• 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.

H-Bridge Thyristor Inverter

Status
Not open for further replies.

mdbrighton

New Member
Hey all, am designing a DC-AC inverter circuit. Converting 12V DC to 12V AC, then put it through a transformer to step-up the voltage. Have found a Thyristor based H-Bridge topology, but the resultant wave is either a square wave or quasi square wave. The topology can be seen here http://www.electro-tech-online.com/custompdfs/2009/04/Chapter2014.pdf

There are a couple of questions I've got:

Firstly, I was advised to use a PIC to generate the pulses to turn the thyristors on. The question I have is, for that particular topology, can the thyristors be all pulsed at the same time? Or is there a better way of controlling the circuit then using a PIC?

Secondly, with a square-wave output, what is the best way to convert it into a pure since wave? I considered using a filter, but was advised against it by my supervisor due to power losses.

If not......what other inverter circuits would you recommend?
 

mdbrighton

New Member
Could a Bandpass filter be used to convert the square-wave to a sine-wave?

Edit:

I have written a simple PIC program which sends out pulses, planning on using C106D thyristors.

RA0 EQU 0
RA1 EQU 1
RA2 EQU 2
RA3 EQU 3
RA4 EQU 4
RA5 EQU 5

RB0 EQU 0
RB1 EQU 1
RB2 EQU 2
RB3 EQU 3
RB4 EQU 4
RB5 EQU 5
RB6 EQU 6
RB7 EQU 7


#Include <P16F84A.inc>

__CONFIG _XT_OSC &_CP_OFF &_WDT_OFF & _PWRTE_ON

CBLOCK 0X24 ;SA OUT OF VARIABLE
COUNT ;ADRESSES.
ENDC ;END OF CBLOCK

;Initialization of PortA pins

Start Org 0x00 ;

CLRF COUNT
SETPORTS BSF STATUS, RP0 ;
MOVLW B'11111111' ; PORTA SETS AS INPUT
MOVWF TRISA ;
MOVLW B'00000000' ; POPRTB SETS AS OUTPUT
MOVWF TRISB ;
BCF STATUS, RP0 ;

GETON CLRF PORTB ;RA3=0 LIGHTS UP LED
CLRF PORTA ;
NOP
NOP
HOLD BTFSS PORTA, RA3 ;IF RA3=1, PROGRAMME BEGINS
GOTO HOLD ;
NOP ;
MOVLW B'11110000' ;
MOVWF PORTB ;SET RB6 AND RB7 PINS
NOP ;

ClRF PORTB ;
CLRF PORTA ;
NOP ;
MOVLW B'11110000' ;
MOVWF PORTB ;SET RB6 AND RB7 PINS
NOP ;

ClRF PORTB ;
NOP ;
MOVLW B'11110000' ;
MOVWF PORTB ;SET RB6 AND RB7 PINS
NOP ;
;
MOVLW 0X10 ;
MOVWF COUNT ;
CLEAR DECFSZ COUNT,F ;REGISTER
GOTO CLEAR ;
GOTO STOP ;
NOP ;
NOP ;
;
STOP CLRF PORTA ;
CLRF PORTB
;........

END ;END OF PROGRAMME.



The only thing is, the pulses are all sent out at the same time, and Im not too sure about it.
 
Last edited:

jrz126

Active Member
It would appear that you setting all four of the outputs high at the same time. You probably need to change some of the "MOVLW B'11110000' ;" lines.

You can vary the pulse widths to get a sine wave out.

This is similar to how Class D amplifiers work.
 

mdbrighton

New Member
So are you saying that with that particular topology, by changing a few lines of program and varying the pulse widths, it would be possible to get a true sine wave out?

Everything in the documentation lead me to believe that the best I could get was a quasi square wave output, which could then be smoothed out using a filter of some sort.

Edit: How about using a logic inverter at two of the ports, because T1 and T2 need to be fired together, and then T3 and T4 are fired together.
 
Last edited:

jrz126

Active Member
I think you would be able to get pretty close. Almost half of the pdf in your link talks about using PWM. It starts on page 13.

It should be pretty simple if you are trying to run at a fixed frequency (50/60 Hz). You could make a lookup table with firing commands that will give you the desired sine wave output and just step through it. Although the firing pattern may have to change based on the load.
 

mdbrighton

New Member
Cheers mate, I actually recently started reading up about PWM techniques and power inverters, so excuse any silly questions. Yes it is at a fixed frequency of 50Hz. The way you put it out though seems like a couple of more days work.

