Author Topic: L293D Overheating  (Read 8155 times)

0 Members and 1 Guest are viewing this topic.

Offline Abdulla M.A.Topic starter

  • Jr. Member
  • **
  • Posts: 30
  • Helpful? 0
    • My personal website
L293D Overheating
« on: October 30, 2010, 06:02:45 AM »
Hi guys,
for a circuit not exactly as the one shown below, I have L293 gone to be overheating
why?? the motors I used is shown in the link.
I used PWM of frequency=4KHZ (i.e. PR2=294 , crystal=4MHZ, Pre-scale).
Battery:9v, 1500mA.
VS of L293 is connected to the battery acros 3.5 Ohm/1w resistor.

http://www.sparkfun.com/commerce/product_i...products_id=319



thankx, Abdulla

Note: there is a mistake in the schematic, the connections of 1Y, 2Y, 3A, 4A. fixed
"A scientist can discover a new star, but he cannot make one. He would have to ask an engineer to do that."
"For an optimist the glass is half full, for a pessimist it's half empty, and for an engin

Offline blackbeard

  • Supreme Robot
  • *****
  • Posts: 575
  • Helpful? 4
Re: L293D Overheating
« Reply #1 on: October 30, 2010, 08:18:57 AM »
have you thought about putting a small resistor on the vs pin? something small like 100ohm
"sure, you can test your combat robot on kittens... But all your going to do is make kitten juice"

First step: Build androids with AI
Next step: Give them vaginas

Offline Abdulla M.A.Topic starter

  • Jr. Member
  • **
  • Posts: 30
  • Helpful? 0
    • My personal website
Re: L293D Overheating
« Reply #2 on: October 30, 2010, 09:22:47 AM »
100 ohm !!! how did you calculate it??

Abdulla
"A scientist can discover a new star, but he cannot make one. He would have to ask an engineer to do that."
"For an optimist the glass is half full, for a pessimist it's half empty, and for an engin

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: L293D Overheating
« Reply #3 on: October 31, 2010, 05:25:40 PM »
for a circuit not exactly as the one shown below,
...We can only provide a solution not exactly correct/working!
If you don't care enough about your problem to provide the actual schematic, why should others care to spend time on a solution?


I have L293 gone to be overheating
Overheating would be how many °C?
Did you make the PCB with a copper area for cooling through the ground pins?


why??
Provide us with an exact schematic and put some numbers on specs, or we'd have to be guessing.


the motors I used is shown in the link.
Sure, a picture tells a thousand words, but I'd settle for engineering specs like voltage range, peak efficiency current and locked rotor current.
However, SparkFun writes that it works with the L293D, which is weaker than the L293.
If you use the L293 (the one without a "D"-postfix), did you remember adding protection diodes?


I used PWM of frequency=4KHZ.
That should not cause any problems.


Battery:9v, 1500mA.
For what? (There's a 5V line and a 12V line).


VS of L293 is connected to the battery acros 3.5 Ohm/1w resistor.
Why the resistor?

Just got the quite scary thought that you connected Vs to +9V through a 3.5 Ohm resistor, but remembered that you are an engineer and such foolish things would an engineer never do, as it's a sure way of snuffing the chip, so what purpose did you have in mind for the resistor?


Note: there is a mistake in the schematic, the connections of 1Y, 2Y, 3A, 4A. fixed
Another good reason for posting the exact schematic.
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline slurp

  • Beginner
  • *
  • Posts: 5
  • Helpful? 0
    • MiniSumo.Org.UK
Re: L293D Overheating
« Reply #4 on: November 01, 2010, 06:26:01 AM »
the motors I used is shown in the link.
Sure, a picture tells a thousand words, but I'd settle for engineering specs like voltage range, peak efficiency current and locked rotor current.
However, SparkFun writes that it works with the L293D, which is weaker than the L293.
If you use the L293 (the one without a "D"-postfix), did you remember adding protection diodes?

I'd second the comment of L293D vs L293 - check the data sheets on current rating. In my experiance the starting and stall current out strips the L293D rating (or is at best to close for comfort).

It's also worth checking your build of the gear box, tight fits and poor running is possible - make sure you apply the grease supplied.

Best regards,
Colin

Offline Abdulla M.A.Topic starter

  • Jr. Member
  • **
  • Posts: 30
  • Helpful? 0
    • My personal website
Re: L293D Overheating
« Reply #5 on: November 01, 2010, 08:09:44 AM »
Thank you guys,

@Soeren
Here in Baghdad we have a wise said "Deal with people in your ethics, not in their ethics"
your are really strange person!!!!! anyway, I solved the problem.

