[SOLVED] Full Bridge Inverter with MOSFET and IR2110 gate driver

[Tahmid]

In that case, check the high voltage bus. It can not exceed 500v (maximum limit of the IR2110). Also, make sure the MOSFETs can handle the high voltage.

Hope this helps.
Tahmid.

---------- Post added at 14:00 ---------- Previous post was at 14:00 ----------

With no input signal, nothing should happen.

---------- Post added at 14:01 ---------- Previous post was at 14:00 ----------

There, I think I see the problem. Connect pin 2 and pin 13 (Ground) together. That should solve it.

---------- Post added at 14:02 ---------- Previous post was at 14:01 ----------

And, the ground for the IR2110 and the high voltage ground should be the same. There should be a common ground.

 

[stoyanoff]

I mean the microcontroller is not connected. Actually the 12V circuit is on and on both inputs I have 0 potential and i think the both transistors should be close for safety. I think even that I should have half of the input potential on the load resistor. Are you sure about the common ground? In the datasheet of IR2110 these is a chart where they are separated. If I connect them is there a chanse a burst form the load to destroy the input circuit of the ir2110???

 

[Tahmid]

You are powering the Vcc and Vb of the microcontroller from the 12v supply. However, if you do not connect the ground of that supply, the IR2110 gets no supply. I'm pretty sure they should be connected.

 

[stoyanoff]

I have a siple question. Is IR2110 a full-bridge drive? Can he fully open the low and high transistors and the voltage on the load to change from MAX to 0??? I saw the datasheets of some drivers as L6201 and these is witten Full-bridge driver, but for ir2100 the datasheet says low ond high side driver. I saw a chart where is used a transformer for creating full-bridge with ir2100. Is the next chart a full-bridge or it`s a half-brigde???

 

[vimalkhanna]

Is there a possibility that we use MOSFET IRFBG50 in Hi-bridge mode and drive thru IRF2110 for 230VAC /50 Hz inverter transformerless design ??
If possible, I wanted to go for IGBT output stages in the main inverter stage

 

[Tahmid]

IR2110 is a high-side low-side driver, not a full-bridge driver. This is because it has 2 outputs to drive the MOSFETs/IGBTs (high-side and low-side). You will find that full-bridge drivers have 4 outputs (or maybe even 6 for 3-phase) to drive all the MOSFETs/IGBTs in the bridge.
IR2110 can be reliably used to control MOSFETs to change the voltage on the output from 0 to MAX, provided MAX is within the safe limits of the IR2110, ie 500V.
The diagram shows a 3-phase full-bridge converter, with 3 high-low side drivers driving it.

Hope this helps.
Tahmid.

 

[stoyanoff]

I`ve created a full-bridge with 2 IR2110. When I measured the output voltage it was fine(I think so. My program for the microcontroller is only the test.For now it`s not proper for a outputing 220V sinusoidal). I have a small concern. The temperature of both IR2110 increases very fast after I connet the power supply. Is this a problem or it`s normal? If it`s not where is the problem???

 

[ian123]

please submit your drawing and i will help you with it.
for full bridge circuit
first off in order for the IR2110 to work correctly you need to switch on the lo side first so it can charge up the bootstrap capacitor then turn on the high side
since it is full bridge I reccomend switching on both lo sides to charge caps then switch on the hi side you require. There also needs to be some dead time between switching from one direction to another otherwise you will get overshoot which will kill the mosfets.

In my experience it appears that your driver is blown but in order to check this the best is to take out the mosfets and tie the VS outputs to ground then check to see if they heat up if they do they are blown. Also to check the pwm while the circuit is connected this way you can use a scope on the gate resistors and see if you get the correct pulses. If the pulses are correct then the IR2110 wont heat up and would be working.

If you post a drawing of your circuit I will check it for you and tell you where you have gone wrong.

I`ve created a full-bridge with 2 IR2110. When I measured the output voltage it was fine(I think so. My program for the microcontroller is only the test.For now it`s not proper for a outputing 220V sinusoidal). I have a small concern. The temperature of both IR2110 increases very fast after I connet the power supply. Is this a problem or it`s normal? If it`s not where is the problem???

 

[stoyanoff]

Here is the one half of the bridge. The another one is the same. I`ve change the boostrap diod with faster one and there is no heating now. But I have another problem. My bridge is not working now. As you can see I`ve created a small reostat with 2 resistors and I`m measuring the voltege between the exit of the half-bridge and this mid point where the voltage is 110V. The voltage between this points is 1.3V. I`ve tryed to open only the low transistor and I`ve expected the voltage to be atleast 70-80V. I`ve used 10ms on time and 200us off time. But there was no result - 1.3V. I`ve trqed to measure the voltage between the both side of the full bridge but the result is almost the same. I`ve tried to increase the supply voltage to 18V - nothing. Any suggestions???

 

[ian123]

what is the voltage across c3

---------- Post added at 18:36 ---------- Previous post was at 18:30 ----------

C2 should be at elast ten times c3 r5 r6 should be lower in ohms 3.3 ohms is good

---------- Post added at 18:37 ---------- Previous post was at 18:36 ----------

what is the voltage on Hin when you switch u3

 

[Tahmid]

Are you sure you used 6.8k for R8? The diagram shows only 6.8 . I assume that is just a typo.

