use of IR2110 for three phase Full bridge IGBT driving

[KlausST]

Hi,

In post #26 FvM told you:

- missing bootstrap diode ...

It is still missing, so your high side will not work.

In post #29 you write:

e in any case its very important to turn on Fully the MOSFET/IGBTs.

To turn a FET/ IGBT fully on within a limited time you need some gate current.
The IR2110 can deliver 2A. But you use a 750 resistor. It limits the current to max. 0.025A. This is 1/80 of it's capabilities.
This slows down turn ON and turn OFF. This causes increased power dissipation in the FETs and maybe causes cross conduction, wich kills the FETs. --> try 10 Ohms.

"PWM_L" seems to be the PWM for the low side FET, but it is connected to the high side.
The same is with "PWM_H".

Klaus

 

[jean12]

Hello KlausST,I am still having the problem with this interfacing process,I provided the bootstrap diode and changed the resistors as you proposed in previosu post #41 but now the Driver are having problem and even the pic,the drivers are becoming too hot and end by burning.Other components are kept the same!!
I also provided between the Gate and Source of each MOSFET a resistor of 750Ohms

 

[FvM]

but now the Driver are having problem and even the pic, the drivers are becoming too hot and end by burning.

"and even the pic" means what, specifically? Your reports are known vague. I also noticed that you reported IR2110 not working at all in a parallel thread. https://www.edaboard.com/threads/349714/#post1493967

There are only a few possible cases where IR2110 can overheat or damage with output transistor still healthy. One is massive undershoot of the bridge output node connected to IR2110 Vs pin causing forward biasing of internal substrate diodes. Happens with bad power stage layout.

This and other unhealthy situations are usually diagnosed by detailed oscilloscope measurements.

I also provided between the Gate and Source of each MOSFET a resistor of 750Ohms

Expect problems for the high side bootstrap by the resistor DC current consumption. I know that the gate source resistors are persisting through many IR2110 hobby circuits, but they aren't founded by any IRF application note as far I'm aware of.

 

[jean12]

I wanted to mean that the pic also has damaged,you are meaning

use of Timer 1 for enhanced PIC PIC16F1936

with the configuration over there the system can`t run??
With the parallel thread I measure with oscilloscope I get the signals on HO,and LO of the driver but the power block (MOSFET) doesn`t go on.how may I connected those drivers output?Is my PIC configuration bad?because I set it in half bridge mode!!

Pls help!!

 

[FvM]

I believe that to control a three phase bridge with PIC16F1936, you have to run ECCP1 to 3 in half-bridge mode, each.

You should be able to verify correct PxA and PxB waveforms with an oscilloscope.

You can also check the IR2110 LO and HO signals without a bus voltage if you short the bridge outputs respectively IR2110 VS pins to ground so that the bootstrap circuits are charged. If the low side MOSFETs are in place and the control waveform is activating low and high side alternatively, you don't need a short.

After verifying the gate driver signals, the bridge output shorts can be removed and a bus voltage supplied from a lab supply, starting at low voltage and unloaded outputs.

 

[jean12]

Exactly that is what I did but not succeeding!!are the PXA and PXB be connected on one half bridge?

 

[FvM]

Not clear what the problem is.

Connection should be e.g. P1A and P1B to HIN and LIN of first half-bridge, P2A and P2B to second, P3A and P3B to third.

 

[jean12]

Yes,the P1A and P1B are on HIN and LIN respectively of the first half bridge and P2A,P2B on HIN and LIN of the second half bridge and then P3A and P3B on the HIN and LIN of the third half bridge.or are there any additional component(s) to add? unless the dead time is not well configured I set

PWMXCON=0x82;

Thx

 

[jean12]

Hello,I am designing a sine wave DC-AC converter but I am having trouble to interface THE IGBT SKM145GB066 to IR2110 which is having signal from PIC16F1936,Could you please guide me on how those different terminal from the IGBT are connected?

Thank you!!

 

[jean12]

Hi,

In post #26 FvM told you:

It is still missing, so your high side will not work.

In post #29 you write:

To turn a FET/ IGBT fully on within a limited time you need some gate current.
The IR2110 can deliver 2A. But you use a 750 resistor. It limits the current to max. 0.025A. This is 1/80 of it's capabilities.
This slows down turn ON and turn OFF. This causes increased power dissipation in the FETs and maybe causes cross conduction, wich kills the FETs. --> try 10 Ohms.

"PWM_L" seems to be the PWM for the low side FET, but it is connected to the high side.
The same is with "PWM_H".

Klaus

Hello,what do you mean by Gate current?The ULN.... like ULN2803?
I want to control an IGBT Module 145GB066D but it is not working very well because the High level at the output of IR2110 (I am using two IR2110 because I have two modules making a single phase full bridge) the voltage at HO is 6V and the LO is very low at approximately 600mV,could you advise how I can increase those levels all at HO and LO so that the module could work effectively?

Thank you,

 

[pic.programmer]

Please provide your latest circuit.

 

[KlausST]

Hi,

Please provide your latest circuit.

Yes. Urgent.

Hello,what do you mean by Gate current?

I think it's rather clear:

To turn a FET/ IGBT fully on within a limited time you need some gate current.

Klaus

- - - Updated - - -

Added:

I just scanned the thread on "ULN" and didn't find anything.