BTW, in the university you just deal with theortical problems, and this is known in all the world universities, so you need
alot of time to get practical experiance, I think you are older than me in at least 10 years, that's mean you took more
time in electronics, so when I will became in your age, I will be better than you ;D

Now, I will answer your questions:
1- No Comments (search for answer above).
2- Very hot, sorry but I do not have thermometer( :o) to measure the Temp!!!  I did not make PCB, it cost alot, about 40$
 single layer, it's too much if we compare it with USA where it cost 9$
3-I used L293D, L293D do not need FWD
4-The resistor to step down the voltage because Pulse voltage must be less than 6v, so what is the problem!!! do you
 have any suggestions??

Abdulla
"A scientist can discover a new star, but he cannot make one. He would have to ask an engineer to do that."
"For an optimist the glass is half full, for a pessimist it's half empty, and for an engin

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: L293D Overheating
« Reply #6 on: November 03, 2010, 01:56:27 PM »
@Soeren
Here in Baghdad we have a wise said "Deal with people in your ethics, not in their ethics"
your are really strange person!!!!!
And that was a prime example of how your ethics work then, I suppose?


BTW, in the university you just deal with theortical problems, and this is known in all the world universities, so you need alot of time to get practical experiance,
You are not allowed to tinker in your spare time during Uni?
Well, extracting info from a datasheet does belong in the theoretical phase where you calculate what's possible in a given circuit.


I think you are older than me in at least 10 years, that's mean you took more
time in electronics, so when I will became in your age, I will be better than you ;D
An interesting thought... At least the one I had about, how you got from the thesis to the conclusion in that single sentence.

Well, anything is (theoretically, at least) possible, but you need to speed up (and perhaps reverse time) then, as I was designing (working) circuits when I was much younger than you are now, before I had any formal education in electronics (among lots of other stuff, I build my own Z80 based personal computer when I was a teenager).
You won't ever be as flexible as a Chinese artist either, even if you start now and work at it for the rest of your life - starting as a kid, when your body and brain has a lot of plasticity is the key to later mastery.

Sure I'm older than you (Isn't that something that usually calls for respect in Baghdad? ;)), my sons are older than you as well, but age has nothing to do with it, it's your way of assimilating, or rather rejecting, info. I have told you at several occasions that you should provide engineering data rather than fluffy descriptions and I would think that it came naturally for one signing himself as engineer (and sometimes teacher/educator), but you seem to have a blazing disregard for anything that doesn't become you and so you refuse to learn - not a good attitude for becoming a better engineer.


Now, I will answer your questions:
1- No Comments (search for answer above).
2- Very hot, sorry but I do not have thermometer( :o) to measure the Temp!!!  I did not make PCB, it cost alot, about 40$
 single layer, it's too much if we compare it with USA where it cost 9$
"Very hot" is still not usable. 10°C is "Very hot" in Antarktis, but very cold in Congo.
If you haven't got a thermometer of some sort, at least compare it to a known factor or make a rough estimate.
Like "burns a finger in 7 seconds", "goes *ftzzz* when I place a wet finger on it" or "glows cherry red". I'm sure you get the point (eh, no I'm not, just being polite here).

Study the data sheet. The ground pins need to go to a copper area for heat sinking! Such integrated circuits are not meant to be used without heat sinking, if you want to use the full power range.
It's not very expensive to make a PCB at home and, if all else fails, you can solder thin copper plate "wings" on the ground pins as a final measure.


3-I used L293D, L293D do not need FWD
Correct, but the data sheet says, under "Maximum Ratings", which is not to be taken as anything but "cross this boundary and we guarantee that things will go wrong"...
"Continuous output current, Io: L293D = ±600 mA". And that's when used correctly with heat sinking through the ground pins.
Sensible designers stay somewhat below that 600mA.
Since you haven't provided any data for the motors you use, I cannot say whether this chip is at all usable, but I'm sure the locked rotor current goes way beyond 600mA, so, all things considered, I'm not surprised it doesn't work.


anyway, I solved the problem.
AND
4-The resistor to step down the voltage because Pulse voltage must be less than 6v, so what is the problem!!! do you have any suggestions??
Well, either you solved it or not!
If you solved it, it would be polite to provide the solution for others that may run into a similar problem and if you didn't solve it (that's where my money is), don't claim you did.
Theoretical knowledge should be quite enough to tell you that trying to drop around 3V from a dynamic (="changing") source via a resistor to a dynamic load is asking for trouble.

Batteries are never nnV. They start at some voltage and drops while discharging.
The load will vary as well, according to how it's commanded.

A resistor of 3.5 Ohm (1W) with a drop of 3V needs a constant load current of 857mA and will then have to dissipate 2.57W. Change the current and you change the voltage (Gee this is theoretical knowledge that you should know).
Assuming the current falls to eg. 500mA, the drop will be 1.75V (and the 1W resistor will be able to handle the power of 875mW) and suppose it goes up to, say 1.2A, the drop would be 4.2V and the dissipated power goes up to a tad over 5W!


