Maker Pro
Maker Pro

High side MOSFET not working @ IR2104 driver

Hi everyone,

I am trying to design variable synchronous buck regulator. I will change the buck output voltage by changing the duty cycle of PWM signal. For triggering, I am using Half bridge driver IR2104. There is no output signal in 'High side out' of the driver IC. But low side out is working properly.

Frequency of driving signal: 25 KHZ
minimum and maximum duty cycles of PWM driving signal : 10% AND 90%
VCC voltage = 12 Volt (Fed to IR2104 and drain of top MOSFET)
Diode used for Boot strap circuit: 1N4148
capacitor between VB and VS = 0.1 uf (Bootstrap diode)
capacitor between VCC and GND = 0.1uf
Resistor at the output of H0 and LO = 10kohm (I am NOT sure about which value to use )

NOTE: Supply voltage on the drain of the top MOSFET is fed from the same main 12V supply used to power the driver IC IR2104. Is there any problem on using same supply or it is fine to use??

PS: Please find the attachment. It contains the circuit diagram.
 

Attachments

  • IR2104_CIRCUIT.png
    IR2104_CIRCUIT.png
    21.7 KB · Views: 1,921

CDRIVE

Hauling 10' pipe on a Trek Shift3
Hi everyone,

I am trying to design variable synchronous buck regulator. I will change the buck output voltage by changing the duty cycle of PWM signal. For triggering, I am using Half bridge driver IR2104. There is no output signal in 'High side out' of the driver IC. But low side out is working properly.

NOTE: Supply voltage on the drain of the top MOSFET is fed from the same main 12V supply used to power the driver IC IR2104. Is there any problem on using same supply or it is fine to use??
Yes! It's not floating if the logic & HV supply are common to each other.

Chris
 
10k is much too high. It should be in the range 10Ω-100Ω. Depending on the FETs you are using, a 10k gate resistor may prevent the gate voltage rising high enough to turn the FET fully on in the time that HO is high.
There is no output signal in 'High side out' of the driver IC.
If HO is not pulsing but LO is, that suggests the IC is faulty.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
Damn it! I knew someone would catch that before I posted it! I can't type as fast as Hop. :)

Chris
 
Thank you everyone for pointing out the mistakes and leading me to the right path.

Yes! It's not floating if the logic & HV supply are common to each other.
I am not able to understand this point. Can you please explain.

Do i have to use higher voltage at the TOP MOSFET DRAIN in comparison to VCC supply of driver IC OR it can be lower and equal also and does the ground of both the supply has to be common?? Really i have no idea on this and thus it's a bouncer for me.o_O



Since resistor at the output should be between 10 ohm to 100 ohm, can anyone explain about the exact calculation for finding out the value of resistor if possible OR we just have to use approx value for application like these. I am using "N-channel MOSFET Z44" at the high and low side.

Surely, i will make the change by using resistor in range between 10 ohm to 100 ohm and using separate supply for TOP side MOSFET and VCC Supply of driver IC.
 
Last edited:
Thank you everyone for pointing out the mistakes and leading me to the right path.


I am not able to understand this point. Can you please explain.
Do i have to use higher voltage at the TOP MOSFET DRAIN in comparison to VCC supply of driver IC OR it can be lower and equal also.

Since resistor at the output should be between 10 ohm to 100 ohm, can anyone explain about the exact calculation for finding out the value of resistor if possible OR we just have to use approx value for application like these. I am using "N-channel MOSFET Z44" at the high and low side.

Surely, i will make the change by using resistor in range between 10 ohm to 100 ohm and using separate supply for TOP side MOSFET and VCC Supply of driver IC.
Did you even read Tahmids blog I posted?
M.
 
Since resistor at the output should be between 10 ohm to 100 ohm, can anyone explain about the exact calculation for finding out the value of resistor if possible
The value is not critical. The purpose of the resistor is to damp any oscillation which might occur as the result of the combination of gate lead inductance and gate capacitance.
 
Did you even read Tahmids blog I posted?
M.

Thank you for the link. I got new and interesting things to learn.

Yes i have gone through it but he is focusing mainly on IR2110. It has two supply VDD as logic supply and VCC as low side supply. Based on his explanation i came to know that there is a direct relation between VDD voltage we select and logic output voltage of microcontroller. For dspic or msp like micro, he suggested to use 5 Volt as VDD supply.

He also suggested to select VCC between 10 to 20 volt (as per datasheet) BUT for the driver IC IR2104 logic supply and low side supply is common (there is NO VDD and VCC separate but single supply as VCC). So now what voltage i should select.

If i select VCC as 12 volt then will IR2104 will be able to drive the load as i am generating PWM drive signal from 3.6 volt logic microcontroller.

PS: i am using MSP430 microcontroller. Its logic 1 corresponds to 3.6 volt.

