0

I designed an H bridge using the DGD0506A gate drivers. https://www.diodes.com/assets/Datasheets/DGD0506A.pdf

I used the FDMC7696 for both high side and low side mosfets. I went with the suggested circuit as provided in the gate driver datasheet enter image description here

I just used a transistor to invert the IN signal to the second chip, so the opposite mosfets turn on (e.g. HO on the U1 and LO on the U2) enter image description here

and this is the arrangement of H bridge itself enter image description here

after ordering the parts and the PCB, I started testing it, but I had no luck and it wouldn't work. I checked everything and found out I'm missing a connection here: enter image description here

I just botched a few wires to make that connection but it still wouldn't work. enter image description here

I have tried it with a DC motor and a 5k resistor connected to the load, no chance with either of them. I checked a few volrages and the Vs is sitting at around 9.5v with respect to ground (drops quickly to 7.8 volts when measured by a multimeter or oscilloscope) but the voltage across both tantalum capacitors is 0. I've created this exact same circuit 3 times and I'm running out of them but I've had no luck finding what the problem is. (both EN1 and DIR1 are 3.3v signals,DIR1 is digital High/Low, but EN1 is a PWM @25KHz. VCC is 12v and VDD is not being utilized here but it would be 3.3)

I was wondering if you guys can help me troubleshoot this issue and perhaps get this board working. I have enough parts to make 1 new board but that's pretty much it.Thank you very much.

OM222O
  • 282
  • 2
  • 11
  • You probably need to post the completed schematic, mark the test points and voltages. You've added jumpers to C1 and C2 but they don't appear to be the 4.7 µF electrolytics on your first diagrams and we have no idea where they go. Can you see the problems for your readers? – Transistor Sep 05 '18 at 21:10
  • they are not electrolytics, they are tantalum capacitors which still do have a polarity. these are all parts of a huge schematic, so I will try to remove the other parts and post it later. – OM222O Sep 05 '18 at 21:14

1 Answers1

1

Unfortunately it appears that you are using static control voltages for signalling to the driver chip and H bridge and this means that despite you fixing the connection, the bootstrap circuit (that is needed to properly drive the upper MOSFETs) will not operate. The data sheet describes it as: -

The DGD0506 A is a high frequency half bridge gate driver capable of driving N channel MOSFETs in a half bridge configuration. The floating high-side driver is rated up to 50V.

Note the words "high frequency" and this means that you control the inputs with a switching waveform such as one derived from a PWM source. The bootstrap circuits rely on the output switching continually so that it can internally generate a power rail higher than Vcc in order to turn-on the upper N channel MOSFETs.

Using the EN input for PWM will not work either - you need to use the IN input: -

enter image description here

As you can see, all that EN does is enable both MOSFET drives to be controlled by IN or turn both MOSFET drives off.

Andy aka
  • 434,556
  • 28
  • 351
  • 777
  • As I mentioned above, I'm using a 25KHz PWM "EN1 is a PWM @25KHz".I'm also not using 100% Duty cycle, I actually ran the tests at about 50% cycle time. Do I need to use PWM on the IN (Dir) line as well? – OM222O Sep 06 '18 at 12:35
  • You don't use a switching PWM signal for EN because, if you look at page 2 of the data sheet it tells you that "Logic Input Enable, a Logic Low turns off Gate Driver" and this means both MOSFET drives are turned off. Also look at figure 2 on page 6. – Andy aka Sep 06 '18 at 15:27
  • You have to use a PWM on EN because it otherwise it's constantly on, which means there is no speed control for the motor which would be useless! I am using the IN to switch the direction of motor spinning, and EN to control it's speed. I'm guessing I should extend the EN duty cycle but turn on the bottom mosfet for about 10% of the period? – OM222O Sep 06 '18 at 16:23
  • A low on EN turns both drives off and that means no control because both mosfets will be off simultaneously and that won’t allow the bootstrap circuit to work properly. – Andy aka Sep 06 '18 at 16:46
  • well how can you adjust the speed of the motor then? I don't need the mosfets to be on 100% of the time. that's basically the whole idea behind using a gate driver, otherwise you can just hook the bootsrap circuit to the micro controller and have it switch everything. – OM222O Sep 06 '18 at 17:06
  • You feed pwm to IN and with a 50% duty cycle the motor does not rotate. Higher duty means rotation in one direction and lower duty means rotation in the other direction. That’s the basic setup. – Andy aka Sep 06 '18 at 18:07
  • I will give that a try and get back to you later. thanks – OM222O Sep 06 '18 at 18:09
  • Or you can control one pair of MOSFETs (half bridge) in a way depicted in this question: https://electronics.stackexchange.com/questions/130069/do-ac-inverter-h-bridges-ever-get-driven-this-way (2nd picture in question). – Andy aka Sep 06 '18 at 18:15