The solution:
Use a voltage regulator and don't forget some kind of heat sink.
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline Abdulla M.A.Topic starter

  • Jr. Member
  • **
  • Posts: 30
  • Helpful? 0
    • My personal website
Re: L293D Overheating
« Reply #7 on: November 04, 2010, 08:39:03 AM »
Quote
You are not allowed to tinker in your spare time during Uni?
Well, extracting info from a datasheet does belong in the theoretical phase where you calculate what's possible in a given circuit.

Spare time!!!! really you do not know what's going on!!! you do not know the Prof. here, anyway I will give you an idea
about the life and the study here:
you have to wake up at 6 am, at 8:30 am the first lecture start, if you reach at 8:35 am, you will not enter, the
lectures end at 2:30 pm, you will be at home 4 pm, small rest and lunch, the O'clock become 5 pm. so you have just
7 hours to study, and this is not too much to get success, many of the students were falled, as I remember we were 138
student
in the first year, in the graduation, we were just 40. yo do not know how much the difficulty of the studying, you have
to learn everything by yourself.

Abdulla
"A scientist can discover a new star, but he cannot make one. He would have to ask an engineer to do that."
"For an optimist the glass is half full, for a pessimist it's half empty, and for an engin

Offline Abdulla M.A.Topic starter

  • Jr. Member
  • **
  • Posts: 30
  • Helpful? 0
    • My personal website
Re: L293D Overheating
« Reply #8 on: November 04, 2010, 08:49:55 AM »
Quote
An interesting thought... At least the one I had about, how you got from the thesis to the conclusion in that single sentence.

Well, anything is (theoretically, at least) possible, but you need to speed up (and perhaps reverse time) then, as I was designing (working) circuits when I was much younger than you are now, before I had any formal education in electronics (among lots of other stuff, I build my own Z80 based personal computer when I was a teenager).
You won't ever be as flexible as a Chinese artist either, even if you start now and work at it for the rest of your life - starting as a kid, when your body and brain has a lot of plasticity is the key to later mastery.

Sure I'm older than you (Isn't that something that usually calls for respect in Baghdad? ), my sons are older than you as well, but age has nothing to do with it, it's your way of assimilating, or rather rejecting, info. I have told you at several occasions that you should provide engineering data rather than fluffy descriptions and I would think that it came naturally for one signing himself as engineer (and sometimes teacher/educator), .
Hmm, you built a computer!!!! let me see it, if what you said is correct!!! I do not think it's possible to build a personal
computer by yourself specially you do not have that enough info. about computer architectures, memories, .... in that
age!! be more logical please.


"A scientist can discover a new star, but he cannot make one. He would have to ask an engineer to do that."
"For an optimist the glass is half full, for a pessimist it's half empty, and for an engin

Offline Abdulla M.A.Topic starter

  • Jr. Member
  • **
  • Posts: 30
  • Helpful? 0
    • My personal website
Re: L293D Overheating
« Reply #9 on: November 04, 2010, 08:58:55 AM »
I changed L293D with L298.
voltage regulator!!!! is that your solution!!! do you think, I did not think in that!!!! the motors take to much current
at stand still (2100 mA), 7805, LM317, not suitable in this application, right??
D.C Motor No.= FA-130, just google it.

Abdulla
"A scientist can discover a new star, but he cannot make one. He would have to ask an engineer to do that."
"For an optimist the glass is half full, for a pessimist it's half empty, and for an engin

Offline WhomBom

  • Full Member
  • ***
  • Posts: 64
  • Helpful? 0
Re: L293D Overheating
« Reply #10 on: November 04, 2010, 09:51:41 AM »
you could buffer the lm317 output with some capacitors i guess to deal with the start up current spike, but ofcourse you probably already thought of that too...

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: L293D Overheating
« Reply #11 on: November 06, 2010, 12:09:56 AM »
Spare time!!!! really you do not know what's going on!!! you do not know the Prof. here, anyway I will give you an idea about the life and the study here:
The "Prof."?


you have to wake up at 6 am, at 8:30 am the first lecture start, if you reach at 8:35 am, you will not enter, the lectures end at 2:30 pm, you will be at home 4 pm,
If it takes you 90 minutes to get home from school and 90 minutes to get there from you wake up. Does that imply that you sleep fully dressed and take off the minute the alarm clock goes off (no breakfast, no bath, no nothing)?

