry that produces a
waveform like that:

/\ /\
/ \ / \
/ \ / \
/ \ / \
/ \ / \
/ \/ \

(Of course, with the frequence stretched to your requirements)
You would control one LED with the signal above, and the other
LED with the inverted wave form.

Recently I had to produce a circuitry producing the waveform
above. If you want, I can send a picture of the schematics by
Email (I'm too lazy to produce some ASCII-art here...)

Regards,
Mario

Well, rigth now your solution sounds the smallest - and size is a
serious factor for this circuit (so serious a factor in fact that every
component is surface mount) so could you please e-mail that circuit to
me? My E-mail address is nleahcim at comcast dot net. Thanks!

Frank Bemelman · Jul 24, 2003

"Michael" <[email protected]> schreef in bericht

[snip]

Well, rigth now your solution sounds the smallest - and size is a
serious factor for this circuit (so serious a factor in fact that every
component is surface mount) so could you please e-mail that circuit to
me? My E-mail address is nleahcim at comcast dot net. Thanks!

The smallest solution is a PIC, 12F629. 8 pin soic. You need
one current limiting resistor, decoupling cap and of course
your 2 leds.

It must be group-wide amnesia/dementia that it has not been
mentioned yet <G>

Frank Bemelman · Jul 24, 2003

Jonathan Kirwan said:
"Michael" <[email protected]> schreef in bericht

[snip]

Well, rigth now your solution sounds the smallest - and size is a
serious factor for this circuit (so serious a factor in fact that every
component is surface mount) so could you please e-mail that circuit to
me? My E-mail address is nleahcim at comcast dot net. Thanks!

Click to expand...

The smallest solution is a PIC, 12F629. 8 pin soic. You need
one current limiting resistor, decoupling cap and of course
your 2 leds.

It must be group-wide amnesia/dementia that it has not been
mentioned yet <G>

Click to expand...

Mario *did* mention something like that, actually. And then
went on to something else. But there may be an analog-first,
micros are for the heathens in comp.arch.embedded, mentality
operating as well.

Oops, missed that. It would be nice to try it analog, but for this
application I'd really go for a PIC. The analog versions tend
to use a triangular wave, but I'm not so sure if that gives
a pleasant and smooth fade-over, the human eye being what it is.
With the PIC you can use a LUT and tweak it for a satisfying
effect.

Spehro Pefhany · Jul 25, 2003

Jonathan Kirwan said:
Jonathan Kirwan said:

"Michael" <[email protected]> schreef in bericht

[snip]

Well, rigth now your solution sounds the smallest - and size is a
serious factor for this circuit (so serious a factor in fact that every
component is surface mount) so could you please e-mail that circuit to
me? My E-mail address is nleahcim at comcast dot net. Thanks!

The smallest solution is a PIC, 12F629. 8 pin soic. You need
one current limiting resistor, decoupling cap and of course
your 2 leds.

It must be group-wide amnesia/dementia that it has not been
mentioned yet <G>

Click to expand...

Mario *did* mention something like that, actually. And then
went on to something else. But there may be an analog-first,
micros are for the heathens in comp.arch.embedded, mentality
operating as well.

Click to expand...

Oops, missed that. It would be nice to try it analog, but for this
application I'd really go for a PIC. The analog versions tend
to use a triangular wave, but I'm not so sure if that gives
a pleasant and smooth fade-over, the human eye being what it is.
With the PIC you can use a LUT and tweak it for a satisfying
effect.

I mentioned it too. Who's the demented amnesiac here?

Best regards,
Spehro Pefhany

Scott McDonnell · Jul 25, 2003

Look at the link again. It is EXACTLY what you were asking for...don't click
on the FADING RED EYES link on the page, just scroll down and look for it.
An excerpt:

"This circuit can be used to slowly illuminate and fade a pair of LEDs, or
with the addition of a second transistor (PNP) the circuit will fade two
pairs of LEDs out of phase so that one pair gets brighter while the other
grows dimmer."

This is what you asked to do, correct?

Scott

Mario Trams · Jul 25, 2003

Hi Michael,

Well, rigth now your solution sounds the smallest - and size is a
serious factor for this circuit (so serious a factor in fact that every
component is surface mount) so could you please e-mail that circuit to
me? My E-mail address is nleahcim at comcast dot net. Thanks!

