+ Post New Thread
Results 1 to 9 of 9
  1. #1
    Member level 3
    Points: 3,330, Level: 13

    Join Date
    May 2004
    Location
    MISR "EGYPT"
    Posts
    56
    Helped
    2 / 2
    Points
    3,330
    Level
    13

    dc offset in MOSFET gate signal

    Hi all


    there is a nearly 4Vdc offset in the signal that drives the high side in my H-bridge
    and i don't know how to remove it .


    i didn't have this problem when i was driving the motor with a high side switch only
    i added a RC snubber circuit across the drain and source and every thing was working well i.e no offset ,sharp on/off , no distorion in either the gate or the source signal and the voltage diffrence between the gate and source are nearly above 12 v .

    but when i have entered the motor between high and low side switches , a dc offset appears on the gate of the driver of the high side , so the mosfet turn off partially .

    and the switching signal across the low side drain is distorted although the gate switching signal of the low side is nearly perfect .

    i think it is the miller effect that caused this shift in the high side source and distortion in the low side drain.

    but how to compansate miller effect ?

    salam
    hindi

  2. #2
    Advanced Member level 4
    Points: 41,560, Level: 49
    Fragrance's Avatar
    Join Date
    Jul 2002
    Location
    East Of Earth
    Posts
    1,177
    Helped
    212 / 212
    Points
    41,560
    Level
    49



  3. #3
    Full Member level 1
    Points: 1,922, Level: 10

    Join Date
    Jun 2004
    Posts
    111
    Helped
    1 / 1
    Points
    1,922
    Level
    10

    dc offset in MOSFET gate signal

    Look for common mode feedback.



  4. #4
    Member level 3
    Points: 3,330, Level: 13

    Join Date
    May 2004
    Location
    MISR "EGYPT"
    Posts
    56
    Helped
    2 / 2
    Points
    3,330
    Level
    13

    Re: dc offset in MOSFET gate signal

    Hi all

    sorry there is no simulation , because i didn't found the IR2110 driver .

    but , here is the schematic

    1- Fig 1 ,dc offset in the driving signal and hence in the source signal but when the collector of the PNP transistor is connected to the ground instead of the high side SOURCE, the dc offset had been removed Fig 2.


    2- this time the signal had been shifted to nearly 3VDC below the zero Fig3


    3-SCHOTTKY diode added the -VE part of the signal had been removed Fig4.


    4- but the signal at the drain of the low side MOSFET is completly distorted in both the amplitude nearly 5 V and shape although the gate signal is completly good and the Motor was not working .


    5- the motor starts to work after adding a RC snubber at the load and the low side , but the drain signal still distorted Fig5 and the high side Mosfet gets hot .


    6-when i have tried to add a snubber circuit to both the high and low side Mosfets and the load at the same time the motor doesn't works .


    7- so may be the problem is in choosing a suitable RC snubber circuit with an adequit values for the above combination [high Mosfet-LOAD-low Mosfet].

    So, how to adjust or tune the snubbers ?

    SALAM
    hINDI



  5. #5
    Advanced Member level 5
    Points: 37,801, Level: 47
    flatulent's Avatar
    Join Date
    Jul 2002
    Location
    Middle Earth
    Posts
    4,632
    Helped
    476 / 476
    Points
    37,801
    Level
    47

    Re: dc offset in MOSFET gate signal

    You should duplicate the suggested circuitry in the data sheet

    http://www.irf.com/product-info/data...ata/ir2110.pdf

    and remove the BJT additions. In your present design the BJTs are not biased properly and do not do any switching. They are just extra networks between the IR chip and the MOSFET.



  6. #6
    VVV
    VVV is offline
    Advanced Member level 5
    Points: 20,960, Level: 35
    VVV's Avatar
    Join Date
    Nov 2004
    Posts
    1,582
    Helped
    362 / 362
    Points
    20,960
    Level
    35

    Re: dc offset in MOSFET gate signal

    I do not think the problem you are describing is related to the addition of the snubber. I think it is simply a matter of not biasing/ using the circuit correctly.
    I assume the bipolar transistors were used just to simulate things, they are not part of the actual circuit. If they are, then remove them. The IR2110 is designed to drive the MOSFETs directly.

    I do not understand why you use two switches on a motor, if you are not using a bridge configuration.

