ir2110 bootstrap capacitor calculation

[abc_de]

Hello
I am trying to build 3 phase inverter using pic18f4431 mcu, pwm frequency =16khz
Power circuit consist ir2110 mosfet driver, irf 840 as switching device.
I am facing problem while calculate bootstrap capacitor value.
VDD=5VOLTS, VCC=12VOLT, DC BUS VOLTAGE=342
Please any one suggest me what should be right value for bootstrap capacitor. I shall be very thankful.

 

[red_alert]

You could find all the answers here:

http://www.fairchildsemi.com/application-notes/AN/AN-6076.pdf

http://www.irf.com/technical-info/appnotes/an-978.pdf

http://www.mitsubishielectric.com/semiconductors/files/manuals/dipipm_bootstrap_circuit_e.pdf

 

[abc_de]

thanks for your reply i will study and re post if i got sucess

 

[clara]

Chuck in the biggest capacitor you can find. When power is disconnected, the voltage will fall at a speed inverse to the size of the capacitor.

 

[red_alert]

Chuck in the biggest capacitor you can find. When power is disconnected, the voltage will fall at a speed inverse to the size of the capacitor.

The bootstrap capacitor has nothing to do with power disconnection but the switching frequency, duty cycle, MOSFET gate charge (turn-on gate input current), rise/fall time and so on.

Larger is better anyway but if it doesn't have enough time to charge (due to small duty cycle of the low side MOSFET) it's oversize is useless.

It's much better to use a moderate size (0.1 - 10uF) but quality (pulse type) capacitor to support short duration impulses. Also, a fast switching diode (for bootstrap capacitor charging) it's a must.

 

[abc_de]

thanks for reply
just for testing purpose i am using full bridge configuration. and fpwm is just 50hz square wave.
but my circuit does not repose. i have attach image of my circuit please help to solve the issue
i am working over it from last 1 week :fight::fight:


https://obrazki.elektroda.pl/6087508600_1423395669.png

- - - Updated - - -

Circuit diagram is shown on above given url

- - - Updated - - -

 

[red_alert]

The circuit it's ok (including bootstrap capacitors). You mean it doesn't work at all? Are the HIN/LIN signals ok? Do you have any output from IR2110?

You could also put some diodes across R1 and R2 (cathode toward IR2110) for faster gate turn-off (though that's not important for the 50Hz testing PWM ).

 

[abc_de]

yes the circuit does give any out put
hin/lin voltage is about 1.2v. i have replace ir2110 many times with new one. still i don't know whats wrong in this circuit

 

[red_alert]

HIN/LIN voltages are way to low (if 1.2V is supposed to be the "1" logic level).

IR2110 is 3.3V logic compatible but you have to use a VDD of 3.3V, too (in that schematic, VDD is 5V).

Anyway, 1.2V is bellow logic "1" threshold even for 3.3V logic. That's it, IR2110 only see "0" logic at its inputs. You need at least 2.5V for the "1" logic state.

 

[abc_de]

Logic supply vdd Is +5v , same as microcontroller power supply then 1.2 volt would be considered as logic 0 right???

 

[red_alert]

How did you measured those 1.2V? Using a regular multimeter? If so, and you've measured the 50Hz square wave signal, there's no indication about high/low (1/0) logic levels (you've just read a rms value).

Anyway, if you're using a uC (and i suppose you don't have an oscilloscope available) you could further lower the PWM frequency (1 - 10Hz) to actually see if the IR2110 is working (check its inputs/outputs).

 

[abc_de]

I found HO is 8volt & LO is 2.5volt with respect to common (vss).
I thing ir2110 is not working properly. :x

 

[red_alert]

I presume you're still measuring the RMS value (50Hz).

Please try to apply a steady signal (5V) at the LIN input then measure the LO output (you should read about 12V). Do the same for LIN = 0V (attempt to measure around 0V). This way you could actually verify the low-side driver (at least a static determination).

For the high side driver, you should connect the VS terminal to ground then do the same test for HIN/HO. Obviously, you should disconnect the MOSFETs during this tests.

 

[abc_de]

Thank you for your valuable reply. Today I will check the circuit again and repost the result

 

[abc_de]

hello
i have done all observations with regular multimeter. while i was testing i found driver circuit working fine. then i replace bootstrap diode 1n5819 with uf4007 (ultra fast diode).
then circuit give output.
as following vdc=342 volt then i observe 165 around unstable(swing from 130-165 approximate) across h bridge. then check my circuit on Proteus. i have attached the file of Proteus.
i don't have an oscilloscope right now i am planning to buy.
please tell me i am going right. :roll::roll:

URL

http://obrazki.elektroda.pl/8686270900_1423502707.png

- - - Updated - - -

http://obrazki.elektroda.pl/8686270900_1423502707.png

 

[red_alert]

What kind of signal do you apply at IR2110 inputs when you got that fluctuate voltage? That 50Hz square wave (test) signal or are you using a sine PWM signal (50Hz/16kHz)?

Do you have any capacitors across DC BUS?

 

[abc_de]

i am just applying simple square wave also using decoupling capacitors .1uf/630volt across dc bus

 

[red_alert]

For a 50Hz PWM signal, you need more than a simple .1uF decoupling capacitor across DC bus. Anyway, it depends on load, too. Is it resistive or inductive? What's the load rated power?

 

[abc_de]

Till I am testing circuit without load. I have drive 1/2 HP motor.
In circuit after rectification I have use 470uf/200v capacitors in series so that 470uf become 400v same kind I have use 4 capacitor in 2 pairs total 4 capacitor then decoupling capacitors.
I found dc bus voltage is remain constant probably while I switch the mosfets at 50 hz
...
Can you share any another algorithm for switch of h bridge so that I can generate 1 phase and run 1 phase motor

 

[red_alert]

You could find details about sPWM algorithms here:

https://www.freescale.com/webapp/sps/site/overview.jsp?code=784_LPBB_HCS08TPM

To generate a 50Hz sine wave, I recommend you the unipolar sPWM technique: