Hi,

I've been using an Lm317 as a constant current source and it works, however, adjusting the current using a pot is useless (inverse relationship means last few degrees of pot movement result in massive change in current). How can i make a variable current source (0ma - 100ma) that i can adjust linearly? What i would like is to have a pot where when its fully turned down, the current is 0ma, when its fully turned up, 100ma, and at 50%, 50ma.

Any help on this would be appreciated, thanks.

ps, It doesn't need to use the Lm317, just need any way of linearly adjusting current.

 

What will be the resistance of your load? I=E/R, so that's why it's important to know your load's resistance. If your load will be in excess of 200 Ohms or so, you'll start getting into some design problems due to the higher voltages required to obtain the desired current flow.

Here's Wikipedia's entry on current sources:
http://en.wikipedia.org/wiki/Current_source

 

The circuit is just to test LEDs, so the load is relatively small.

I've had a look at Wikipedia, but to be honest, i can't make any sense of it as i'm pretty new to electronics, knowing which transistors or op-amps to use is above me.

I have a couple of bc547, bc558, bc327, and also a tl071 and an lm741, would I be able to use these?

 

You need a supply of about 15V to get a descent current ,

the formula is 1.25V/R= current, so if your looking for 100mA thats 12.5 ohms
so your variable resistor would be better if it was a 10 turn one say 100 ohms, like this type,

http://www.ebay.co.uk/itm/534-100R-...al_Components_Supplies_ET&hash=item5643d94580

that would give you current from 12.5mA to 1A.
its possible your trimmer pot is too large in value.

http://www.ae5d.com/images/Lim.png

 

Thanks Dodgydave, i've already tried that circuit with an lm317, but the trimmer does not vary the current linearly. If a standard pot has 270 degrees of rotation, then for every 2.7 degrees i turn the pot, i'd want 1ma of current.

 

Have you thought about an LM3914 or LM3915?

 

I can't access my library until daylight, but one circuit from the index stands out.

Practical Electronics May 2005 p322.

Adjustable current source : 100na to 10mA

 

If you want to adjust the current, refer to here that I'm using four different combination of transistors, you can choose anyone you like.

I haven't translate to english version, but I'm using google to make you more easier to understand, specially you just care about the current part of the circuits.

You can also move the current adjustment to the input of the LM317.

 

Below is a constant current source with a linear relation between pot position and current.

The transistor has a maximum rating of 100mA so you are pushing it with that setting. A higher current transistor such as a 2N2222 could be used.

The op amp shown is a single supply type. If you want to use one of your dual supply types, you will also need a negative supply voltage of at least 5V for those.

If you use a different positive supply voltage, you will need to change the value of R1 accordingly to maintain a full-scale current setting of 100mA (maximum pot voltage output of 1V).

Notice that the LED under test goes in the collector of the transistor.

 

Excellent circuit and it goes all the way to zero ma.

 

You will not be able to use your LM741 nor your TL071 for crutschow's proposed circuit, as neither can "see" within the required range (close to ground; the negative rail). Both the LM741 and the TL071 are considered "dual rail" opamps; as they can't see within about a volt of the positive supply, and the TL071 can't see within about 3v of the negative supply.

The LM741 can see closer to the negative rail than the TL071 and variants, but the 741 is such an old, slow design with terrible crossover problems, there really is not much of a reason to use one. There are LOTS of newer opamps with far better specifications on the market nowadays.

 

Thank you all for the help, much appreciated. Looks like I need to buy a different op-amp then. As I buy parts at maplin, my options are... LM358N, LM3900, TL081CN, NE5532N, TL084CN, TS358, NE5534AN, CA3140EZ, LF347N, TL082CN, LM13700N, TL074CN, TL072CN.
I've looked at the specs for each but can't work out if they will work?

 

You need one that will operate from a single supply voltage. I know the LM358N will. Don't know about the rest. The LM3900 is a Norton type (current input) op amp and is not suitable for that circuit.

 

Here's the simulation with the pot stepped from position 0 (%) to 1 (100%) in 0.1 (10%) steps, showing the linearity of current with respect to pot position.

Edit: For a reasonable accurate indication of current versus pot position, you could use a ten-turn pot with a suitable 10-turn dial indicator such as this.

 

Hi, What software did you use for this simulation ? Is it free?

Here's the simulation with the pot stepped from position 0 (%) to 1 (100%) in 0.1 (10%) steps, showing the linearity of current with respect to pot position.

Edit: For a reasonable accurate indication of current versus pot position, you could use a ten-turn pot with a suitable 10-turn dial indicator such as this.

View attachment 55495

 

Hi, What software did you use for this simulation ? Is it free?

I used LTspice, which many on these forums use, and it's free from Linear Technology.