Using 2 L293D to drive 2 high torque servos

Thread Starter

bobson_h

Joined May 5, 2010
3
Dear folks,
I try to drive 2 high torque servo motors with 2 L293D, but encountered several problems. I'm hoping someone here has clues for me. Thanks in advance.

I hacked the servo motors so that they are now 2 geared DC motors that operate 4.5v~6v. The 2 L293D were piggybacked to provide the current needed. I wired the servos directly to 4.5V (3 AA batteries), they raised the heavy arms attached to them. I wired the servos and L293D (piggybacked) and provide L293D power and input signals directly from 6V and ground with a SMPS wall wart, they raised the heavy arms. So I think that the power source and L293D output are able to handle my servos.

Besides L293D, there are also an Op-amp (ua741), a 4011, a 4049, and a couple of pull-up resisters in my circuit. Anyway, here are my problems:

1. I used the 6v wall wart for motor power and 4 AA batteries for logic circuits. The op-amp output signal is "windowed" (output high is about 4v, low is about 1.5v). I learned that a poor SMPS can have terrible effect on op-amps. What better solutions do we have to remove the noise of a poor SMPS. (I have tried to put a capacitor between 6v to ground, not helping)

2. Gave up the wall wart for the time being, I used 2 6V (4 AAs) separately for motor power and logic circuit. The circuit works fine when motors are not attched. (6V/1xx mV output for motors from L293D). But when motors were attached, they stalled. Apparently, L293D is not providing as much juice "in my circuit". (Probably not enough current, since I got 6V/1xxmV output.)

3. I also noticed that my logic circuit voltage is dropped to about 4V instead of 6V even when I pulled off the 6V motor power. Why is my 6V battery drop voltage to 4V in my circuit? Is it the same reason causing my motor power providing not enough current?
 

SgtWookie

Joined Jul 17, 2007
22,230
For starters, you really need to attach a schematic of your circuit. You mention a number of components, but a schematic tells the story in a much more complete manner.

You also need to provide more details about your servomotors. We currently have no way to know what their power requirement is. You mention "high torque", which implies that a good deal of current will be required.

"Stacking" the L293 may result in a smoking tower of ICs, if you have not provided sufficient heat sinking for them.

The 741 opamp is antique, and is best suited for display in an obsolete electronic components museum. It's cheap, but there are a huge number of modern opamps available with far better specifications.
 

Thread Starter

bobson_h

Joined May 5, 2010
3
My apologies, the circuit diagram is kind of messy and the servos are SCANNER SSV-9724 and SSV-9764MG. Their specs looks identical, operate at 4.5~6V, the running current is 410 mA, stall current is 970 mA.
About my circuit:
L293D Pin 11/14 are determined by comparing LM393 1IN+/1IN-
L293D Pin 3/6 are determined by SW5 on/off
Pin9 is disabled when the controlled motor is going one way and SW3 closed, or going the other way when SW4 is closed. Same story for Pin1 with SW1/SW2

I recalled that when I test drove 2 L293D directly with batteries, the motor power was from a wall-wart. (A cheap SMPS, Input AC100-240V 50/60Hz, Output DC+6V/2A)
However, the wall-wart seems to mess with my Op-Amp, so I did replace 741 with a LM393, the output low signal shows 2xx mV, but the high signal is at about 4V when it should be 6V. The corresponding L293D motor output also showed 4.xxV/1.xxV instead of 6V/1xx mV and stalled. Therefore, I replaced the wall-wart with another 6V battery source (4AAs). This time, the logic signals look good, but the servos stall. This morning, I tried to put 2 6V battery source (2 x 4 AAs) in parallel (Piggybacking the batteries, how sad.) for L293D motor power, and it worked. So I guess the 2 L293D cannot source enough current from one 4AAs.

Does anyone have suggestions to improve my circuit so that I can put the wall-wart with my LM393? The SMPS seems to produce a lot of noise in my logic circuit.
 

Thread Starter

bobson_h

Joined May 5, 2010
3
My bad, please ignore the unconnected wire on the left of 1OUT pin of LM393. And "LM293" is a typo, it's 393. Sorry!
 
Top