开云体育

Patrick.

The following assumes that you will put your red LEDs in
series with the anode of the top one connected to a positive
supply of at least 12 volts (more volts = more LEDS).

The basis of the solution is a constant current sink. It
uses one small zener diode (5.1 volts), any general purpose
NPN transistor with a hFE of at least 100 (common 2N2222
will do), two resistors (1K and 270 ohms) and a small pot
(500 ohms, linear) to adjust the current between about 6 mA
and 16 mA.

Connect the 1K resistor between your +12volt rail and the
base of the transistor. Connect the cathode of the zener to
the base of the transistor and its anode to ground. Connect
the emitter to one side of the 270 ohm resistor. the other
side of the 270 ohms goes to one side of the pot. The other
side of the pot goes to ground. The wiper goes to either the
ground or the junction of the pot and the 270 ohm resistor.
The collector of the transistor goes to the cathode end of
the LED series.

When the pot is adjusted so that its value is zero, the
current flowing through the LEDs is (5.1 - 0.6)/270 = 16.6
mA. When the pot is at maximum resistance, the current will
be (5.1 - 0.6)/770 = 5.8 mA. This should give you a
reasonable range of brightness.

Regards,

Ross McKenzie ValuSoft Melbourne Australia

Nice idea. Another concept is to use a PWM circuit. Essentially, you
use a Transistor (or several) as on/off switches, varying the amount
of time the circuit is "on." I do not have a fully formed circuit I
can show you, but the pieces should be easy to find on the 'net:

1) Take a 555 circuit that can give you an adjustable on / off time.
A quick search revealed:



check out the last circuit on the page and just replace R1 & R2 with
your POT (hint: wiper would be the point between R1 & R2). Circuit
just has a 555, a POT, 2 Caps, and some wire.

2) Attach the output of this multivibrator to a FET (or similar)
circuit that is rated to supply the current you need.

3) Attach the FET to the LED's in parallel, with individual
resistors from the LED to GND.

4) Then, as you turn the POT, the LEDs will get dimmer and brighter.


NOTES:
- I am glossing over a lot of details. However, this type of circuit
will be much more efficient than dropping a decent amount of voltage
across a POT (typically rated at 1/8 watt or maybe 1/4).
- Make sure to calculate the right values for your resistors from
the LEDs to GND. At most, your duty cycle should be On 50% of the
time with the multivibrator (although theoretically, you will never
actually reach 50%). So, figure that IF you want the equivalent
brightness of 20mA, you need to calculate the resistors to give you
40mA (since it will be off half the time, at least) given the supply
voltage. At the other end of the swing, it will be off virtually all
the time.


Just an idea...

Bob H.

P.S. Others will probably criticize this approach. Listen to them
carefully, because they know more about this than I do.

Patrick,

A simple and very stable current source can be made with a regulator

LM317 family. A single resistor sets the current (and in fact, the

uses a BJT internally to control the current, so BJTs aren't

worse than JFETs or MOSFETs for the job). The application circuit

many LM317 (and other 3-terminal regulator) datasheets. By the way,

current sink is desired rather than a current source, the LM317

replaced with an LM337.

Good luck. You can build a similar circuit with an op-amp, a

and a pass transistor, but why go through all the trouble when an

all these components "built in"?

Jeff


--- In Electronics_101@..., "Patrick" <patrick49418@>

I tried to search through previous messages to find what I was
looking for; but, nothing seemed to be what I needed. Also, I

through the files for this group and nothing there helped me

I want to control the amount of current to a string of LEDs

controlling the brightness of the LEDs. Simple POTs will not

due to their low power ratings. To my understanding, BJTs are

much on and off; there really isn't anything in between(I think).
I've begun to read about MOSFETs and JFETs in my books. Those seem

be what I am looking for. Basically, I want to use a POT
to 'manipulate' a transistor thereby controlling the current

the transistor which will in turn go to the string of LEDs. (I've
gotten quite confused about transistors; it's hard to teach myself
just by reading my books.)
Can anyone point me in the right direction? Am I right about BJTs
being mostly On and Off? Mostly used for switching?
Umm, can't think of much else to tell ya'. I'm pretty sure for my
application I'll have to use the TO-220 package since it will have
higher power ratings.

That's really all I'm trying to do at the moment - trying to

the current to strings of LEDs without having all of their current

through a POT. Also, I have to use basic components, no
microcontrollers or BASIC Stamps. Could someone specify a
[transistor] part number that may work for my application so I

place to start poking around?

Thanks in advance.

-PatMan

Hi John
I would just like to make this point, in case you are using your

any high frequency AM/FM circuit there may be issues because zener is a
noisy component.

On such a case usage of preset pots(for precision setting) as voltage
divider network may be useful.

