L298 not working

Discussion in 'General Electronics Chat' started by miniElectronics, May 4, 2010.

  1. miniElectronics

    Thread Starter Member

    Aug 4, 2009
    20
    0
    Hi,

    I need a little help to get the L298 to work. Please take a close look at Project2.png. Where i am trying to connect a motor. The motor have to work forward and reverse using a microcontroller. All design before L298 is working. But when signal is sent to L298 so it doesn't work according to datasheet.

    I have connected 4 blocking diodes to get the motor to work. But i think there is a trouble or short circuit maybe.


    Many thanks
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You do not have anything connected to Sense A, so there is no path to ground for the motor. It won't work like that.

    Usually, you would use a current sense resistor from Sense A to ground, somewhere in the range of 0.1 Ohm to 0.5 Ohms - if you want to monitor the motor current. If you do not plan on monitoring the motor current, then connect it to ground.
     
  3. kkazem

    Active Member

    Jul 23, 2009
    160
    26
    Hi,
    I noticed another problem with your design. You are using 1N914 diodes as catch diodes around the L298 power devices. The 1N914 is a small-signal diode and not rated for the higher currents that it needs to handle. You need an ultrafast diode. You can probably get by with the On-Semiconductor's MUR410, which is rated at 4 amps and 100 volts and a reverse recovery of about 25 nanoSeconds, or the MUR415 (150V) or MUR420 (200V). I'm giving you these multiple part numbers as sometimes distributors are out of one but have lots of the others. These are about $ 0.50 to $ 0.75 each or so. The 1N914's will either short out, or do nothing. And the layout is critical. These diodes should be as close to the L298 pins they attach to as possible. Also, your 0.1uF cap should be very close to the supply and ground pins of the L298, with the cap leads less then 1/4" long. Finally, I don't know why you're using a L298, which is a dual power H-bridge when you only have a single load. You're wasting half of the IC. You can probably put the two H-bridges in parallel and drive them with the same signal, which will keep the stress on each H-bridge to 1/2 the level that you are now stressing the single H-bridge. But do consult the datasheet and any app notes from STS before doing so.
    Regards,
    Kamran Kazem
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Good catch. I probably would've caught the 1N914's if I'd looked at the 2nd image; too hard to read on the first one.

    FR303 diodes could be used as well. Digikey has the MUR410, MUR420, FR303's all in stock.

    Don't try using 1N400x or 1N540x series diodes; they are too slow.
     
  5. miniElectronics

    Thread Starter Member

    Aug 4, 2009
    20
    0
    Hi,

    I would like to say thank you to all for helping. Specially for diodes & SensA. Actually i need a motor of 2.5A to run on H bridge. So therefor i choosed this L298. And want to connect two channels in parallel.

    Posting here working part of L298 which i have corrected now. Just have another question is that in manual for L298 it is written a non-inductive capacitor have to be connected to ground and a normal. Is there difference between them?
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You have connected Sense A directly to ground. This means that you have no way to determine the current flow through the bridge and motor. An 0.5 Ohm sense resistor is suggested; when voltage across the sense resistor reaches 1v, current through the bridge and motor has reached 2A, and the power needs to be cut for a brief period to ensure that the current does not exceed an average of 2A. You could use a comparator in your uC to measure the voltage on Rsense; when the comparator trips, have an interrupt routine either use the inhibit or control lines to turn off the bridge for a period of time.

    The L297 companion IC has a built-in chopper driver that will take care of this for you, if you don't wish to perform the chopper driving via software.

    The L298 is rated for 2A continuous current per channel maximum. You'll have to connect both sides in parallel to get up to 4A current.

    Posting here working part of L298 which i have corrected now. Just have another question is that in manual for L298 it is written a non-inductive capacitor have to be connected to ground and a normal. Is there difference between them?[/QUOTE]

    Aluminum electrolytic capacitors have a good deal of parasitic inductance and resistance. They will get hot if you are charging/discharging them at a fast rate, and may even explode on you.

    You could use low-ESR capacitors, or you could use a number of metal poly film or ceramic caps.
     
  7. miniElectronics

    Thread Starter Member

    Aug 4, 2009
    20
    0
    Hi,

    The motor i am going to use is of 2.5A. Now i have connected the L298 in parallel so currents can be upto 4A. I don't see that is there any point to use Capacitors as mentioned in the manual of L298?

    I have test the simulation and it is working good without connecting the capacitors which are mentioned that must be connected between supply voltage and ground. The motor i am going to use will not work more than 6 minutes constantly and maximum current of motor is 2.5A.

    You can see the attached on pin4 and pin9 it is connected a 100nF capacitor. But i have not understood that it is something must be connected or it will work without that?

    Thanks
     
    Last edited: May 6, 2010
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    So, why don't you test it for battery life WITH and WITHOUT the capacitors, running at 50% PWM.

    Report back as to the useful run-times you get in each case.

    You must have a 0.1uF cap connected from L298 pin 9 to ground.

    I strongly suggest that you follow my advice and include the other capacitors as mentioned.

    You will get much better performance and longer battery life.
     
Loading...