6 replies to this topic

[Matt Brailsford]

Hi Guys, I was wondering if you could help explain what I did wrong in the following scenario? I'm a bit of a noob, so apologies if it sounds like a dumb question. So basically, I had my netduino connected to an Adafruits motor shield, which in turn was connected to 2 DC motors (2 x Tamiya FA-130 Motors). I then created a simple program to alternate the direction of the two motors continuously using this Mshield driver http://forums.netdui...-shield-driver. To make sure I had enough power, I connected a 10v - 1.2A external power supply to the netduino jack, and attached the power jumper on the motor shield. When I ran the program, the motors turned (though not as I expected) and gradually slowed down, and finally resulted in the L293D chip overheating and getting fried. I swapped the chip for the other chip on the board, and used a lower voltage, battery power supply which then worked (to some degree, it kept cutting out, which I'm pretty sure was due to not having enough power as I was only connected to 3 AA batteries). Anyways, I'm pretty sure the chip fried because I was applying either too much voltage or too many amps, but I'm not which and for what reason. Could someone explain what it is I did wrong, and what calculates I should be making to choose a suitable power supply? Many thanks Matt

[427v8]

from http://www.ladyada.n...shield/use.html Current requirements: The second thing to figure out is how much current your motor will need. The motor driver chips that come with the kit are designed to provide up to 600 mA per motor, with 1.2A peak current. Note that once you head towards 1A you'll probably want to put a heatsink on the motor driver, otherwise you will get thermal failure, possibly burning out the chip. Stall current on the motors can be about 2 amps, http://www.pololu.co...alog/product/77 , so if your driving the motor one direction and them slam it to the other dir, guess what...Stall! in fact, not only Stall, but I'm thinking since the motor is turning the wrong way it will suck even more current. Bottom line? you smoked the driver because you drew too much current...just as you expected Keith

[Matt Brailsford]

Hi Keith, Thanks for your reply. So it was more about the ampage the motor was drawing exceeded the supported peak ampage of the chip that fried it. I now see the error of my ways =) One further question then. Moving forward with my project, what would be the best option to get this to work? as I guess the motors are always going to draw that ampage at peak. Would it still peak out if I gradually lowered the motor speed and then gradually increased it in the other direction? Or would I just be best doing the thing of piggybacking 2 L293D chips to increase the peak on the motor shield to 2.4a? Or is there something else I can do to protect the chip? Many thanks Matt

[Magpie]

Hi Keith I think it would pay to put an inline fuse in series with your power supply, maybe rated at less than half the current rating of the chip. And buy a few fuses. Better still use a current limiting supply while developing your project. Fuses will save you a lot of time and money on any project that involves more than a smidgen of power. Also it means that you don't have to size your power supply by the load. So you can use as a large power supply and not worry. ps. My last fuseless disaster blew $120 worth in leds in one brief burst of light.

[427v8]

Your drivers should be capable of supplying more than 2x the peak current. So like a 5 amp capability for a 2 amp motor. I'm not a hardware designer tho, so there may be better ways... Like adding a small series resistor or they do make PTC resistors that are low resistance untill too much current flows, then they go un in resistance to protect the circuit Using fuses while highly important, are only to protect for failure, not a weak design. the power supply should be enough to supply anything you need, current limiting should be done elsewhere! Current limiting and protection are different

[Magpie]

Yes in any good design it would be prudent to have your driver chip capable of sourcing as much current as your motor can pull. This is so you can drive your motor to its full power. But if you only have 1 amp drivers and a 2 amp motor then you can still safely run your circuit, if you use a fuse to protect it. Then just make sure that you don't drive your motor too hard, especially dont suddenly reverse the H bridge or else you will keep blowing fuses. I wouldn't use series resistors except as a last resort, they waste power and create heat. PTC resistors are not very good as resetable fuses because they are not very accurate and if they run hot for long periods, which they will if they are near their high resistance point, then they will eventually fail anyway, (I seem to remember them failing open like a fuse). The best way to protect a motor circuit is a current sensing ciruit read by your micro. Then back this up with fuse or circuit breaker protection. And when developing or breadboarding a power circuit, or power control software, nothing is more convenient than a current limiting power supply.

[Matt Brailsford]

Thanks guys. You've given me some great feedback, and I'm now confident I should be able to move forward with the project safely. Many thanks Matt