LTspice simulation of H-bridge using IR2110

Discussion in 'Homework Help' started by afwef1, Apr 1, 2016.

  1. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    I need to simulate H-bridge using ir2110 and irf540 mosfets. At some point in simulation, high output of ir2110 stops working, and I don't know why, hope someone here could help me out.

    IR2110sim.as is the point where it works. I am using ir2110 typical connection as guide how to connect it. Anyone knows what is the problem?

    I am using LTspice IV to simulate it.
     
  2. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
    Could you upload the components as IR2110 and HCPL-2200?
     
  3. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    Yeah sure, just to confirm ( I am new to LTspice), u want .sub files right?
     
  4. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    I apologise for not being more helpful, but I have made both componets from these files
     
  5. eetech00

    Active Member

    Jun 8, 2013
    652
    112
    try these parts and symbols: IR2110 and HCPL-2200
     
    ScottWang likes this.
  6. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
    The IR2110 still didn't match with your attached file -- IR2110simNOTworking.asc.
    I tried your IR2110 and from eetech00, I also downloaded from the library site, but they are the same, they all too big compared to your asc file.

    What is your original screen shot?

    (I was used hcpl-2200 from eetech00 provided), irf540 was downloaded from here.

    IR2110simNOTworking_afwef1_ScottWang.png
     
  7. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,813
    1,105
    I don't see any connection from the top FET source to the VS pin? Without that the bootstrap won't work.
     
  8. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    [​IMG]


    This is SC of IR not working. I am trying to find out when does HO stop pulsing, and also why. I Know it is not the best option to connect half of circuit and then watch what happens , but I have no other idea what to do. I have included eagle schematic which is my final goal how simulation should look like. I have tried to connect it fully ( same way that it looks in eagle schematic ) and I still had problem with HO on both sides in simulation. This schematic provided in eagle was made into PCB and it worked when I used oscilloscope to check it out.

    If you want I can connect full circuit and then send .asc file, if you think that would be helpful.

    Thank you all for the time you are taking to help me.
     
  9. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
    Do you have the lib of IR2110 components as the same with that screen shot, it is different from what you uploaded on #4.
     
  10. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,813
    1,105
    V2 and V3 should be out of phase, non-overlapping clocks.
    V2 and V3 should have series resistors for driving the optocoupler LEDs.
    The VS pin should go to U4 source, not to ground.
    Since the IR2110 has CMOS inputs, the optocoupler outputs will need pull-up (load) resistors (e.g. 2.7k).
    The lower end of C2 seems to be disconnected. It should go to VS.
     
  11. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    Scott this is the only one I have. I uploaded it from my C:/some-path-to-LTC/LTC/LTspice/lib/sub.

    I made it by opening .sub file in LTspice, finding where .SUBCTK is, right click on component name and then he gave me option to make element and did all work for me. Then I opened new schematic, and just clicked on add element and he was there.

    Does this make sense?

    Alec I will now try to do what u said and will come later to report what happend. Thanks for advice :)
     
  12. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    I have got to this.

    Ho is always Vcc, Lo is 0V.

    Typical connection of ir2110 provided in datasheet says that VS and Dradin of U3 go to load. R5 is "simulating" load.

    I don't Know what is the problem now, why Ho isn't pulsing, but I am guessing it has something to do with VS. Volatage on VS is around 14V and according to datasheet max is 0.3V

    I didn't put pull up resistor because im not sure how to connect it.

    Li and Hi are pulsing as they should.

    I have no idea what to do anymore.
    [​IMG]
     
  13. eetech00

    Active Member

    Jun 8, 2013
    652
    112
    I don't think the spice model from IRC is any good.
     
  14. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    Actually that would be great, because I could ditch this ridiculous simulation under the reason that model doesn't work ( I have to do it it for university). That would be happy end for me.

    I will consult my mentor in few days (Tuesday) about this and will post replay. If anyone has any idea until that I will gladly hear them out.

    Anyway, great forum, will definitely stay here.
     
  15. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,813
    1,105
    I can't get the 'official' model to work either. Attached is my own model for a IR2110. I had to rename the .sub file to get it to upload here, so rename it back to IR2110g.sub.

    Edit: Model updated.
     
    Last edited: Aug 20, 2016
  16. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    Better, Lo is now pulsing, but Ho is still always Vcc.

    I have connected it like this.

    Hope I didn't make connection misteake , im doing this very late at night.

    Thank you all for effort.

    [​IMG]
     
  17. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,813
    1,105
    I note you're not using the symbol file I posted, so perhaps the spice pin order for your symbol doesn't match mine?
    Edit:
    You still don't have pull-up resistors (say 3k3) from the Hin and Lin inputs to Vdd.
    You have U4 pin 2 shorted to ground.
     
    Last edited: Apr 3, 2016
  18. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,813
    1,105
    If you're still having problems, here's a test .asc I've used successfully with my IR2110 model. It works as is, but you could try adding opto-isolators on the front end to check if they're the culprit.
     
  19. afwef1

    Thread Starter New Member

    Apr 1, 2016
    15
    0
    OK I think we have finally got it working.

    I have now used your symbol, and puted pull up resistors.

    I have also made some changes in my schematic according to your IR2110g-Test file.

    [​IMG]

    It's a bit messy, sorry about that, I was eager to test it out.

    Ho and Lo are pulsing nicely, and both sides of R3 resistors have pulse signal , not perfect pulse but it should be ok.

    I have never seen anywhere that VB should be connected to M1 source, not even in ir2110 datasheet typical connection , however when I did it worked a lot better. Why is it necessary ?

    Also huge thanks for your effort and making new component.
     
  20. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,813
    1,105
    Your circuit isn't right.
    It shouldn't be. VS should connect to M1 source. The bootstrap capacitor should be connected between VB and the M1 source.
    And I've just spotted an error in the test circuit I posted: I show VS connected to M2 drain instead of to M1 source. It happens to make little difference there, but apologies for that.
    N.b. in most circuits using an IR2110 the source of the top FET (M1) connects directly to the drain of the bottom FET (M2); the load goes from that connecting point to ground (or to another half-bridge).
     
Loading...