Well, a 6 hour day at uni doesn't sound hard - you cannot get away with such a short day in DK, except if you skip lectures of course, which most students do now and then, since most can handle that and read up on the relevant subjects at a later time. Most students here have to work at least a part time job to get them through financially, a lot of them spend hours weekly in cafés (coffee shops that sell anything but plain coffee) and goes partying at weekends (hence the need for a job ;D) - even so, some of them (a lot of them actually) graduate and moves directly to engineering jobs.
So, how do they pass? Not because of low standards, I can assure you. Danish engineers are popular targets for head hunters around the World, for exactly that reason.
Cultural differences may have a part to play, but I think the major reason for all this is that Danish engineers are quite geeky and a good lot of us started playing with electronics at an early age.
I find it strange though, that the internet haven't been a better equalizer in that respect - I would have loved the massive info base it is, when I was a kid and probably was something of a trial to the librarians in our public libraries, where I went very frequently to scour their technical sections and to have them get books from other libraries for me.
Nowadays you can get whatever info you need by just using a search engine - no heavy bags of books to drag along and a virtual endless supply of info - only drawback is, that you still have to think, to be able to filter out wrong info. Well, a small price to pay IMO.


[...] small rest and lunch,
Tinkering IS resting for the dedicated, so you have time each day.


[...] the O'clock become 5 pm. so you have just
7 hours to study, and this is not too much to get success,
Fact: The brighter the mind, the less studying is needed to achieve the same goal.
Following that train of thoughts...


many of the students were falled, as I remember we were 138
student in the first year, in the graduation, we were just 40.
Yes, lots of people flunk (probably in any country), but perhaps they should have raised the Q of that filter a bit - a freshly graduated engineer should still be able to do engineering work and not expect the internet to save their sorry behinds.


yo do not know how much the difficulty of the studying, you have to learn everything by yourself.
Then what's the point in going to the lectures?

If you claim they taught you nothing and that you spend 3 hours daily just to get there, it means you skipped the opportunity to study by yourself for a total of 9 hours each day (assuming you didn't study or ponder private projects while en route), so how smart do you think that was in retrospect?

Easy to complain about how hard you think that your life is, but very much on a need to know basis - and we don't need to know. Life is hard and then you die, nothing's fair etc. - deal with it.


Hmm, you built a computer!!!! let me see it, if what you said is correct!!! I do not think it's possible to build a personal
computer by yourself specially you do not have that enough info. about computer architectures, memories, .... in that
age!! be more logical please.
So, now you're calling me a liar?
Just because you weren't able to, nobody else could be?
You are so much in denial of the fact that even if your LED is a bit dimmed, it doesn't mean that the rest of the world is on the same level - you act like a blind man screaming liar, at a seeing person that explains what vision is.

Boy am I glad that I didn't know then, that you would think it impossible, or it would have shattered my confidence in a way that I could have made errors or had problems getting it to work ;)

Somehow, I don't think you should tell others to be "more logical".


I changed L293D with L298.
And you think that this will somehow magically transform your 9V supply voltage to 6V (which is still too much)?

What exactly did they teach you at your selfproclaimed "theoretical education"? (Not about electron flow, Ohms and Watts Laws and similar "trivialities" it seems).

And why don't you pick up on what people tells you?


voltage regulator!!!!
That's what I wrote.
(Here's an article that you might wanna read).


is that your solution!!!
Yeah, and a very easy one.


do you think, I did not think in that!!!!
I've stopped guessing about what you think, but since you didn't use one, very simple logic deduction tells me that you cannot have given it that much thought.


the motors take to much current
at stand still (2100 mA), 7805, LM317, not suitable in this application, right??
Ah, engineering data... Perhaps you are able to learn a bit after all.

Did I mention anything about LM317's or 7805's?
I just wrote "voltage regulator" and if you had given your project a bit of your own footwork and actually Googled it, you'd have known by now, that voltage regulators comes in a multitude of current handling capabilities - like this:

Which would handle both motors without even working up a sweat.

You'd be wise to make a circuit to cut (or seriously decrease) power in the event of a locked rotor though, or you could melt down the motor windings in a short time - like eg. when running into something that won't budge while the obstacle sensor(s) failed. This regulator have an enable input that just lends itself to an over-current shutdown


D.C Motor No.= FA-130, just google it.
Eh?
You're the one with the problem, so you do the Googling and provide the data and links as relevant - no nursing- or spoon feeding service provided. This is not Uni, this is real life.

Are you trying to learn, or are you just trying to find someone telling you that your ideas will work?
You disregard the info you get, if it doesn't fit into how you think it should be (or would like it to be), but your lack of success getting things working might be a pointer, that you need to step back a few paces and re-evaluate a bit.

And before you spend your entire quota of exclamation points on a new post only displaying substance-less attitude and spitting on people, consider this...
Whom of us will be sorry if your project never gets finished and working correctly?
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: L293D Overheating
« Reply #12 on: November 06, 2010, 12:15:11 AM »
[...] but ofcourse you probably already thought of that too...
Oh sweet irony :D
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

 


Get Your Ad Here