Thanks for coinsidering my suggestion ;-)
I put the schematic here:
http://www-user.tu-chemnitz.de/~mtr/zeugs/sawtooth_osc.jpg

At least you have to increase C1 and/or R9 (or R1+R25 appropriately).
And also do not forget to select a right buffer OpAmp (for IC1D) that
is able to provide the current needed by your LEDs.

And for your second LED you have to use an inverting Buffer (i.e.
+ connected to Ground, say 20K between out and -, and 20K between
- and output of IC1B.

Have fun!
Mario

PS: If you really need an ultra-small design, you should really
consider the micro controller solution mentioned by Frank and Spehro.
Of course, it is "shooting with cannons on birds", but who cares...

--
----------------------------------------------------------------------
Digital Force / Mario Trams [email protected]-chemnitz.de
[email protected]
Chemnitz University of Technology http://www.tu-chemnitz.de/~mtr
Dept. of Computer Science Tel.: (+49) 371 531 1660
Chair of Computer Architecture Fax.: (+49) 371 531 1818
----------------------------------------------------------------------

Frank Bemelman · Jul 25, 2003

Spehro Pefhany said:
I mentioned it too. Who's the demented amnesiac here?

Not only that... am I now dyslectic too?

hahaha... let's forget it

Fred Bloggs · Jul 25, 2003

Hi - I'd like to have two LEDs (a white and a blue) throb in a sort of
strange way.

Throb?- sounds like a porn thing.

Let's say the white one is at full brightness, I'd like the
blue to be completely dark. The white would slowly start fading, and the
blue would start lighting up. At the point where the blue was fully lit,
the white would be dark.

The simplest way is probably to use a 555 timer- since you say the
alternation frequency is slow, the CMOS LMC555 is best. A circuit for 9V
is shown below:

Please view in a fixed-width font such as Courier.

~
+------+----------+----+----------------+-----------+---|>|---+
| | | | | | Db |
| | +--+----+-----+ | | blue \
| \ | RST VCC | / | 470
| 1M | | 1.5M 4.7k | Rb \
| Rd \ | OUT+-/\/\-+ / | /
| / | | Rt | \ | |
| | | LMC555 | | | c e
| | | | | | |/ |/
| | | | +-------+----+----+---------+
--- | | | | | | | n2|\ p2|\
- 9V | | | | | | | e c
| | +--+TRIG | | | e | | |
| | | | | | | |/ | | +-----+
| +----+--+THRESH | | +-+p1 | | |
| | | | | | | |\ | +---|-----+
| | | | DIS GND | | | c c | |
| \ | +--+----+-----+ | Ct| | |/ | \
| 1M | | | | === +--+n1 | 470
| Rd \ | | | 330k | 1u |\ | Rw \
| / +-------------/\/\--+ FILM e | /
| | | | Rd1 | | | ~ |
+------+----------+----+------------+--------+----------+-|<|-+
Dw
n=2N4401 p=2N4403 white

The alternation period will be 2.5 x Rt x Ct = 3.75 seconds for the
values shown. Increase Rt and Ct for indefinitely longer periods, use no
larger than 10M for Rt and an inexpensive metallized polyester film
capacitor for Ct. p1 and n1 buffer the capacitor voltage onto the
parallel connected bases of n2 and p2 which are complementary emitter
follower drives for the LEDs so that the LED currents have the required
opposite phase. Since you are using blue/white LEDS, the current
limiting resistors, Rb and Rw, are sized for approximately 5mA at a
forward LED voltage of 3V. The THR/TRIG bias network, consisting of 2x
Rd and Rd1, is there to ensure complete LED turn-off and is optional.
The values shown modify the effective 555 switching thresholds to be
approximately 2V of each rail which ensures that each LED is turned off,
or very dim, for practical purposes. In the case of lower voltages, the
bias network may not be necessary so you can remove the 2x Rd and
replace Rd1 with a short to the THR/TRIG inputs.

Watson A.Name - 'Watt Sun' · Jul 27, 2003

Or use this circuit.

http://ourworld.compuserve.com/homepages/bill_bowden/page5.htm#eyes.gif

Use the bottom circuit of the fading red eyes.

[snip]

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS? Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted. *All* email sent to it
goes directly to the trash unless you add NOSPAM in the
Subject: line with other stuff. alondra101 <at> hotmail.com
Don't be ripped off by the big book dealers. Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html
@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

Watson A.Name - 'Watt Sun' · Jul 27, 2003

I built a circuit aimed at doing this only yesterday - LM324, 2 sections as
quasi-sawtooth oscillators, one 2kHz, one variable around 0.1Hz-2Hz, with
feeds from each to the other 2 sections used as comparators, reversed as you
describe. Outputs switched small NPN transistors with leds as load.
Schematic at http://www.cpemma.co.uk/graphics/ult_throbber_sch.gif

Looked fine in simulation, but real results were disappointing - despite 1%
resistors to try to get oscillator amplitudes identical, one led faded to
(past) fully off, the other to only almost off.