Thank you
 
Most of the features of these series are common, just that 2110 the VCC and VDD are separate but can be coupled if needed to 10-20v.
I usually feed both with 15vdc and use a high speed opto on the input from a μp or Picmicro.
I would use 12vdc minimum for VCC.
Just my preference, I have used the principles shown in the blog and they work fine.
COM and VSS are the same in the iR2110 also.
M.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
Most of the features of these series are common, just that 2110 the VCC and VDD are separate but can be coupled if needed to 10-20v.
I usually feed both with 15vdc and use a high speed opto on the input from a μp or Picmicro.
I would use 12vdc minimum for VCC.
Just my preference, I have used the principles shown in the blog and they work fine.
COM and VSS are the same in the iR2110 also.
M.
Minder, you are too kind. I made a statement earlier that the supplies should not be common to each other but I was obviously in error. You didn't call me on it though. Don't be shy in doing so. I do make mistakes that seem to be increasing with age.

Right now I have Hurricane Mathew heading my way and I'm nervous as a hatter. It isn't helping the process of clear thinking. I'm a well seasoned Hurricane veteran but they still scare the sh!t out of me! It's not something that comes easier with experience. Just the opposite. Being nailed with the eye (makes a believer of you) of Wilma (the nasty bitch) in 2005 has left an ever lasting impression on me .... TERRIFIED! :eek: It's an enemy that I can't strike back at.

Amit, sorry for the brief derailment of your topic but I think those 10K resistors was your problem. I would guess that if you had an appreciable load on those FETs while using those 10K resistors on the Gate it would have fried the FETs from turning on so slowly.

Chris
 
Thank you for the reply everyone.

I usually feed both with 15vdc and use a high speed opto on the input from a μp or Picmicro.
I would use 12vdc minimum for VCC.

I found in Tahmid's blog :
It is common practice to use VDD = +5V. When VDD = +5V, the logic 1 input threshold is slightly higher than 3V. Thus when VDD = +5V, the IR2110 can be used to drive loads when input “1” is higher than 3 point something volts. This means that it can be used for almost all circuits, since most circuits tend to have around 5V outputs. When you’re using microcontrollers the output voltage will be higher than 4V (when the microcontroller has VDD = +5V, which is quite common).

According to graph using 12 volt dc as VDD logic 1 input threshold is around 8 volt. Does this mean, i will require a controller whose logic 1 corresponds to 8 volt min. but I am using MSP430 (Logic 1 corresponds to 3.6 volt only). In that case MSP430 will not able to drive IR2104.
Did i misunderstood something?? or my question to this point is valid. Please correct me if i am wrong.

I think those 10K resistors was your problem. I would guess that if you had an appreciable load on those FETs while using those 10K resistors on the Gate it would have fried the FETs from turning on so slowly.

Chris, thank you for your suggestion. i am working on it.
I am sorry if i am asking silly question but i want to understand the concept.!!
Can i use same VCC supply (+12 volt) to power both IR2104 Driver and Top MOSFET gate with COM supply of IR2104 Driver and source of down MOSFET as a common ground.???

waiting for response.
 
ccording to graph using 12 volt dc as VDD logic 1 input threshold is around 8 volt. Does this mean, i will require a controller whose logic 1 corresponds to 8 volt min. but I am using MSP430 (Logic 1 corresponds to 3.6 volt only).
The IR2104 needs a Vcc voltage of at least 10V (so your 12V is fine) and is compatible with 3.3V logic (so your MSP430 can drive it directly on the IN pin).
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
Thank you for the reply everyone.



Chris, thank you for your suggestion. i am working on it.
I am sorry if i am asking silly question but i want to understand the concept.!!
Can i use same VCC supply (+12 volt) to power both IR2104 Driver and Top MOSFET gate with COM supply of IR2104 Driver and source of down MOSFET as a common ground.???

waiting for response.

If I understood Minder's reply (below) then I was wrong and the answer is yes.
Most of the features of these series are common, just that 2110 the VCC and VDD are separate but can be coupled if needed to 10-20v.
I usually feed both with 15vdc and use a high speed opto on the input from a μp or Picmicro.
I would use 12vdc minimum for VCC.
Just my preference, I have used the principles shown in the blog and they work fine.
COM and VSS are the same in the iR2110 also.
M.

Chris
 
Please find the attachment. It contains the waveform and measurement of Low side MOSFET gate.

I got PWM output at both LO and HO side of the MOSFET but frequency at the LO is unstable. It is switching between 35 KHz to 50 KHz. (As per oscilloscope output at the right bottom corner) but measure switch of the oscilloscope shows 25 khz. While frequency at the HO is stable 25 Khz . I am generating a PWM signal of 25 Khz from Micro.
Can anyone tell about this misbehavior??

Right now i am NOT driving any load. All the testing is going on without load.
I replaced resistor value at HO and LO to 10 ohm and used same supply to power both driver and TOP MOSFET'S DRAIN
.
Bootstrap cap used is 100 nf. i got same result even with 2.2 uf tantalum cap. Does this problem is anywhere related to bootstrap capacitor or its the oscilloscope issue. (Oscilloscope is brand new "OWON")


Thank you.
 

Attachments

  • 20161006_002701.jpg
    20161006_002701.jpg
    247.1 KB · Views: 226
Last edited:
Top