Hi Peter.

Let's assume that PWM frequency is 256 times by 60 Hz which is 15.36 KHz
PWM switches simply add and/or deduct more current (different polarity) into
primary windings. ( Like approximating curve with very short lines method )

What kind of load dump condition we are talking about in this example ?

There is no abrupt current changes in primary wirings on frequency close
to 60 Hz. ( since PWM frequency is much higher and it assumed to be filtered
by inductance to smoothly changed sine shape of AC current. )

What is wrong with such approach ? Effectiveness ?
( There is feedback also that watches load condition and keeping
output AC wave in certain amplitude and frequency range. )


WBR Dmitry.


> *>What is the most difficult part with PWM type of DC-AC with transformer
> *>converter design? Either regular PWM or some sort of magic sine waves.
> *>(minimal efforts to receive optimal output sine wave with minimal
> *>distortions)
>
> The load dump condition. Assume a load is attached and suddenly turns off,
> f.ex. at voltage zero crossing. If the UPS reactive elements contain X
> Joules and are not unity damped these X Joules will become voltage that
> keeps rising until it finds a place to put some current through. The usual
> PI filter, VDRs, spark arrestors, and fuses at the input of any circuit
> connected to mains are there to prevent this, and sometimes succeed. Now
> if you drive a transformer with relatively clean sine (class B or better)
> there will be no load dump (well, a little) even if there is no feedback
> loop since the transformer output impedance is hard-coupled to the driver
> stage. The voltage will not rise (well, a little). If the transformer is
> driven with a switch then even if there is feedback the energy in the
> transformer will become a fearsome spike and thefeedback circuit can do
> nothing to stop it. There are several ways to deal with this (like dynamic
> crowbars etc) but they add complications to the high voltage side.
>
> *>Negative feedback in addition thru say some opto couple to make sure that
> *>output AC signal is sine and required voltage and frequency.
> *>
> *>Probably main problem is to drive high current into primary windings
> *>but with on/off type control it will actually save you on heatsink
> *>and will increase? converter efficience as well.
> *>
> *>Anybody was successfull with doing this type of design with PIC ?
> *>(different than class B)
>
> Not with PIC, with logic circuits (divide by 4 and gates, driving
> 00-nothing 01-one phase 10-nothing 11-other phase), after a scheme from a
> book. The secondary had to be damped with a string of neon lamps to
> prevent it from doing stupid things when not loaded. It had a capacitor
> across it. The voltages that appear when undamped are quite enough to
> break down the insulation of just about any transfomer and will arc over
> 5mm spark gaps easily.
>
> Peter
>
> --
> http://www.piclist.com hint: The list server can filter out subtopics
> (like ads or off topics) for you. See http://www.piclist.com/#topics

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads