My Boost Converter Notes

[s4w2099]

Hello everyone,

Since last year I have been studying switchmode power supplies and building different prototypes. I have worked specially with inductive boost converters so I decided to write a little bit about them in my website.

This one is the second part of a series of writeup which main purpose is to describe and develop a pure sine wave inverter. The link to the boost converter webpage  and full schematics is:

http://www.s4wsbox.com/?q=node/26

and the first part of the series discussing mainly the inner workings of a sine wave inverter can be found here:

http://www.s4wsbox.com/?q=node/25

I am still developing all my circuits and learning new stuff so any criticism and suggestions are welcomed.

[commanda]

Hysteretic Class D amplifier.

read everything here.

http://www.diyaudio.com/forums/forumdisplay.php?s=&forumid=49

If you're running at 50Hz, a 3.2768Mhz crystal and binary divider will give 50Hz square wave. A low pass filter will give a 50Hz sine wave. Then your digital clocks will keep correct time.

60Hz is a little more difficult, but can be done.

I have a working prototype of the crystal oscillator, divider, LPF, driving a class B bridge amp, into a transformer; outputs 240 volts at 50.00000 Hz and 10 watts sine wave, runs my digital clock. Yet to refine the class D amp using an LM339, with the other 3 gates of the LM339 for the LPF, amplifier, and virtual ground.

You did ask for suggestions.

Amanda

[s4w2099]

I know what you mean. My inverter right now does not have a frequency control as stable as yours. It can power reliably all resistive loads and MOST motors. I have noticed that powering a cheap pedestal fan it lacks some power. Not very noticeable but I can tell. I was guessing that it had something to do with the frequency control.

I am trying to avoid the big, expensive, heavy and lossy transformers and thats why I am using the boost topology to boost the voltage to the required levels.

I have been thinking about using a micro controller to generate my encoded sine wave. A micro timed with a crystal should give me a precise frequency. It will also lower the costs associated with the many discrete components needed by the current version of the inverter.

Some time during this week I will post a video of the booster working. Maybe powering a bunch of CFLs with some efficiency and temperature charts etc.

Thanx for the suggestion

[oztules]

S4w,

I haven't worked on an inverter (commercial) that has used a booster for the HV step up. All seemed to use push-pull step up stages (600W and upwards).

As the run of the mill elcheapo inverters in modified sine are cheap as chips, (probably cheaper than building the front end yourself), it may be reasonable to use the step up pushpull stage of a modified sine inverter to cheaply achieve your HV DC.voltage, and homebrew your D class output from there.

At the least this will allow you to test your D class output stage with an established current controlled drive voltage.

It is interesting to note that in the modified sine inverter, the H bridge is there, the crystal oscillator is there, but they don't modulate the output with a 50hz sine, but a lousy  square wave....... there must be some reasonable explanation which escapes me at present.

..........oztules

[commanda]

there must be some reasonable explanation which escapes me at present.

Most of those circuits have simple gate drive circuits which excludes using it in the way you envisage. Driving an H-bridge running on high voltage, at high frequency (you'd probably want 100 times the output sine wave frequency to run it effectively in class D) is not trivial.

I'd love to work on it if I didn't already have too many projects.

Besides, I recently scored 2 x 1600 watt true sine UPS's, so I think I'm covered there.

Amanda

[oztules]

yes, ....

What I was getting at was the main things are already in place, the H bridge is there, the HV DC is there, a stable 50hz oscillator there already, it seems not too much of a jump to use say a tl494 running at 35khz, a totem pole driver, small toroid driver coil to isolate the high and low sides of the bridge, modulate the error amps of the 494 with the 50hz, and I imagine we would be close...... but your right, may have to do a project on it.

(Just picked up some MC33033 chips from the Post Office today, so first I want to build a 1000w three phase brushless motor drive... too much to do as you say)

........oztules

[s4w2099]

As commanda said one needs to have High frequency driving the H bridge. The way I like to think of it is that the driving frequency is the "output resolution" of my desired sine wave (less distortion of the wave).

I thought about using a push pull converter instead and I think its a totally sensible solution. The thing is that I dont really need this inverter. I am just putting it together for educational purposes and using a boost topology instead helps me cover more ground with only one project.

Here is my TODO list for the booster:

Current limiting

Synchronous rectification

High Temp shut down

The inverter does what its supposed to do. At the moment I just have to make some mods to get a more stable frequency. Right now its working at about 60+-0.05Hz if I remember correctly, not too bad.