Maybe 2 kHz is too fast for the opamp. Didja try slowing down the
upper oscillator, with a bigger cap? LM324s and LM358s are known for
not having a lot of output current, and for an asymmetrical output.
They recommended putting a (6.8k?) resistor to ground to help with
this problem.

I'll try your 555 version.

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS? Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted. *All* email sent to it
goes directly to the trash unless you add NOSPAM in the
Subject: line with other stuff. alondra101 <at> hotmail.com
Don't be ripped off by the big book dealers. Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html
@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

Fred Bloggs · Jul 31, 2003

cpemma said:
I built a circuit aimed at doing this only yesterday - LM324, 2 sections as
quasi-sawtooth oscillators, one 2kHz, one variable around 0.1Hz-2Hz, with
feeds from each to the other 2 sections used as comparators, reversed as you
describe. Outputs switched small NPN transistors with leds as load.
Schematic at http://www.cpemma.co.uk/graphics/ult_throbber_sch.gif

Looked fine in simulation, but real results were disappointing - despite 1%
resistors to try to get oscillator amplitudes identical, one led faded to
(past) fully off, the other to only almost off.

Right it makes better sense to use the two transistors as a Schmitt,
with switchpoint centered on OA span, and single-ended drive like so:

Please view in a fixed-width font such as
Courier.

12V
|
+------------+--+--+
| | |
| / /
| 390 390
| / /
| \ \
| | |
| +-----|-----+
| | | |
| L L |
| E E /
| D D 3.6k
| \ | | | /
Vmod-|+ \| c c \
| \ 10k |/ \| |
| >--/\/\--b| |b--+
| / |\ /| |
Vosc-|- /| e e |
| / | | | |
| +--+--+ |
| | |
| / /
| 560 4.7k
| / /
| \ \
| | |
+---------------+--------+
|
-+-
0V

The speed of this device puts the possibility of undefined metastable
operation into the ppb of an alternation period, eliminates an OA, and
adds just two resistors.

The modulation, Vmod, has to overdrive the switching oscillator
amplitude, Vosc, for complete turn-on and turn-off of the LEDs. The
simplest way to do this is like so:

Please view in a fixed-width font such as Courier.

12V 12V
100k | 100k |
+--/\/\---+ +--/\/\---+
| | | |
| 100k | | 100k |
+--/\/\---|----+ +--/\/\---|----+
| | | | | /
| | | | | RA
100k | | \ | | 100k | | \ | / 50k
+-/\/\-+-----|+ \| | +-/\/\-+-----|+ \| \ pot
| | \ | | | \ |
| | >---+ | | >---+
| | / | | | / |
| +---|- /| / | +---|- /| /
| | | / | Rt | | | / | 47k
| | | / | | | /
| | | \ | | | \
| | | | | | | |
| +------------+--> Vmod | +------------+--> Vosc
| Ct | | | 15n | |
| === | | === |
| | | | | |
+--------+-------+ +--------+-------+
| |
--- ---

Adjust RA to attenuate Vosc below Vmod, for best effect. With
Vosc/Vmod=0.8 then each LED remains full OFF for 20% of the time etc-
the adjustment is slight.

Fred Bloggs · Aug 2, 2003

cpemma said:
Thanks, I'll try that

Now that the unnecessary dual OA LED drive is eliminated, it is possible
to reduce the entire circuit to the aforementioned Schmitt plus a single
dual-OA like the LM358. But I see no advantage over the single 555
analog drive- especially when it's for a simple display.

Moore's Lobby Podcast

Menu

Categories

Platforms

Content

Connect With Us

Network

any way to get two LEDs to blend into one another?

any way to get two LEDs to blend into one another?

Michael

Mario Trams

Sir Charles W. Shults III

Spehro Pefhany

hvac200

Michael

Michael

Frank Bemelman

Frank Bemelman

Spehro Pefhany

Scott McDonnell

Mario Trams

Frank Bemelman

Fred Bloggs

Watson A.Name - 'Watt Sun'

Watson A.Name - 'Watt Sun'

Fred Bloggs

Fred Bloggs

Similar threads