Decrease the gate resistances to about 10 to 33 ohms.

Connect a 1k resistor between the gate and source of the MOSFETs.

You haven't connected Vss to ground. Vss needs to be connected to ground.

Have you connected the 15v ground and the 220v ground together? You should. I think this is your main problem.

Hope this helps.
Tahmid.

 

[stoyanoff]

I`ve created a full-bridge with 4 transistors. Here is the chart.

As you can see I`ve connected the Vss and COM together. I`ve changed the boostrap capacitator with 1 22uF electrolite cap and 1 220nF non-electrolite.
I tried to create an output sinusoidal signal. I load in the microcontroller(PIN18F2550 quartz 40MHz) a program with this function for PWM - sin(point*2*3.14/200)*100. As you can see I`ve devided the whole period to 200 point and with this equation I`m calculating the time of every output impuls. I`ve added an extra delay(100us) between the change of every + and - half-period.
In this case when I measure the voltage between the both mid points of the bridge(LOAD) I`m registrating an output AC voltage in large borders - from 0 to 170V.
With this fast changing freqency I think it`s OK, but when I connect a load(an old fan) nothing happens. I supposed that my boostrap capacitator(2.2uF in my previous tries)is too small to maintain the high transistor open long enough, so I replaced it with this 2 capacitors.

I`ve not change the gate resistors because I think they just controll the time of swithcing of the transistors, but if it`s a big problem I`ll do it.
Can you give me an advice where could be the problem???

Thanks a lot!

 

[ian123]

change the gate resistors

also i need to know what your 220v psu looks like.
what is the current draw of your old fan and voltage 220 ac or 220 dc
I have had a problem where the micro was locking up this was caused by interference from the hi side circuit i had to separate the power supplies totally.
when i get home i will calculate the timming for your caps but i doint think this the problem.

---------- Post added at 15:43 ---------- Previous post was at 15:39 ----------

one more thing if you are trying to run a motor make sure you put fast recovery diodes across the source and drain of each mosfet the diodes included in the mosfet are not strong enough.

---------- Post added at 15:45 ---------- Previous post was at 15:43 ----------

Why are you using different mosfets in the circuit

 

[stoyanoff]

The bootstrap diods are SF38 - Trr=35ns. The fans is small. It`s on AC voltage and I don`t think it`s draining a lot of current. I don`t think it`s a problem because the mosfets have 20A max current. All of my MOSFETs are IRF840. I forgot to change the names of the 3th and 4th.I`ll put extra diods to MOSFET, but I don`t believe this would resolve my problem. The evarege output voltage is low - around 10V.
I`ve measured the output voltage of a real frequency invertor and it`s output voltage has peaks of about 1000V and evarege voltage around 140V.
I think my bootstrap cap is small, but it`s already atleast 10 time bigger then my calculations. Could I try with bigger one - 220uF may be???

 

[Tahmid]

I don't think the bootstrap cap is the problem. Do one thing. Remove the connections from the IR2110s to the MOSFETs. Then, connect pin 5 of both IR2110s to ground. Connect your PIC output to the input of the drivers via the optocouplers. Watch the output at pins 1 and 7 with an oscilloscope. Check to see that you get proper waveforms, similar in shape to those at the PIC output, but greater in amplitude. Is the 15v power supply isolated from the PIC power supply? If not, you can omit the optocouplers if isolation isn't necessary and you can connect the microcontroller outputs to the IR2110. What is the value of the resistor between the optocoupler emitter and ground? Change R5, R6, R9 and R10.

Hope this helps.
Tahmid.

 

[stoyanoff]

Tahmid, this is a good idea!
I changed the gate resistors to 50 oms each and I connected an extra diod in paralel with each of them. I decreased the points of the sinusoid to 10 for each half-period(1 ms impulse and 30us pause). The output voltage is not constant and it`s varring in large borders. I think the efective value is lower then 10V in spite of it`s rising above 160V. I`ve connected a electric lamp for a load, but it didn`t turn on.
I`ll see what I can do with the output of the drivers.
My optocouplers are slow(CNY17-4), but when I`d increased the impulses I expected the output voltage to be more stable but it wasn`t

 

[Tahmid]

Check the output waveforms from the driver and check it with that coming from the PIC. Do you have spare IR2110s so that you can use them to check if the ones you're using are burnt or not. Also, you might want to put a 1k resistor from the gate to source of each MOSFET.

Hope this helps.
Tahmid.

 

[stoyanoff]

One more thing. I`ve noticed that when I use a pulses with pwm(not static 1ms and 30us pause) the value of the output voltage is bigger. I suppose because of the dead time. How long should be the dead time? Could the problem be here in my program???

 

[stoyanoff]

I`ve checked the output signal from the controller and the driver. The differnce is enormous. The IR2110 registrates less impulses. Any ideas why??? I`ve removed the capacitator between Vcc and COM, because it`s a obstacle when I was creating the desing. I remember in the application note was writen this capacitator prevents the GRD to sank below 0. Could it be the problem???

 

[Tahmid]

So, your problem lies in this drive circuit. No, the capacitor shouldn't be the problem. Ok, check the pulses at the IR2110 input, meaning the opto-coupler output. What is your modulation frequency? Maybe it is too high to be used with the opto-coupler you are using.