Sine wave power inverter

[mat janssen]

Normaly I also make the flowdiagram first and then write the program and when it is a simple aplication i don't make a flowdiagram.This program was made a few years ago so I have to look to the comment and in that time I did not make a flowchart.
Here is the data of the pwm of a sine wave in 200 steps.

128,132,136,139,143,147,150,154,158,161,165,169,17 2,176,179,182,186,189,192,195,
199,202,204,207,210,213,215,218,220,223,225,227,22 9,231,233,235,237,238,240,241,
242,243,244,245,246,247,247,247,248,248,248,248,24 8,247,247,247,246,245,244,243,
242,241,240,238,237,235,233,231,229,227,225,223,22 0,218,215,213,210,207,204,202,
199,195,192,189,186,182,179,176,172,169,165,161,15 8,154,150,147,143,139,136,132,
128,124,120,117,113,109,106,102, 98, 95, 91, 87, 84, 80, 77, 74, 70, 67, 64, 61,
57, 54, 52, 49, 46, 43, 41, 38, 36, 33, 31, 29, 27, 25, 23, 21, 19, 18, 16, 15,
14, 13, 12, 11, 10, 9, 9, 9, 8, 8, 8, 8, 8, 9, 9, 9, 10, 11, 12, 13,
14, 15, 16, 18, 19, 21, 23, 25, 27, 29, 31, 33, 36, 38, 41, 43, 46, 49, 52, 54,
57, 61, 64, 67, 70, 74, 77, 80, 84, 87, 91, 95, 98,102,106,109,113,117,120,124

The data is the on period of the pwm.

[batee]

I disassembled a DOA APC ES500 UPS the other day, and found that it contained a 28bit PIC (can't remember which). There were other ICs between the PIC and the output FETs (SG????'s).

[sougata]

Hi everybody,

Thanks for the table. The schematic block that you have posted indicates that you are using an already boosted battery voltage to drive an IGBT bridge. I am trying to drive it straight into a transformer for boost up. I wonder if this is going to work.

By the way batee any PIC with an onboard HWPWM and ADC will do. You are damn right about the APC UPS.

Regards

Sougata

[mat janssen]

When I understand it good, you are going to switsch the 24 V over the bridge into an output transformer, and the voltage what is cumming out of the transformer, will pass the low pass filter for making a sinewave.
I hope you will find a good output transformer for it !!!???

[sougata]

Hi Matt,

I understand your concern but I am hopeful here since I have got my own transformer manufacturing unit. So from paper to product and fine tuning issues I have got a plus point over here.However I am not sure of the actual losses and bulk of the design for practical use so I might need to design a battery boost circuit.

Regards

Sougata

[mat janssen]

Please let me know if it works and how you did it. And if you have some questions about the hardware, just let me know, maybe I can help.

[sougata]

Okay Matt,

That would be great. I need to design a RCD snubber and I don't know the maths may be you can help. I am concerned about the switch off peaks.
I would send you the entire schematic as soon as I finish it up. It will take some time but I will keep in touch.
Regards
Sougata

[psdayama]

Dear Sougata,
I haven't seen the entire code at all but just one table of delays for
sinewave. Also I have faint recollection of the block diagram.
Truly speaking Im not much PIC er. Im doing everything with ATMEL
micros. I have found this forum while searching for sinewave inverter.
I think it doesn't matter which company U use but design is more importatnt.
So I can understand code very well.
I also have Transformer Unit but it is for HT Instrumentation type.
I think that varying amplitude with delays and error adding will work only
maximum upto 70% of basic amplitude. After that U will have gross THD due
to most of delays at zero crossing very small.(It appears in class D amplifiers
of analog kind). I haven't seen whether such approach works with truly
variable amplitude or from 0 to 120% of the amplitude.
What I m saying is not double conversion but switchng of Buck Boost
mosfets with 2 ports from same PIC outputs and CORRECT amplitude there.(See attached bmp file.) In this s1 and s2 are mosfets whose timings is controlled to vary the voltage.
Im in India but near Mumbai so here not much Inverter brouhaha is there.
Melanie has agreed that this DSP method is not so good although everybody
is claiming otherwise.
I am also looking for design in which there should be no such bulky
transformer of 50Hz. If I have read right then proper driving of mosfets
would result in 20-40Khz output which is double modulated. this can use
high frequency transformer with centre tapped secondary and demodulator cum filter circuit which will give 50Hz output!!
As presently the transformer is costing about 50% of the total invertor
cost and one can imagine savings. Also as many circuits are having SMPS type charger for battery the advantage is obivious.(Compare APLAB 30V 1A
regulated supply with 250W computer smps,)
Hope to have more discussion on this.
regards,
dayama