Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

SPWM Inverter using IGBT's and PIC18F4520

Status
Not open for further replies.
According to you above circuit, you are showing current and voltage at a resistive load.
Yellow - output voltage on the load;
Blue - output current on the load;
The voltage looks like a clean sine waveform, the current doesn't.

Three possible explanations (there may be more):
- the load isn't resistive
- the measurement is incorrect
- the cirscuit is different

My previous remark about the filter dimensioning being unsuitable was mainly related to the observed resistor overload, which also shouldn't happen.
 
  • Like
Reactions: ledoo20

    dgs117

    Points: 2
    Helpful Answer Positive Rating

    ledoo20

    Points: 2
    Helpful Answer Positive Rating
The load is 5x100W incandescent light bulbs. I assume they are resistive.

Other parameters and circuitry did not change. Only different capacitor values were tested.

And what about measurement? What am I doing wrong? How to measure properly?

Added after 2 hours 33 minutes:

You were right about the measurement. I changed the position of the current clamp at different places and look what I got.

6.8uF cap, 0.6mH inductor, 500W load.

So it was a problem of distortion of some kind which caused the incorrect current readings.

Now the question is, how to measure current correctly?
 

Today I added a shunt resistor of 0.015 ohm in series with the load and measured the voltage across it. I got the following waveforms. First picture contains image of sampled waveforms and the second contains averaged waveforms of 64 samples.

Doing simple calc one can determine the current flowing through the load:

Umax = 50 mV = 0.05 V, R = 0.015 ohm.
Imax = 0.05 / 0.015 = 3.33 Amps.
Irms = 3.33 / √2 ≈ 2.357 Amps.

The current clamps, which measure the rms current, showed 2.2 Amps current flowing through 500W load.


Now the question is how to measure THD of the voltage and current?

And one more observation - the inductor is getting hot during the operation. What is the cause for that?
 

Well, I found out why the inductor core is getting hot. It is connected to the iron powder core (#26 material) losses which depend on the rms voltage, number of turns, frequency and cross-sect. area of the core. So, according to my calc, I get about 35W of wasted energy through the heat. And these losses should not depend on the load, right?

Please, could someone explain, where do these losses come from in the iron core? Or could you suggest some books and other sources where I could study this thing.

Thanks in advance.
 

I am a bit confused by the switching arrangment. Any way you switch this H-bridge you are getting shoot through at the zero crossing. Any h-bridge we have put together, Q1 and Q2(caddy corner) are fed the same PWM, with Q2 input inverted. Then Q3 and Q4(caddy corner) are fed the other. See PDF. This should help your heating problem also, since the current is not being "forced" through the load, instead it will flow. Correct me if i am wrong but it looks like you just connected two half bridges together. Other than that everything else looks very nice.
 

Any way you switch this H-bridge you are getting shoot through at the zero crossing.
You ignored the circuit details given by dgs117. As far as I understand, there's no "shoot-through" or cross-conduction problem
with the present circuit. According to the posts, there are no excessive IGBT losses. The discussion was about distortions in
the filtered output and inductor losses.

Regarding inductor losses, they depend on both DC (respectively 50 Hz) current bias and AC magnetization. Core size and
windings dimensioning always involves a tradeoff between losses and size respectively costs.

For a inductor with DC/low frequency bias current, it's important to select a suitable air gap respectively sufficiant low µ
parameter for a powder core. Personally, I'm not designing with powder cores, but I have a Magnetics data book and
know, they offer all cores with a wide µ variation. The data book is also suggesting a design procedure for biased cores.
 

    dgs117

    Points: 2
    Helpful Answer Positive Rating
Indeed the IGBT switching losses are minimal.

Now I am experimenting with different LC filter configurations, trying different iron powder cores, different number of turns, different capacitor values.

So far I found out that the losses mainly depend on the size of the core and number of turns (value of L).

Can you direct me towards right literature I could study for my design process?
 

dgs117 said:
Indeed the IGBT switching losses are minimal.

Now I am experimenting with different LC filter configurations, trying different iron powder cores, different number of turns, different capacitor values.

So far I found out that the losses mainly depend on the size of the core and number of turns (value of L).

