Current source

[k7elp60]

I have been using a single transistor circuit like windmiller posted. I use as a load bank for testing DC power supplies or to discharge batteries at a constant current. I chose a RE with a control voltage VIH of 0 to 5 volts ,I can select from near 0 current to about 10 amps. I have use darlington NPN transistors and N channel powermosfets for the transistor. Of course my transistor is on a large heatsink. I used a 741 opamp. <p>I can visualize with the proper RE the 0 to 10 volts would adjust the current from 0 to 250Ma.

[fsdenis]

Gentlemen,<p>To move the image from my website to post it here
I pushed the IMAGE button below in the POST REPLY
area and then entered the exact image address on
my website. <p>Instructions come up when you push the image button to tell you what to do.<p>I'll try it again just to be sure. Sorry for the
reply delay. I've been in the wilderness for a few days.

[fsdenis]

<p>OK. Here is the image again. I pushed the IMAGE button and entered:<p>http://paralleltransistors.com/ed0-2_fin.jpg<p>that's ed zero, not ed "O". and _ not -.<p>You guys are welcome to use this website image to practice if you wish. A little clutter won't hurt this thread much.<p>[ September 21, 2004: Message edited by: windmiller ]</p>

[fsdenis]

By the way, I've taken the paralled current source circuit out past 50 amps. Each LM324 has four opamps in it.<p> I've found no limit to the number of channels possible to assemble in parallel nor upper limit to total current.<p>Of course, the assembly begins to resemble a spagetti factory after you have several LM324 chips tied to a lot of transistors mounted on an aluminum yardstick as heatsink.

[fsdenis]

Forgot to mention that I invented (so far as I know) this paralleling scheme for multiple channels of current controlled circuits in June
2003. My drawing is copyrighted by me, Fred Denis.

[erkanyigiter]

windmiller: For a 0-2A current source do i need parallel stages or can i use only one opamp and transistor with 0,5 ohm R instead of 1 ohm R?. is there a difference?

[fsdenis]

yigiter:<p>If you wish to use Re = 0.5 ohms and you want 0-2A out and you wish to use a single op amp and a single transistor, you can make this work.<p>Vin = Ic * Re, so for 2V out you would have Vin = 1V.<p>You will probably want to use a darlington transistor so the opamp will put out enough base current to drive the transistor to 2A. Darlingtons come with current gains in the neighborhood of 1000, so the roughly 20mA out of the opamp will easily drive the darlington to 2A.<p>And mind the transistor heating. You might want to use a darlington transistor in a TO3 package and put a heat sink on it, depending on how much power your application will dissipate in the transistor.<p>The parallel approach shown in the above diagram is what I usually use as it is easy to build and adapt to just about any application. <p>1)Any total maximum current can be handled.<p>2)Slew rates can be very fast, limited only by the opamp or the transistor, whichever is slowest.
Choice of opamp, transistor, Re and Vin is whatever suits the application requirements.<p>Note that the darlington choice we are discussing will be fairly slow to respond to changes in Vin, but if you only need a manually adjusted voltage controlled current supply here, it should be plenty fast.<p>3)I will also usually give the opamp its own isolated +/- 5V and mid-0V power supply so I can easily put the current control circuit anywhere in any load circuit without worries about biasing, relating it to the load supply, mysterious ground loops and so on.<p>3)And the parallel current controllers approach often allows me to dodge the need

[fsdenis]

yigiter continued after modem dropout. Sorry:<p>3)And the parallel current controllers approach often allows me to dodge the need for expensive heatsinks. Usually opamps and transistors are MUCH cheaper than heatsinks and the transistors can be compactly mounted off-board. This eliminates need to use expensive board space for transistor mounting.<p>There are many ways to skin the voltage controlled current power supply problem, but I like the parallel approach as very general purpose, easy to design, build and debug. And conceptually very simple. When I've been away from electronics for a few months, it is a comfort to know I can just build this and stay out of trouble with intricate biasing, relating and other mysterious problems.

[erkanyigiter]

windmiller:
thanks for your helps. it was very clear....

[erkanyigiter]

a question about current source again....
i have tried it and it works perfectly. but if i have a grounded load, what can i do? i can not find a solution, any idea?<p>-Erkan

[fsdenis]

yigiter:<p>Yes. You can easily operate your current regulator
(current source) circuit anywhere in the load
circuit by giving the current source circuit it's
own ISOLATED +5V, 0V, and -5V power supply.<p>Say your load is grounded (connected to the
negative terminal of the battery shown in my
diagram. Then:<p>1)Connect the bottom of Re to the top (positive)
side of your load.<p>2)Keep your current source circuit as before, except for new connections to the isolated power supply:<p> 1)Connect the bottom end of Re to 0V of your isolated power supply. (Re is also connected to the top of your load, as above.)<p> 2)Connect opamp + (power supply, not input) to the +5V of the isolated supply.<p> 3)Connect opamp - (power supply, not input) to the -5V of the isolated supply.<p> 4)Connect your Vin governing circuit to between
0V and +5V of your isolated power supply.<p>This is not as complicated as it might look. All
you are doing is connecting the transistor collector to the positive terminal of the battery and the bottom of Re to your load. All other connections are to the isolated power supply used to operate the current source.<p>Is any of this comprehensible?

[erkanyigiter]

i think i can not do this. because my grounded load is a "car battery" so i can't connect the positive side of my load to 0V.(12V->0V) and also i have not got a isolated power supply. i have only a 24V dc source, thats all.<p>(sorry if i misunderstand your schematic)<p>-Erkan

[fsdenis]

<p>yigiter:<p>I've done a drawing for you, trying to make more
clear that an isolated power supply can make it
possible to put the current regulator anywhere
in your load circuit.<p>Do you know what an isolated power supply is?

[Ron H]

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by yigiter:
i think i can not do this. because my grounded load is a "car battery" so i can't connect the positive side of my load to 0V.(12V->0V) and also i have not got a isolated power supply. i have only a 24V dc source, thats all.<p>(sorry if i misunderstand your schematic)<p>-Erkan<hr></blockquote>
There are other ways to do this. Do you need to control this current with a ground-referenced voltage, or can you use a pot referenced to +24V?

[fsdenis]

yigiter:<p>If I understand your last post:<p>1) you are trying to build a 12V (adjustable current) battery charger.<p>2) you have a 24V source separate from this 12V
battery.<p>3) you do not have a transformer input electronic
power supply (isolated) to work with here.<p>Is the 12V battery mounted and connected in a car
or may your charging circuit work with the battery
out of the car?