Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Full-bridge Inverter (spiking current)

Status
Not open for further replies.

dm927

New Member
I am trying to build a full-bridge inverter circuit that will take 170VDC and give a 340 (peak-to-peak) or 120 VRMS output at 60Hz. In doing so, I am using IRF740 transistors [Vdss = 400V, Id = 10A] driven with IR2181 gate drivers at 18kHz. The voltage and current at the load (100-ohm, 225W) seem correct. The problem is when the DC rail voltage is increased to 100VDC, the current immediatly spikes. Then, the transistors instantaneously heat up and the multimeters that were set up to examine the input and output current read "Error". Does anyone know what the problem could be?
 
I suggest you post the circuit diagram as it is impossible for us to help without one.

Len
 
I have examined your circuit and have some questions and suggestions. But first, some house keeping.

You should have labelled the components in order to make it easier to discuss by post. So I have labelled the transistors Q1 - Q6 where Q1 - Q4 correspond to G1 - G4 and Q5 & 6 are the upper and lower transistors of the "half bridge DC- DC converter". I also labelled the rectifier bridge RB1 and the 220 uF capacitor C1.

Questions
1. I assume the transistors that are over heating are the Q1 - Q4, not Q5 & 6. True?

2. I assume the PIC generates PWM signals that, in conjunction with the MOSFET driver circuitand the 0.01 uF capacitors across the Vb & Vs pins of the MOSFET drivers somehow create a sinusoidal drive signal for Q1 - Q4. True?

3. You said that when the voltage reaches 100 Volt, the "current spikes" and the transistors overheat. I assume this voltage is measured across C1. True?

4. If 3 is true, then I assume that you disconnected RB1 and applied a variable voltage source to the drains of Q1 & Q3. True?

Conclusion
If 1 is true and my other assumptions are true, then it would appear that transistors Q1 & Q2 and Q3 & Q4 are all turning on when the voltage reaches 100 Volt thus causing an excessive current through them and causing the overheating and "current spikes".

Have you looked at the gate waveforms with an oscilloscope? This would show whether Q2 & Q3 are off when Q1 & Q4 are on and vice versa.

Len
 
Full-bridge problem

Sorry about the schematic. I will go back and clean it up.

Answers to your questions:

1. Yes, Q1-Q4
2. true
3. Yes, the 100VDC was measured across the 220uF filter capacitor (C1)
4. Yes, I have a isolation transformer connected to a variable transformer. The variable transformer is then connected to a full-wave bridge rectifier.

I am not sure how to measure two channels simultaneously because I am using 2 channels at once to make a differential measurement. How can I do this?

I have place a copy of the transistor waveforms at the following link:

https://www.electro-tech-online.com/custompdfs/2004/02/mosfetwaveform.pdf
Today I found that the Vgs increases with an increase in rail voltage
(Vgs should remain at about 12V). Do you know how to fix this problem? The IR2181 drivers are supposed to keep the Vgs at the required voltage.

I have attached a document showing the waveforms that were recorded today. Thanks for your help.
 
Full-bridge problem

I have not studied the diagrams yet. Will do so later.

You asked "I am not sure how to measure two channels simultaneously because I am using 2 channels at once to make a differential measurement. How can I do this?

I suggest the following:-

1. connect the Gnd of the scope to g2 and observe the waveforms at G2 and G4. HOWEVER, be sure that the 12 Volt supply is not connected to GND! ie. the only Gnd should be from the scope. I have not looked at the data sheet of the IR2181, is it an Optical isolation device? If so then this may not be a problem. Given also that your transformer is also isolating the 12 Volt from the high voltage side.

2. then (leaving the Gnd of the scope to g2) connect the external trigger input of the scope to G2 so this can be used as a time reference and use the differential inputs to the scope across G1 and g1. HOWEVER, be sure that the scope can withstand the common mode voltage. I expect that you are using x10 probes, so this should not be an issue, but please check otherwise it may be an expensive error.

3. repeat 2. with the differential inputs connected to G3 an g3.

It is essential to see the time relationship between all of the waveforms in order to see if Q1 and Q2 are turning on at the same time. And ditto for Q3 and Q4.

Len
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top