Difference l293D and SN754410NE?

I found this circuit (on the picture) on the Arduino site. Could i replace the SN754410NE H-bridge to a L293D? If no, why not?


Hi bestanamnetnogonsin,
just DON’T the SN754410NE is the better device! I usually do it the other way around and replace the L293! But now I tend to use drivers that use FET outputs, as transistor outputs like on the L293 lose about 2V and more as higher currents are drawn.

But yes they are pin compatable, if you really want to do it!!

You should try something like the DRV8833 or the DRV8835 both have FET outputs and lose little power… Pololu produce a number of small SMT modules that work very well.

Hope it helps.



PS. Here’s a Pololu DRV883 module on one of my bots, green PCB lower left side…

Well its just that it seems to be so much more expensive...

Here it is cheaper http://www.technobotsonline.com/semiconductors/integrated-circuits/linear-and-drivers/h-bridge-and-motor-drivers.html The L293 still relies on it's old legacy as being very good (at the time).

Description: Faster, cheaper, smaller, better, right? The SN754410 Quad Half H-Bridge is just that. Capable of driving high voltage motors using TTL 5V logic levels, the SN754410 can drive 4.5V up to 36V at 1A continuous output current! Please see datasheet for more information. This is a pin to pin compatible replacement for the L293D.



@Cactusface, what would you recommend for higher currents, say 3A?

Hi Chilli,
To be honest I've never built anything that used more then 1.5A. But the VNH5019 from ST and available on a Pololu module will work at 12A and about double that peak. Also a few driver shields work at higher currents.



That looks like quite a beast!

Incidentally my reason for asking was that I was thinking of making some motor control shields, and didn't want to copy a design based on a chip no one likes!

I do a small one that can handle 800mA a channel which is quite impressive given the size of the actual chips on the board.

L293D is good for only 600mA per channel, SN754410 is good for 1A per channel. Both are bipolar devices, and you'll lose about 1V to 1.5V of your battery volts in the chip.

the VNH5019 is not a proper stepper driver. the OP has a sketch for a stepper connected motor. If that is required, this device may not be the best choice.

the L298 is a proper stepper driver, with a 4 amp max rating. I think 4 amp max means destruction at 4.001 amps. and I would never suggest getting it near max without proper protection.

however, I would not recommend it for a new design board, there are better ones that use FET technology and have lower consumption of power.

however, I would not recommend it for a new design board, there are better ones that use FET technology and have lower consumption of power.

What would you recommend for a newer design of board?

You want to look at a chopper driver. They act like a constant current switch mode regulator, using the inductance of the motor windings rather than an external inductor.

You can drive a stepper motor faster with more torque by doing this. You would also then use a power supply voltage much higher than what the stepper is rated at, and adjust the driver for the motor’s rated current.

In addition, those driver chips also generally are driven by sending a direction and step pulse, rather than driving each winding separately. This reduces both the number of IO pins used, and the amount of time the program must spend driving the motors.

The TB6560 is rated for up to 2.5A and is easily connected to a heat sink.
The A4988 can drive up to 2A if you properly solder the bottom pad and add a heat sink and forced air cooling.

These can also do microstepping, another advantage of using a chopper based driver chip.

With good heatsink, the L298 is good for about 1.5A continuous per chan. A liitle more for short periods. It's also a bipolar device, so will lose 1-1.5V in the chip.