Another question I had was about high end drive, in a MOSFET H-bride, you have that issue, but what about in this particular topology?
 

tcmtech

Banned
Most Helpful Member
I would not try to run an SCR based switching circuit on this type of drive application.
SCR's are a very fussy about all that high speed high accuracy PWM switching during each half of the cycle. They tend to lock up to easy! Even with a fairly complex control system attached to them.

But using Mosfets or IGBT's as the switching devices would work very well!

You dont need to get close to having a sine wave going into a properly designed filter stage. The power losses are actually rather small.
Whats being shaved off during part of the cycle gets put back in an other part.
 

colin55

Well-Known Member
A simple way to get a sinewave from the output of a micro is to use a resistor-capacitor network and even an inductor. I used this principle to get a near-perfect DTMF output for a dialing alarm. And you know how accurate the DTMF waveform needs to be, for reliability.
 

mdbrighton

New Member
I would not try to run an SCR based switching circuit on this type of drive application.
SCR's are a very fussy about all that high speed high accuracy PWM switching during each half of the cycle. They tend to lock up to easy! Even with a fairly complex control system attached to them.

But using Mosfets or IGBT's as the switching devices would work very well!

You dont need to get close to having a sine wave going into a properly designed filter stage. The power losses are actually rather small.
Whats being shaved off during part of the cycle gets put back in an other part.
Hmmm....so would MOSFETs work in that particular topology? Would you happen to have a schematic of something which could be controlled by a PIC?


Another question I had was on how to know how much of a delay I should use to get the pulse widths? That is the length of the pulses. Is there any particular maths involved? Any link or help with that would be appreciated. Oh and could I use a PIC16F84 for it?
 

tcmtech

Banned
Most Helpful Member
I've got a bunch or different inverter schematics. I personaly dont really care for the micro controller stuff.
They are great and very usefull little devices but they tend to get used too much for basic stuff that really does not need that level of sofisticated control.
a large amount of micro controller missuse is from people trying to force a system to act in the theoreticaly perfect condition. Not the natural condition.

There are countless IC's out there that can do the same job and are dedicated to controling that type of inverter circuit you are designing. They are simpler and easier to work with and have proven themselves to be very adaptable too!

For low voltage, high current, high speed switching control you will find Mosfets are hard to beat! IGBT's Work on the same voltage control principles as Mosfets but are able to do the switching at much higher power and voltage. They are a super output stage switching device.

Just something to think about.
 

mdbrighton

New Member
Ok, I've narrowed it down to using IRF530 MOSFETs(IRF530 Datasheet pdf - N-CHANNEL POWER MOSFETS - Samsung Electronic), planning to drive them with HIP4080 driver chip(Datasheet Archive - HIP4080AIP datasheet - HIP4080AIP application note - HIP4080AIP pdf), which has two input ports.

I could write the code in assemly if I knew the delay times for each width(Just testing abit of code right now with that), what kind of IC's are you talking about? I dont mind doing the work programming but if there is a ready made chip, it would be good to compare it with to a PIC based inverter circuit in my report.
 

tcmtech

Banned
Most Helpful Member
I will try to come up with the right schematic if I can still find it.

I have about 14 gig of rather poorly sorted tech info data.

The most basic one was a simple 2 channel class D audio amplifier IC that drove a set of mosfets that made up a standard push-pull half bridge inverter input.
It used a simple 60 hz sine wave signal being fed to the amp IC's inputs trough an AGC control IC.

Literaly three simple and cheap IC's could give you control over any size of sine wave inverter you could build!

One channel of the amp IC drove one set of mosfets as a non inverting output. The other channel was set up as an inverting output. That was about all there was to it!
Given the right size and number of mosfets used on each side of the push-pull circuit it could theoreticaly run almost any wattage of inverter.
There was a simple voltage feedback loop comming from the output side of the step up transformer that fed the automatic gain control IC that was between the sine wave generator and the amp IC.
If the output voltage dropped it told the AGC to turn up the volume and thusly raise the PWM timing!

Do a search on any of the IC data sheet websites for a Signal Generator IC, An Automatic Gain Control IC, and a Class D audio IC.

It will give you a good idea of how simple this is!
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top