I am new to electrical engineering and I would like to build simple circuit with potentiometer as voltage regulator.

I have simple circuit, as you can see in attached picture below.


I have 12V/1000 mA DC source connected via potentiometer to LED strip (unfortunately I don't know the resistance of this component, strip is about 1.5m long - type 5050 if it will help somehow)

I would like to know what kind of POT should I use to regulate source to range 6 - 12V. At the moment I have chosen 1k Ohm POT and it dims LED strip only in first 10% of range then nothing happens at all because strip barelly shine.


What you are doing is not voltage regulation, but rather adjusting the resistance in series with the LEDs. This is a fine way to adjust brightness in principle - though it won't be linear, as you've observed - except for the fact that a potentiometer is not rated to carry the sort of current that flows through an LED strip. Potentiometers tend to be rated at 100mA or less, but 1.5 meters of LEDs will draw a lot more than that.

Your best option for controlling brightness fairly linearly is an adjustable constant-current supply - such as one like this:

enter image description here

An alternative that solves the power dissipation issue but not the linearity issue is to use a linear regulator like the LM317 with a potentiometer in the feedback loop - see that part's datasheet for an example.

  • 1
    \$\begingroup\$ BJT would be more adequate ;) although, the best is off-the-shelf led driver. \$\endgroup\$ – Gregory Kornblum Jul 10 '15 at 10:23
  • \$\begingroup\$ @GregoryKornblum A BJT on its own will have a high dependence on its beta, which doesn't make for a really reliable regulator. \$\endgroup\$ – Nick Johnson Jul 10 '15 at 10:27
  • \$\begingroup\$ But you made a current sense resistor! The circuit will almost not depend on beta. On the other hand, the control circuit will be much more linear, while with MOSFET you will have hard time with stability and regulation. Trust me, i did this mistake once :) \$\endgroup\$ – Gregory Kornblum Jul 10 '15 at 10:30
  • 1
    \$\begingroup\$ @GregoryKornblum Oh, sorry, I thought you were suggesting using a BJT on its own as a current regulator, not replacing the FET with a BJT. I actually manufacture a series of dummy loads for hobbyists that use this exact circuit, and with a little compensation it's perfectly stable. \$\endgroup\$ – Nick Johnson Jul 10 '15 at 10:55
  • 1
    \$\begingroup\$ @GregoryKornblum BTS117/BTS133/BTS141. They're rated for linear operation, and include various protections too. \$\endgroup\$ – Nick Johnson Jul 10 '15 at 12:10

First, note that this way you are not making regulator. Regulator is a circuit with feedback, this one has no feedback.

Second, to understand your circuit you need the following basics:

  1. voltage and current definitions
  2. Ohm's law
  3. LED simplified model (this is for LED, and it's the only one worth talking here: suppose that while illuminating, it has certain forward voltage, like 1V to 4V, depending on color, and current that is proportional to light intensity).
  4. Kirchoff's law of currents.

just google the points 1, 2 and 4, it's too long to explain here.

Anyway, change 0R to minimal resistance you want to use- otherwise in certain position of the potentiometer the LEDs will burn, or the power supply will fail (it kind of has his own resistor inside).

Also check that you don't have too many LEDs in series, because if you do, they will not illuminate because they will not reach their forward voltage.

Last- if you are serious about engineering, take a course on edx or coursera.

  • \$\begingroup\$ A few points (1)"Regulator is a circuit with feedback" - a regulator circuit can also be open loop (no feedback) e.g see ko4bb.com/e102/e102-4.php .(2) The LEDs will not burn but the pot may well at low resistance if its current rating is exceeded.(3)The LEDs in a 5050 strip are series (3 LEDs + resistor)/parallel connected and rated at a particular voltage so no need to check you have too many LEDs.With increasing length of strip you increase the current required so what is needed is a circuit to control current rather than voltage e.g. a controlled current source or perhaps PWM \$\endgroup\$ – JIm Dearden Jul 10 '15 at 11:53
  • \$\begingroup\$ Well, if we are being so precise, PWM itself is a method that may not be compared to current control. \$\endgroup\$ – Gregory Kornblum Jul 10 '15 at 12:08
  • \$\begingroup\$ As an engineer I always strive for precision as should you. However, the point of the last sentance is "to control the current rather than voltage". This can be done in the two basic ways I have suggested - a series linear control element such as a BJT or MOSFET (an analog approach) (see Nick's answer above) OR by averaging the current through the LEDs using PWM (a digital approach). An example of such a circuit is given here circuitlab.com/circuit/c8m48y/pwm-led-dimmer-12v-8a-96w \$\endgroup\$ – JIm Dearden Jul 10 '15 at 12:31
  • \$\begingroup\$ You do understand, that even in this paragraph you could be much more precise, don't you? My point is that for each conversation there is appropriate degree of precision, same as for any application there is appropriate degree of complexity or robustness or whatever. \$\endgroup\$ – Gregory Kornblum Jul 10 '15 at 12:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.