Hey guys. I am trying to build a DC to AC pure sine wave inverter. The circuit for which is attached. The gate driver ICs are being supplied by one power supply and the H-bridge is being supplied Vcc from a separate supply. The problem is that when I make the grounds common for both of them, one of the supply shorts.
What could be the possible reason for this?

P.S. I have tried rebuilding the circuit numerous times.

Attached circuit.



Also, I am feeding a square wave to the IR2104 and not a PWM.

Moderators note : a resize of the picture reduced the size from 1.7 Mb to 195 kb, insetred full image

 

So you have a problem when you connect the grounds of the the two power supplies? This problem is probably independent of your circuit?

You have created a path for current to flow from one power supply to the other. You always have to be very careful connecting line-powered power supplies. I usually start with a high ohms resistor, say 1KΩ, between grounds. I measure the voltage across it to see if there is much current. If not, I'll try something like 220Ω. If I don't see anything more than a mV or two, I'll go ahead and connect them.

Why so careful? I blew a computer sound board once when I connected its audio ground to the ground of some circuit I was working on. Two power sources, both coming from line power. Poof.

 

Use an ohmmeter to trace the short.

 

But for the correct operation of the circuit I need to connect the grounds.
And an ohmmeter is out of question because I have to disconnect at least one supply as soon as it shorts.

 

Hello,

You are only using 33 nF as boost capacitors.
Do you think that is really enough?
Also the capacitors accross the drain and source will not do any good to the fets.
The slow transistions will heat-up the fets.

Bertus

 

What makes you think that the H-Bridge will make a sine wave?

 

I changed the boost capacitor to larger values as well. For now, this works.
And yes I know that the output of the H-bridge will be a squar wave, but I plan to filter out a sine wave. This is just a part of the entire circuit.

 

Hello,

When using SPWM, you will get a better output signal that is much closer to a sinewave.
The following page of the EDUCYPEDIA has some info on the subject:
http://educypedia.karadimov.info/electronics/powercontrolspwm.htm

Bertus

 

Why are there 1k resistors connected between the bridge outputs and your hi and lo gate pins on the IR2104? Is there a purpose for this feedback arrangement? I'll admit to only reading the data sheet and not any app notes.

 

When a pure sinwave is needed than you should generate one with a PIC even allowing you to generate 3 phase signals when wanted.
do a level shift to voltage wanted.

 

The 1k resistors are only for protection purposes.
Thank you everyone for the valuable advice, but the problem that I am currently facing has nothing to do with the sine wave quality,that's a problem for later. I can't even get started because of the shorting problem, let alone checking the output of the H-bridge.

 

Back up one step and try again. Change from two separate power supplies to one.
Get a three leg regulator and try using just the 24 volt source and a 7812 or 7815 let's say, to supply the 2104. Eliminate the second source altogether.

 

All right. That sounds like a good idea. Thanks.