    The high-side driver in the IR2110 uses a capacitor to act as a "flying" power source, connected between the source of the upper MOSFET and the Vb pin, which would be the collector of the NPN transistor in your drawings. This cap would allow the actual high-side driver to turn on the upper MOSFET.
    BUT, this capacitor needs to be charged up periodically. That happens when the top MOSFET is off and the bottom one is on. Then the cap charges up from a low voltage source (say 12V), through a diode. As you can see, there should be a time when the bottom MOSFET is on and the top one is off. I don't think this happens in your circuit. Thus, the capacitor (I don't see it in your schematic, but I assume it exists), does not have time to charge up and the high-side driver does not work properly (it should stop, since it has an undervoltage lockout).

    Now, when you only had the high-side drive, whenever the MOSFET was off, its source would go to GND potential, through the motor, and the cap could charge up correctly. I suspect with the addition of the bottom MOSFET that does not happen anymore, since I suspect you do not turn on JUST the bottom MOSFET for any period of time to allow the cap to charge up. You are forced to turn on both MOSFETs at the same time, which is not the way the IR2110 was intended to work. If it is used in a bridge, then the bottom and top MOSFET are not driven simultaneously (one bottom MOSFET is on and the top MOSFET in the second branch, controlled by THE OTHER IR2110 is on, then the roles reverse)

    In conclusion, since you are not using a bridge configuration, remove one of the switches. I do not see what advantage you get from using both MOSFETs.
    It makes a lot of sense to remove the high-side one and use the bottom MOSFET only, since it is easier to drive. (You may also find that the external diodes are not necessary, some MOSFET body diodes are rated for use in bridge motor drivers).

    Or, use a bridge configuration, which will allow you to reverse the sense of rotation, too. However, the caps must be charged periodically in this configuration too, so you still need to have the bottom MOSFETs on periodically (and when the motor is stopped, to make sure the caps are always charged up properly). This is easier, though, since the top and bottom MOSFETs controlled by the same IR2110 are NEVER on at the same time, unlike in your circuit.

    I hope this helps.


    1 members found this post helpful.

  7. #7
    Member level 3
    Points: 3,330, Level: 13

    Join Date
    May 2004
    Location
    MISR "EGYPT"
    Posts
    56
    Helped
    2 / 2
    Points
    3,330
    Level
    13

    Re: dc offset in MOSFET gate signal

    Hi all

    First , thanks allot for reply .

    the circuit is a complet H-bridge and the used coniguration was for simulation only , i did that for simplicity , and i think
    as VVV said that this was the problem , because it 'll distort the high side capacitor charging discharging cycle .
    So , thanks alot VVV for your help i think that this was the missed point .

    on the other hand i had used the 2 bipolar transistors as a totempole as a protection for for the IR2110 , please see .
    http://focus.ti.com/lit/ml/slup169/slup169.pdf

    from my work with this driver it is very sensitive specialy the high side , so i have used these BJTs as a isolation between the driver and the Mosfets and it wasn't working as a result of the abscence of the charging discharging path in the high side as VVV said before but when i had connected the emitter of pnp transistor to the ground instead of the high side Mosfet source it works .

    also , there was a missed point in the schematic which was that there was 40 ohm resistance betwee the driver and the driver and the npn collector transistor and a 1 k ohm between the driver and the 2 bases , this was for high side to make the totem pole baising floating with high side .

    but when i had connected the compelemtary signal of the high side to the low side , the 2 signals goes in phase although they where out of
    phase completly before applying them to the driver
    .

    so , what do u think about this problem ??


    Salam

    Hindi



  8. #8
    VVV
    VVV is offline
    Advanced Member level 5
    Points: 20,960, Level: 35
    VVV's Avatar
    Join Date
    Nov 2004
    Posts
    1,582
    Helped
    362 / 362
    Points
    20,960
    Level
    35

    Re: dc offset in MOSFET gate signal

    I cannot understand why the signals would go in-phase. That means the circuit is producing them is not working properly.
    Can you post the schematic?



  9. #9
    Newbie level 4
    Points: 1,136, Level: 7

    Join Date
    Sep 2006
    Posts
    6
    Helped
    0 / 0
    Points
    1,136
    Level
    7

    dc offset in MOSFET gate signal

    want schematic of mosfet driver for full bridge