--> please confuse us (and yourself) with such new and useless informations.
Post your actual and complete schematic.

Klaus

 

[jean12]

The circuit remain the same with what I posted on #25 but the problem I am talking is only about increasing the voltage on the driver output which are connected to the Gate of the IGBT.
>>>The IGBT datasheet says that the VGE is <=15V whereas I am having VGE=6V;clearly you see that this voltage can`t fully turn on the IGBT therefore I am asking how to proceed for increasing that voltage from the Driver or if I have to increase the voltage(VGE,PWM Signal) before applying to the driver.

>>>On Post #41 you said:

To turn a FET/ IGBT fully on within a limited time you need some gate current.

;so I was asking you

what do you mean by Gate current?

Hope this is understandable.

 

[FvM]

The circuit remain the same with what I posted on #25.

Really?

Then IR2110 would never work because there is no bootstrap diode. The schematic also shows a Vdd of 9V which is too high to control the IR2110 input reliably with 5V logic. Vdd should be 5V.

 

[jean12]

Sorry,the bootstrap diode is provided and the voltage on pin 9(VDD)is +5V and VCC=15V,it is not shown here on the circuit.

 

[KlausST]

Hi,

we are going in circles.

I´m on the point where I´m not motivated to go on with this thread.
We repeatedly ask for informations. And it seems you are not interested in giving it to us.

--> The schematic in post#25 is far away form being complete
It is not actual because the bootstrap is missing

There must be some other issues, because with your schematic of post#25 it is about impossible for the PIC to heat and burn.
It´s now 55 post ...

It´s on you to provide all informations:
Measurement points, frequency, duty cycle measurement methods, scope pictures, schematic, photos of your circuit and so on.
We don´t know what´s happening on your desk, so come on, tell us.

I hope you don´t see this (my post) as impolite. We really are willing to help you.

Klaus

 

[jean12]

Hello KlausST ,thank you for your reply and advices,so let me give you more details about my current project,this one is different to the previous one but I posted it here because I didn`t want to hijack the forum with some questions which has relationship,so here are details now:

1.I would like to design a Full bridge with PIC16F1937(using its ECCP Modules) where my power switches will be controlled by SPWM (referring to https://tahmidmc.blogspot.com/2012/10/600w-50hz-sine-wave-inverter-test.html)
2.I want the system to be able to display on the LCD the following:
A.Output voltage and Load (in percentage),I do have a current transformer but not now to interface it to a microcontroller
B.Solar voltage
C.Battery voltage (DC Power bank)

A push button may be used to turn on-off the system
unfortunately when I started by displaying on LCD the LCD does n`t want to display anything,here in attachment is my circuit.
Experiment

Before adding those voltage and LCD,on the system I referred to https://tahmidmc.blogspot.com/2012/10/generation-of-sine-wave-using-spwm-in_10.html and the signal I got I applied to my power components to see if at the end I will get something but
**** With the IGBT I have SKM145GB066 (dual modules),I don`t get the output with 48VDC I get 10VAC/50Hz,which is not right I thought its because the IGBT are not fully turn on and from that I asked my question on #50.

When using the simple MOSFET IRFP460Z I get the output but the transformer of course is sounding.

The SPWM Generated is of 12KHZ of carrier frequency and the output voltage has to be of 50HZ and maximum voltage be 230VAC/50HZ.

With the technique of reading the output voltage (in AC) I would like you to help me to control so that the maximum and minimum possible output be fixed,in function of the battery voltages.

I hope here I provided full of information regarding my system.Find also the attachment.

 

[KlausST]

Hi,

Sorry, I should have been more clearely:
(The problem we talk about is the PIC PWM to IGBT output. We don´t talk about voltage measurement of your application, nor display control.

Correct me if I am wrong:
* We are talking about a "real" circuit, not a simulation only.
* In your circuit there is a problem. The problem is that the IGBT output is not like expected.
Is this correct?


If yes:
Is the folowing correct: (PWM signal path)
* The PIC generates PWM signals (12kHz, varying duty cycle)
* The PWM signals are fed to input of IR2110
* output of IR2110 is fed to IGBT gates
* IGBT outputs are fed to discrete filters (LC)
* Output of filters should be sine waveform
--> Do you agree that the problem lies somewhere in this signal path?

If all is correct:
(PIC)
The PIC needs supply voltage. I can´t find where this is connected. Please show a schematic where we can see the power connection.
Please perform the folowing tests and give us the results.
* use a voltmeter (DVM) and measure PIC supply voltage directely at the PIC supply pins.
* use a scope and measure the two PWM signals (two scope channels. PIC: P1A and P1D for example) of one (of a total of three) halfbridge. (You just choose ONE of three halfbridges and we further talk abut them)
Please say what pair of PWM channels you used.
In the one scope picture we need to see: Waveform, frequency, low level, high level, duty cycle (of both halfbridge signals)
post the scope picture of the signals directely measured at the PIC pins.

(IR2110)
* check if both signals (as shown in the scope picture) are the same at the IR2110 inputs LIN and HIN.
* Use a DVM and measure VCC and VDD directely at the IR2110 pins. Tell us the values.
* check with a DVM the voltage at both VSS and COM, both referenced to your PCB GND. Both must show below 0.05V.
* use a scope and measure the voltage at LO. Show us the scope picture.

*****************
Please do it step by step in the written order. Give us feedback on every single step.

Klaus