current regulator -- explanation?

R

Rex

Guest
I found my way to this page today:
http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?
 
Rex wrote:

I found my way to this page today:
http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?
Click to expand...
The circuit in question is a current mirror. The idea is that the current
through the first transistor sets the same current in the second transistor
as a result of Vbe amtching.

As shown it's a poor way to drive LEDs ! Not least because you have to
'waste' as much current through the resistor as flows in the LEDs.

Graham
 
On Dec 6, 2:56 am, Rex <r...@yahoo.invalid> wrote:
I found my way to this page today:http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?
Click to expand...
Hi, Rex. Current mirrors are a good choice if you're making and IC,
because you can match the transistors well, and temperature tracking
between transistors is easy.

If you want an easy current source that has pretty good compliance
over a fairly wide range, you could do worse than this (view in fixed
font or M$ Notepad):

|
| VCC
| +
| |
| .-.
| | |
| | |
| '-'
| |
| |
| Vb |<
| <------| PNP
| |\
| |
| |
| |
| |
| ^
| Isource
|
| |
| ===
| GND
|
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

By controlling Vb, you can achieve pretty good control of current over
a fairly high voltage range with only one transistor and one resistor.

The emitter voltage will be one diode drop higher than the base
voltage. Let's say you wanted a current source of 20mA, and you had a
power supply that varied from 8 to 16VDC. You might do something like
this (view in fixed font or M$ Notepad):
|
| VCC VCC
| + +
| | |
| V .-.
| - | |36 ohm
| |1N914X2 | |
| | '-'
| V |
| - |
| | |<
| o-------| 2N3906
| | |\
| | |
| | |
| | |LED
| |3.3K |
| .-. V ~
| | | - ~
| | | |
| '-' |
| | |
| === ===
| GND GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

By adding the two diodes and the 3.3K resistor, you're ensuring that
Vb will be two diode drops below Vcc. That means Ve will be one diode
drop below Vcc. A 36 ohm resistor is chosen using Ohms Law:

0.7V / 0.02A = 35 ohms

This will not work perfectly, but it is an acceptable current source
for an LED.

Cheers
Chris
 
matched transistors are required to be truly accurate, might find an ic with
four matched pairs in some older manufacturer catalog

i prefer using those voltage regulators in current mode configuration,

http://en.wikipedia.org/wiki/Image:LM317_1A_ConstCurrent.jpg

many other voltage regulators can be configured to do this very easily with
only a single resistor


"Rex" <rex@yahoo.invalid> wrote in message
news:g3dfl3t46pfb5pa4gdp5j0jgj0f3jo8ju2@4ax.com...
I found my way to this page today:
http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?
Click to expand...
 
On Dec 6, 3:56 am, Rex <r...@yahoo.invalid> wrote:
I found my way to this page today:http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?
Click to expand...
I've had good results with another 2-transistor circuit, found here:
http://wolfstone.halloweenhost.com/Lighting/litled_LightEmittingDiodes.html
Scroll about 4/5 down the page, to the section titled "LED Tester".

No transistor-matching required. And the exact type of transistor
isn't too important, I use 2N3906. With that circuit, I can run two
20 mA LED (in series) for 20 hours on a 9V battery.

Mark
 
On Dec 7, 8:05 am, redbelly <redbell...@yahoo.com> wrote:
On Dec 6, 3:56 am, Rex <r...@yahoo.invalid> wrote:



I found my way to this page today:http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?

I've had good results with another 2-transistor circuit, found here:http://wolfstone.halloweenhost.com/Lighting/litled_LightEmittingDiode...
Scroll about 4/5 down the page, to the section titled "LED Tester".

No transistor-matching required. And the exact type of transistor
isn't too important, I use 2N3906. With that circuit, I can run two
20 mA LED (in series) for 20 hours on a 9V battery.

Mark
Click to expand...
Also, while that circuit calls for pnp transistors, you can build an
npn version if that's what you happen to have lying around.

If you reverse the polarity of the source, then just use two npn's in
place of the pnp's. Use the same configuration of the base,
collector, and emitter.

Mark
 
On Fri, 7 Dec 2007 05:10:42 -0800 (PST), redbelly
<redbelly98@yahoo.com> wrote:

On Dec 7, 8:05 am, redbelly <redbell...@yahoo.com> wrote:
On Dec 6, 3:56 am, Rex <r...@yahoo.invalid> wrote:



I found my way to this page today:http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?

I've had good results with another 2-transistor circuit, found here:http://wolfstone.halloweenhost.com/Lighting/litled_LightEmittingDiode...
Scroll about 4/5 down the page, to the section titled "LED Tester".

No transistor-matching required. And the exact type of transistor
isn't too important, I use 2N3906. With that circuit, I can run two
20 mA LED (in series) for 20 hours on a 9V battery.

Mark

Also, while that circuit calls for pnp transistors, you can build an
npn version if that's what you happen to have lying around.

If you reverse the polarity of the source, then just use two npn's in
place of the pnp's. Use the same configuration of the base,
collector, and emitter.
Click to expand...
---
And reverse the polarity of the LED.


--
JF
 
On Dec 7, 9:02 am, John Fields <jfie...@austininstruments.com> wrote:
On Fri, 7 Dec 2007 05:10:42 -0800 (PST), redbelly



redbell...@yahoo.com> wrote:
On Dec 7, 8:05 am, redbelly <redbell...@yahoo.com> wrote:
On Dec 6, 3:56 am, Rex <r...@yahoo.invalid> wrote:

I found my way to this page today:http://www.gizmology.net/LEDs.htm
looking for information about light output.

As I was reading, I found a circuit for regulating the current through
LEDs. It has two transistors with one having the base-collector shorted.
The circuit is about 2/3 down the page and has the caption,

"LEDs driven by a simple current regulator".

I don't see exactly how this senses the current through the LEDs and
regulates it. Does it work? If so, can someone explain or give a link to
how the circuit works.

Posted here because maybe this is basic stuff I have missed or am just
too dumb to see.

Assuming it does work (or even if not), is this a good implementation
(function/parts)? Is there a better way?

I've had good results with another 2-transistor circuit, found here:http://wolfstone.halloweenhost.com/Lighting/litled_LightEmittingDiode...
Scroll about 4/5 down the page, to the section titled "LED Tester".

No transistor-matching required. And the exact type of transistor
isn't too important, I use 2N3906. With that circuit, I can run two
20 mA LED (in series) for 20 hours on a 9V battery.

Mark

Also, while that circuit calls for pnp transistors, you can build an
npn version if that's what you happen to have lying around.

If you reverse the polarity of the source, then just use two npn's in
place of the pnp's. Use the same configuration of the base,
collector, and emitter.

---
And reverse the polarity of the LED.

--
JF
Click to expand...
Yes! Thanks John. -- Mark
 
On Thu, 06 Dec 2007 07:43:48 -0500, default <default@defaulter.net>
wrote:

http://www.discovercircuits.com/DJ-Circuits/constantcurretled1.htm

This is from Bob Pease (National Semiconductor) and drops very little
voltage (0.65 volts) to sense current
Click to expand...
Correction: 0.065 volts drop
--
---- Posted via Pronews.com - Premium Corporate Usenet News Provider ----
http://www.pronews.com offers corporate packages that have access to 100,000+ newsgroups
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top