5.1 V (1.0 A) bipolar stepper + L293B = HOT!

I have a bipolar stepper that I got from an old scanner. From its tag it is a 5.1V, 1.0 A, 1.8degrees stepper. I built the circuit based on the schematics found here:

I'm using an old Nokia cell phone charger to feed the stepper. It is a 5.7V 800mA. Every thing is working really well, except from the fact that the IC (L293B) is getting very hot! Even after just about 30 secs of the motor spinning.
So I was wondering:

  1. Is it normal for this IC to get uncomfortably to touch hot?
  2. Should I use a cooler?
  3. Could I be doing something wrong?

One thing that I noticed is that before issuing the "step( )" function, the stepper spindle is "loose" as if there was no power going to the stepper, but after issuing "step(n)" the stepper moves "n" number of steps, stops and the spindle gets "hard" (as expected) but the IC starts to get very hot.

Thank you for any help,

PS: L293B datasheet in case it helps: ST Microelectronics 1328 datasheet pdf

Is it heat-sinked?

Is it heat-sinked?

No... It is in a protoboard, although I'm using a socket.
How should I heat sink it? (newbie here :slight_smile:
Kind regards,

the 4 ground pins in the middle of the chip are being used as 'heatsink' There is a example on page 9 on how to extend the surface (capacity). The example is with the copper plating on a circuit board but you could make something similar with a small/thin sheet of copper/steel. Just be careful not to short-circuit something.


There is a example on page 9 on how to extend the surface (capacity).

Hi Yot. Thanks for the fast reply. I just got a bit lost here... page 9 where? Here at the forum? :-/

Sorry for being to criptic. Page 9 in the datasheet you linked.


edit: I was talking about figure 12. Figure 13 is not so simple to make compared to figure 12.

Generally don't want to use high amps or voltages with a breadboard, can have some negative results! :slight_smile: Generally, anything over an amp I make my own proto-board to test with.

And I didn't read the datasheet.. but I believe the chip handles 1 amp max? That'll make the chip hotter than your little fingers can handle! You can piggy back the chips to handle more amps. (piggy back meaning.. just that, put the chip on top of the other chip, pins need to be the same direction, solder the chips together)

But yeah, I'd look into some proto-board for testing motors and such, don't need your breadboard melting on you. :slight_smile:

the reason the motor is loose before you initially issue a step command is because at that point the stepper coils haven't been engergised, so the L293d is ENabled via the enable pins and ready for you to step, once you step the first time the appropriate coils become energised and stay energised until the state of the EN 1,2/3,4 pins change or the motor is 'stepped' again, I'm guessing you've got yours tied to VCC? which is why it all gets hot straight away and stays that way. I believe you should be able to use a couple of pins on the arduino to toggle the state of the EN pins so that once you have finished stepping you turn off the EN pins which will keep the heat down. The downside of that is that once the pins are disabled the motor can no longer hold whatever it is turning, so gravity etc. will come into play. you might want to adjust your strategies to cope with the different scenarios. like:

enable, step, disable (single step mode, not worried about 'holding' feature of motor)
enable, step0, step1, stepN, disable (Multiple steps, using holding/detent torque)

Or leave it as is and have all of the features the stepper gives :slight_smile:

WOW!! Thanks a lot pwillard, Reggie, CaptainObvious and Yot!
I learned a bunch from this thread.
I will heat sink it, put it on an appropriate circuit board and see how it goes.
But maybe I will need a better solution mechanically speaking.
I'm building a tiny elevator. Actually a drawer that goes up and down from the ceiling, so I would need the stepper to stay on at all times, which would be very inappropriate on a long term basis, otherwise the entire thing would come down due to gravity.
Any suggestions?
Thank you once again,

A real elevator that is driven by a chain uses chain locks to hold it in place. Perhaps you could use some sort of ratcheting device that would allow the chain to slip past going up, but not slip down. Use one on each side, and a couple of servos to flip them into or out of place, to change direction.

Counterweight the elevator so it's close to equilibrium when lightly loaded. Then put a solenoid-driven friction brake somewhere on it that you have to energize to release in order to move the elevator. That way if anything fails, the brake clamps down by default and stops the elevator from moving.