First ROBOT with L293D problem

[ericgibbs]

Hi Eric,
I should have taken your advice seriously anyway, The ckt seems safe so far.I will try those 1N4001, and still wonder that do I actually need to change the code you suggest while the ckt works fine digitally and even with motor (I have written the program 5 months ago and almost forget the logic or at least most of it and now I just have 6 days to complete the project so don't want to mess with the software part ), the only problem is of hanging when motors are rotating fast and any of the LDR detects light. If I slow down the motors then they work fine I mean completely fine but slow motors can not carry the load of robot.

hi,
As you are on a tight schedule, what input do you want from me.?

 

[pkshima]

Hi Chris,
I did advise against connection a PC's psu.!

Hi Eric, I thought you were concerned about the safety of the computer. Since Chris has a seperate psu, shouldnt it be safe to use it ? Chris doesnt probably have any other means to provide enough power to the robot.

Chris, if I were you, I would fix the robot in 2 parts ...

1. The PIC part and its SW.
2. The L293 part.

I would start with building a simple L293 circuit on a breadboard and assure myself that it works the way I think it does.
Then I would proceed to the PIC part.

If you are in a team, you can ask someone else to look into one part.

Also, you might want to start a new thread altogether in the microcontrollers section about the PIC part. The microcontroller section on this forum is much more active than the robotics one, so things should move faster there.

You might also want to take a day off school/college and work full time on this bot to make max use of 6 days.

 

[c36041254]

O.K. first Eric you have helped me so much and I really appreciate that as of now I want to try the diode ckt and will see if it works. As you pointed out the code problem I think I should forget that as the ckt works fine save for the back emf problem, but let me know what do you think of this and what would be the easiest and simplest correction that can work with my existing hardware if that correction is really necessary.



And to you Pradeep, the steps you have suggested about the project are really good for the beginning of the project and earlier when I made this robo I made it on the bread board and I solder it down only after confirming that the ckt works fine. The only problem is the back emf and of course the power supply in a financial view, but I would like to stay away from the ATX PC supply as I found over some sites that it needs some modifications to be used as lab supply.
I really thank you too for giving your time. I would like to know that if my L293D is burnt by that short ckt or whatever that was , Can that other IC you have suggested fit exactly in the same ckt , I mean can I replace it for my L293D ?

Hope I'll be able to finish the project in time.

Thank you all !

 

[pkshima]

Chris, the other IC BA6209/BA6222 will not fit exactly in your circuit.
But it will save you a lot of headache. Following are some of the benefits

1. Its a simple 10 pin IC with pins in a single row. So fits on simple generic PCBs.
2. Also I never had to use any diodes with it so the layout and fabrication is easy.
3. Gives 1.6 or 2.2 A current based on which IC you choose.
4. Cheaper. BA6209 (Rs 30) and BA6222 (Rs 50)
5. Easily available. Tell the shopkeeper that it is used in VCRs.
6. The datasheet contains an application circuit for tape deck motor.
7. You can stack multiple together to increase power.
8. The motor voltage can be set much different than the supply voltage.

Read the datasheet for BA6209 and see if you like it.

BTW, 6 days is hell lot of time !!!

 

[c36041254]

Hi There,
I tried the ckt with 1A adapter and both motor runs very fast , the ckt was still hanging so I pu 47 ohm resistors in series of each motors and WOW ! the robot runs fine but either 1A or 500mA both makes 7805 and L293D slightly hot, I probably can touch 7805 heat sink but only for 10 or so seconds after that I really feel the heat ( this is for 500mA so imagine for 1A ) while L293D doesn't get that much hot but still it seems to me that I have a disaster in waiting, or is this normal ? Will a heat sink do ? What with L293D ? I can't attach a heat sink to that.

 

[ericgibbs]

Hi There,
I tried the ckt with 1A adapter and both motor runs very fast , the ckt was still hanging so I pu 47 ohm resistors in series of each motors and WOW ! the robot runs fine but either 1A or 500mA both makes 7805 and L293D slightly hot, I probably can touch 7805 heat sink but only for 10 or so seconds after that I really feel the heat ( this is for 500mA so imagine for 1A ) while L293D doesn't get that much hot but still it seems to me that I have a disaster in waiting, or is this normal ? Will a heat sink do ? What with L293D ? I can't attach a heat sink to that.

hi,
Fit a good sized heatsink to the 7805 and refer to this image for the L293.

 

[c36041254]

Thanks ! Eric for the idea, now my next post will take some time and hope that will be the final one as I have to fix ckt and motor and all that stuff in to a ROBOT form so see you then ,
Thank you all !

 

[c36041254]

Hi there,

It's working very nice and quite speedy, just some times ambient light disturbs it but, drawing curtains solved that. I need to clear some points that I think I understood .


1 : The large capacitor across +5v and GND of 7805 makes it work as a current source as voltage can not change instantaneously across a capacitor and thus offering a huge parallel resistance.

Quastion :
If my understanding is right than why it needs to be large capacitance ? , I mean the resistance offered by capacitor is inversely proportional to the capacitance so 1 pico will offer more resistance than 1000u .

2: Can this ckt works if I don't want to use MCU ? In attachement I have just shown the ckt for single motor. The IC is L293D and consider it as properly supplied.