Improving modified sine wave inverter waveform.

[corvairhound]

I have searched, but maybe I didn't have the proper description. Looking to build a tank circuit to change the output of my 3000 watt Vector inverters into something more like a sine wave. Even if I need a seperate one for each circuit to match power factor. I have lots of coils and caps, even a 200 lb single to 3 phase converter "add a phase" and a few large battery chargers and welders.

I am afraid to hook anything with a capacitive input, thinking it may blow the finals, maybe an inductor inline, then a parallel lc tank to ring at 60 hz?

Need advice as to values and placement of varistors and such.Most of my equipment works as is, but some little  pumps and motors tend to buzz from the squarewave.

Anybody make one?? Thanks, Corvairhound.

[Flux]

Good luck it's not going to be easy. Unless the inverter protection is very good you may end up with smoke.

Motors may respond with a bit of series inductance and a bit of capacitance across them.They will be happy if you can get the current waveform more like a sine wave, voltage waveform doesn't bother them.

Resonant LC tanks will go crazy when not loaded and it is very difficult to tune these things when loaded. You may manage to tune them with a series resistance to limit the resonant voltage, but almost all loads will have reactive components that mess up the tuning when you connect them.

The very early stand by inverters were square wave with filters. Most used a resonant shunt tank at line frequency, a resonant series rejector at line frequency and 3rd and 5th harmonic traps. Never completely satisfactory or efficient.

I don't think you have a hope with a fit all filter but with care you may manage to tame individual loads. Let the motors growl if they are doing the job or look for a sine wave inverter is the best advice i can think of.

Flux

[boB]

Happy Gobble-Cluck !!

You might be able to round the corners a weeee bit and reduce some EMI by

filtering but that's probably about all you can accomplish with passive

filtering of modified sine waves.  (I don't know why anybody would want

to modify a perfectly good sinewave !)

We played with filtering a mod-square-wave inverter at Trace Engineering

years back and the consensus was that to get any useful filtering, the

insertion loss is way too high to give you any decent output voltage

after that filter... So, I doubt it will really be a way to go.

You could take the mod sine square wave output and full wave rectify it

to get new DC rails and then modulate that, sort-of, except then the peak

output voltage wouldn't be quite high enough to get 120V RMS without

going through ANOTHER transformer or autoformer.

I guess it's just better to start with a sinewave inverter and use that,

un-modified.

boB

         Happy Turkey Day !!

  (Ooops !  wrong turkey...)

[corvairhound]

I think you saved me a lot of time and effort. I didn't realize how hard it would be..had hoped I could just dump square wave into a tank circuit. and it would take the corners off.Anyone try a  "green plug" powerfactor controller on inverter output? Thanks.

[hydrosun]

Green plug into modified sine wave inverter made instant smoke.

Chris