L293D trouble

[dhruvkaran]

hello all,
we are trying to make a line follower...
what we have done is that we generate 38KHz from the uC, connect sensor o/p to a not gate which is in turn connected to a L293D.
Now, we are facing two problems.
1) initially, the whole circuit works fine, but the problem is that, both motors(each connected to one channel of theL293D) work only one at a time, when the sensors are both set that both the motors should work, none of them work... although they do work individually.
2) after some time, the motors stop working totally. i.e the L293D o/p does change in accordance with the sensors but the motors dont work.
The motor supply to the L293D is coming from a 2A(max.) 15 V supply and that to the sensors and uC is coming from a wall adapter(500 mA max.) 5V(7805)
Please help us.
Thank you,
dhruv

[rao]

1.)Kindly check the pins you are using for your Vm i.e. motor voltage (pin 8) and Vcc i.e chip input voltage(pin 16)..Vm should be greater than or equal to Vcc.... Usually people screw up at that..

2.) I hope you know that the motor will turn only when both the inputs given for a single motor are not same.. i.e 10 or 01... you can ground one input permanently if you want the motor to move in a single direction...

3.) what is your motor voltage specification? what voltage are you supplying at the motor enable pins..

Dont think that I am underestimating your knowledge and these are too obvious to be asked.. but we tend to make mistakes in the most obvious and simple sections...

Please check your circuit again..

Rao

[dhruvkaran]

hey,
we've checked the circuit for these points.
still doesn't work.
the motors are 12V 500 mA, we've tried voltages from 12 to 15.....
still aint working
dhruv

why are you connecting ur sensor to L293d?
what does the uC do then, only generate 38khz freq?

[alimakkaya]

Although the chip can handle 1.2A of current at short bursts, it can not withstand the heat generated. You might have burnt the chip out. Just build a testing circuit to check. The chip acts wierd when it is burnt out. Some expect the chip totally stop functioning but no. I had a problem something like that when i was just trying to test PWM signal on a breadboard. Make sure you have soldered 5-6-12-13 pins to a copper surpace big enough to disipate the heat. Yeah it might not seem neccessary but it is vital to the chip. Try using L293 and externally add the diodes which in 293D are builtin. This will provide a higher stable and bursts current outputs.

[addy]

go through the forum for all sorts of details of l293d....
there has been too many discussions on tat before.
u ll get everything ranging from ckt diagram to its limittions....