Buck Boost Driving with IR2110 ?

[geobabu]

Hi,
I was just trying to implement a Buck-Boost converter with a driver method as shown in the image. Since the IR2110 have individual High side and low side driving, i actually
thought this will help to avoid the using discrete components. But now I came to realize that I was wrong !
The issues I was facing ,

1) Boost operation
For boost operation (ie, switching the LOW side MOSFET Q4 ), the HIGH side MOSFET Q3, normally should be ON for provide the supply . This is not happening because the
Q3 MOSFET is not coming to ON state . I believe its because of the boost strap capacitor is not getting charged , which is usually happening through the LOW side MOSFET's ON
state if its in a HALF bridge circuit .

2) Buck Operation
The Buck circuit normally formed by switching HIGH side Q3 mosfet and keeping the LOW SIDE MOSFET Q4 always in OFF state. I believed atleast this part will work, but its
seen that, till 30% Duty Cycle, the buck part is responding as mentioned above , but after that , the PIC controller is getting heated up and all voltages are diminishing. What
would be happening here ?

The i/p voltage V_PS is just 8-18V & Vout is between 11.5 to 13-5V, and later planning to expand.


Thank You,
regards,
Geo

 

[johansen]

this is a classic problem of nothing to pull the source of the buck mosfet to ground, so there's no path for the boot strap capacitor to get charged.

one solution is to use a synchronous buck architecture.

if the pic is "getting warm" and it works but only at limited duty cycle ranges.. then you have other problems.. can you post a photo of the circuit construction?

 

[ronsimpson]

Look at the data sheet but I think the IR2110 does not work if the supply voltage is below 10 volts.

Change R8 to 100k. The top diver supplies 12mA for that resistor. It is discharging C7,8 too fast. The driver uses 150uA so you are discharging C7,8 100 times faster than if you had no resistor.

Q3 can not be left on 100% of the time. It needs to be off part of the time to charge C7,8. I think with R8=100k you can go from 0% to 90% of the time. With R8=1k you probably can't go much past 50%.

I don't know what mode you are using. Some people use boost OR buck. You can use buck/boost (buck AND boost), where both MOSFETs are turned on part of the time. There are several options but don't run the top MOSFET on for more than 95% of the time or C7,8 will not charge.

 

[geobabu]

Thank you for your replies. The controller used was PIC18F2341 (with 6channels of 12-bit PWM). In this particular model , the PWM pins PWM0-PWM1 , PWM2-PWM3 , PWM4-PWM5 form 3 pairs , which can be configured as half bride type. All these 3 pair sets was working from a single period timer and each pair have its own duty cycle generator. Each pair can be configured as either in Complementary mode (with dead time delay) or Independent (with no delay and exact same pulse outs in each pin having equal duty cycle). For the Buck-boost mode a pulse required as shown in the attached image.

So I have done some modifications like, Controller is changed to PIC18F24K22, and PWM pins from the CCP and ECCPs are arranged in such a way to meet these conditions.
Also as you mentioned , I understood that IR2110 is not good for these kind of use. So I removed IR2110 & done according to the image.


Is there any particular High side driver is available , which can replace this transistor based circuit.

Regards,
Geo

 

[geobabu]

Also In the Circuit, The Gate to Source Resistor , is increased to 10K. If we use 100K , is it sufficient to suppress the accidental switching ON of the MOSFET due to the gate capacitance in cases where the driver is presented there. ? Diodes D9, D8 are kept as optional for later advancements.
Regards,
Geo

 

[ronsimpson]

The circuit you made is the same as the IR2110. You have not changed anything.
It is good to change the gate resistor. When operating the gate is turned off by a transistor not the resistor.

 

[ronsimpson]

What drive wave forms can you make?

 

[geobabu]

yes functionally its same as the IR2110, but in this case the problems as mentioned in the above post are not there !! . May be that's why its telling this IC is not good for below 50V operation.
Also, this IC is bought from the market here, I doubt whether its a genuine item !!
Regards,
Geo