Can you direct me towards right literature I could study for my design process?
Hi,
just joined this forum. How did you go with your sine wave filter? What self-resonant freq's observed? Can your filter be used with an inverter (1.1kW) at 400 Hz output?
Also, with "carrier" (modulation) freq of 15kHz?
Are you intending to feed filtered energy back to the DC Link?
Cheers,
Qtron.
 

qtron said:
dgs117 said:
Indeed the IGBT switching losses are minimal.

Now I am experimenting with different LC filter configurations, trying different iron powder cores, different number of turns, different capacitor values.

So far I found out that the losses mainly depend on the size of the core and number of turns (value of L).

Can you direct me towards right literature I could study for my design process?
Hi,
just joined this forum. How did you go with your sine wave filter? What self-resonant freq's observed? Can your filter be used with an inverter (1.1kW) at 400 Hz output?
Also, with "carrier" (modulation) freq of 15kHz?
Are you intending to feed filtered energy back to the DC Link?
Cheers,
Qtron.

Well, the resonant freq depends on the chosen parameters of L and C. You must take your modulation freq into account. To find the freq characteristic of the output filter I recommend using free LTspice simulation software (just google it). Designing filter you must ensure that the majority of the high order harmonics will be filtered out (in your case harmonics around 15kHz).

The power of the inverter mainly depends on the power of the output filter's inductor (thicker wire, bigger core and etc.)

For now my task is to experiment with output filters. The next step will be to make some kind of control loop to ensure the stability of the DC link.
 

There is a Sinusoidal filter, type CNW M 933/2
see Reo.co.uk This filter uses 29mH & 0.22uF
your values are vastly different. Can U pls shed some light on this?
If U use an Amorphous Ribbon core (see https://www.hilltech.com/products/power_components/amorphous_nanocrystalline_cores.html)
then heat will dramatically reduce!
I wanted to know what self-resonance you actually measured, not simulated, with your cores.
Some manufacturers of sine filters permit a max base freq. o/p of 150 Hz, & a max mod. freq. of 5-6kHz. These filters are therefore useless on modern drives.
Why is there a need for a control loop for the DC link? Surely the motor load would
prevent over volts if that is what you are referring to, or are there other factors?
Do you think that it is possible to do LC bench tests with a safe much lower voltage & current, keeping the impedances real?
Cheers,
Qtron.
 

hi friend i am also doing this project,i am using pic 16f877A please send me the one part of the code my volt 220v,180 ohm
 

hi every one, am trying to work on this project using PIC18f4550 but i dont have the mains .......

PLease i need the mains for the code below


// Half period 32 points sinewave lookup table
const unsigned short int sinewave[32]=
{
0,25,49,73,96,118,137,159,177,193,208,220,231,239,
245,249,250,249,245,239,231,220,208,193,177,159,137,
118,96,73,49,25
};

unsigned short int i=0;

void interrupt()
{
if(TMR0IF_bit) // TIMER0 interrupt
{
TMR0IF_bit = 0; // clear TMR0 overflow interrupt flag bit

if(i==32)
{
CCP1M3_bit = ~CCP1M3_bit; // CCP1 PWM mode togle
CCP1M2_bit = ~CCP1M2_bit; // ...
i=0;
CCP2M3_bit = ~CCP2M3_bit; // CCP2 PWM mode togle
CCP2M2_bit = ~CCP2M2_bit; // ...
}

CCPR1L = sinewave; // put a new value for PWM1 duty cycle
CCPR2L = sinewave; // put a new value for PWM2 duty cycle

TMR0L = 66; // inter. for 50Hz output @ 20 MHz, prescaler 8

i++; // increment table counter
}
}
Code:
Code:
 

Hi,

I am also working on an inverter project using igbt. I am switching them with unipolar spwm @ 12.2KHz. No matter what i do, I get lower order harmonics in 100-250 range.

I am not using any o/p filter, just 10 uF capacitor in parallel with the transformer secondary.

Would really appreciate some help here.

Thanks
Manish
 

hi
i think you must be check this other one wiring-scheme

 

Hi,dgs117 did you try to check the wave form from your microcontroller how is the wave like and after your igbt's driver's did you notice any noise in the signal if no, kindly upload your circute diagram and let see where this noise is coming from because Many inverter uses only capacitor across the output as filter.
 

Hi dgs117
Could you share us the full c programm code including the configure?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top