Regards
Gaurav


On 10/8/06, Bob Hyland-PMP <BobHyland@...> wrote:

--- John Baker <johnbaker_erie_pa@> wrote:

First of all, thanks for everyone's quick responses. I thought
using a voltage divider would take a toll on my 9V cell. Thanks
for clearing that up. I don't really want to go the route of
buying another chip...but I haven't put any thought into diodes. I
like the zener idea with a virtual ground. I'll give that method a
go.

Thanks again!
John

*snip*
John:

Here are some single supply circuits, "A Single-Supply Op-Amp Circuit
Collection":



Bob H. (Google Master...)

The first one from QST is actually a bit more along the lines of what I
am looking to make, though I'm looking for something a bit more "elegant"
with 2 channels, multiple ranges, and an ability to automatically alternate
between channels or sweep through the ranges to assist in testing. Think
"poor mans Huntron". Either of those would probably suite Gaurav, the
original poster well though.

One thing that amazed me about curve tracers once I discovered them is
that they are so simple to make and so usefull once you know how to use one.
No one ever even mentioned a curve tracer, let alone taught them durring any
of my training. But then again there isn't all that much component level
trouble shooting anymore. Everything is modular, insanely tiny, or potted to
prevent repair.

----- Original Message -----
From: Bob Hyland-PMP
To: Electronics_101@... <Electronics_101%40yahoogroups.com>
Sent: Monday, October 09, 2006 12:17 AM
Subject: [Electronics_101] Re: Question around sound card curve tracer for
William Kroyer

--- In Electronics_101@... <Electronics_101%40yahoogroups.com>,
"Gaurav Verma"
<gaurav.verma.mca@...> wrote:

Hi
Just the way we have PC based scopes are there any PC based curve
tracers available that one may purchase for there work?

I have been unable to find something I would like. Anny
suggessions will be welcome.
Gaurav

I may have missed part of this discussion. However, curve tracers
are straight forward to make if you have a 2 input scope with X-Y
mode. Many of the sound card O-Scope software can do X-Y. Certainly
all the PC Scopes I have seen do.

Anyway, here is a full project to do just what I think you are
trying to do for less than $2 ("A Low Cost Automatic Curve Tracer"):



Here is a generic Curve Tracer circuit for use with any X-Y scope:



Bob H.

Patrick.

The following assumes that you will put your red LEDs in
series with the anode of the top one connected to a positive
supply of at least 12 volts (more volts = more LEDS).

The basis of the solution is a constant current sink. It
uses one small zener diode (5.1 volts), any general purpose
NPN transistor with a hFE of at least 100 (common 2N2222
will do), two resistors (1K and 270 ohms) and a small pot
(500 ohms, linear) to adjust the current between about 6 mA
and 16 mA.

Connect the 1K resistor between your +12volt rail and the
base of the transistor. Connect the cathode of the zener to
the base of the transistor and its anode to ground. Connect
the emitter to one side of the 270 ohm resistor. the other
side of the 270 ohms goes to one side of the pot. The other
side of the pot goes to ground. The wiper goes to either the
ground or the junction of the pot and the 270 ohm resistor.
The collector of the transistor goes to the cathode end of
the LED series.

When the pot is adjusted so that its value is zero, the
current flowing through the LEDs is (5.1 - 0.6)/270 = 16.6
mA. When the pot is at maximum resistance, the current will
be (5.1 - 0.6)/770 = 5.8 mA. This should give you a
reasonable range of brightness.

Regards,

Ross McKenzie ValuSoft Melbourne Australia

Hi
Just the way we have PC based scopes are there any PC based curve
tracers available that one may purchase for there work?

I have been unable to find something I would like. Anny
suggessions will be welcome.
Gaurav

First of all, thanks for everyone's quick responses. I thought
using a voltage divider would take a toll on my 9V cell. Thanks
for clearing that up. I don't really want to go the route of
buying another chip...but I haven't put any thought into diodes. I
like the zener idea with a virtual ground. I'll give that method a
go.

Thanks again!
John

I want to control the amount of current to a string of LEDs thereby controlling the brightness of the LEDs. Simple POTs will not suffice due to their low power ratings. To my understanding, BJTs are pretty much on and off; there really isn't anything in between(I think).

I've begun to read about MOSFETs and JFETs in my books. Those seem to be what I am looking for. Basically, I want to use a POT to 'manipulate' a transistor thereby controlling the current through the transistor which will in turn go to the string of LEDs. (I've gotten quite confused about transistors; it's hard to teach myself just by reading my books.)

Can anyone point me in the right direction? Am I right about BJTs being mostly On and Off? Mostly used for switching?

Umm, can't think of much else to tell ya'. I'm pretty sure for my application I'll have to use the TO-220 package since it will have higher power ratings.