Basically this could be easily done by using a BJT NPN, but since I want to control two motors(3 V, 45-60mA) and 5 LEDs was thinking of using a Bi directional level shifter instead of using transistors for each application. Could this be possible ?


No. You should not use a "level shifter" as a motor driver "instead of using transistors for each application"

Although your motors are relatively low current as such go (though may well draw more in some situations), they still require at least three times more current than what relatively high-current "logic" applications would require - and at least an order of magnitude beyond what typical modern logic needs on a steady-state basis.

Motors are motors; logic is logic. For driving motors, you should be using at least light duty power electronics rated for the anticipated loads. There are definitely IC drivers in existence with sufficient capability - but they are marketed as that, not as logic level shifters.

Ultimately, it is possible that some particular level shifting designs using discrete components might be made of parts with sufficient rating for a light duty motor-drive application - but that would be something you would determine only after a qualified review of their specific data sheets, rather than something you can guess from the "level shifter" functional description of a circuit. Further, motor drives typically incorporate protection components, such as catch diodes and suppression capacitors.

  • 1
    \$\begingroup\$ Well according to my knowladge MOSFET are good in swtching circuits. For an example BS138. What I have is a Level shifter with 4 BS138 MOSFETS. So I wonder, given an arduino signal, why can't it to the same purpose. Ofcourse I may need to have include a capacitor and a diode for each motor get rid of noise and stuff. Apart from the fact use BJT why not FET? \$\endgroup\$ – OmegaD Jun 4 '15 at 5:40
  • \$\begingroup\$ I am in the same boat. I have some extra BSS138 level shifters and the motor controllers I do have won't go to al ow enough voltage for my 3v motor. \$\endgroup\$ – Matt Williamson Nov 29 '17 at 23:00
  • \$\begingroup\$ @MattWilliamson - you can generally supply a higher voltage to a motor drive bridge and apply a lower PWM duty cycle. The classic BSS138 level shifter circuit is utterly and entirely unsuited to motor drive as it has no current gain whatsoever. You could make a very weak low side switch with a BS138, but it would be an inverting switch, and not a "level shifter". And it would be a terrible choice for such a task, compared to NFET's actually suited to motor drive. \$\endgroup\$ – Chris Stratton Nov 30 '17 at 1:39
  • \$\begingroup\$ Thanks. I ended up just ordering some TB6612FNG H-Bridges. They work at 3.3v. \$\endgroup\$ – Matt Williamson Dec 8 '17 at 6:57

Yes, you can use a bidirectional level shifter if you want. However, unless you are receiving data back from the motor controller on the same data lines that feed the controller, there isn't much point.

Bidirectional level shifters are used when you need bidirectional communication between two or more devices that operate at different voltage levels.

It sounds as if your application is unidirectional: from microcontroller to motor controller.

For What It's Worth, one of the simplest non-inverting level shifters uses a single resistor and a single small N-channel MOSFET per line. I'll upload the schematic when I get in front of a computer.

  • \$\begingroup\$ Thank you so much for your promptly reply. I got it, as you said it is ttly uni directional, I had this bi directional level shifter in hand. Anyway I am looking forward to your schematics at your earliest convenience. Cheers . \$\endgroup\$ – OmegaD Jun 4 '15 at 3:44
  • \$\begingroup\$ I am waiting for your diagram :) \$\endgroup\$ – OmegaD Jun 10 '15 at 5:04
  • \$\begingroup\$ No. This is entirely erroneous - at best, Dwayne misunderstood the question - there is no motor controller, rather the question naively proposes using a level shifter as a motor driver. For the popular level shifter topologies that is not workable, and even in push-pull topologies that could work if made with power FETs, the results when constructed of logic FETs would be dismal. \$\endgroup\$ – Chris Stratton Nov 30 '17 at 1:42

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.