Connect with us

"Improving" a Modified Sine Wave

Discussion in 'Photovoltaics' started by Graham Parkinson, Dec 28, 2003.

Scroll to continue with content
  1. Has anyone experimented with RLC filtering the output from a simple modified
    sine wave inverter?

    I have to run a "computerized" Makita 24V drill "smart" charger from my
    "Statpower Prowatt cigarette lighter" style inverter and was wondering if
    the charger will not deal well with the stepped output.

    The only other option is to find a small inexpensive portable true sine wave

    (We are testing the concrete liner inside an 8km long water supply tunnel at
    several places - the tunnel is too long for an extension cord. I have 3
    drill batteries but will have a little "shopping cart" with some big 12V
    batteries for other testing gear and can run an inverter off them.

  2. Nelson

    Nelson Guest

    Graham, "Nothing 40"...
    I'm running a Trace DR inverter in a remote cottage location, and have a
    lot of trouble with the RF hash. Here's my experience.

    1) I put a cap, about .1 uf, across the output. It reduced the hash a
    bit, but not a long time later I had to replace the output FETs. There may
    be no cause/effect connection, but still I suspect the FETs were seeing a
    short across the cap for at least the leading edge of the square wave, and
    it may have toasted them without kicking a breaker. So I would definitely
    use an inductor input filter, trying it again.

    2) I put up a sizeable building using all screws last summer, and the
    main tool was a DeWalt rechargeable drill. It charged quickly and well, and
    never faltered.

    3) There's a 486 computer out at that little log cabin running off the
    inverter. Not a hickup.

    4) Finally resorted to setting up a remote antenna with a shielded cable
    back to the cabin for the radios. At least we can listen when the stations
    are on daytime power.

  3. Luc Collin

    Luc Collin Guest


    You will spend lots of time and money to develop a reliable LCR filter that
    may just end up frying your inverter because it can't handle inductive

    If I was in your shoes, I would just invest in a true sine inverter and
    protect my tools. A modified sine wave will be hard on your charger.

    But if you do insist on developing LCR filter for your inverter, get
    guidance from Xantrex. It will be hard to develop anything without knowing
    the output impedance profile of the inverter (both time and frequency

    Luc Collin
  4. Luc Collin

    Luc Collin Guest

    Guess work with filters will end up frying something.

    To develop a filter you will need to analyse the frequency profile of the
    inverter to figure out what needs to be filtered.

    Once this is known, you need to figure out which order of filter that needs
    to be used. The higher the order the lossier the filter will be, but the
    cleaner the signal.

    I would investigate first and second order filter. The problem with a first
    order will be a larger inductor that will be needed to handle the relatively
    large power in the harmonics plus the attenuation can only be -20db/decade
    (may not be enough to clip the primary harmonic).

    The second order filter needs to be looked at. The problem with a second
    order (and larger) are your zeros and poles in the frequency domain (points
    where frequency will have constructive cancellation) and may make the whole
    thing unstable and fry something. Added to this will be the added loss on an
    extra inductor plus misc. components.

    You may even need a stiffer filter than that. I would guess that the 120Hz
    or 240Hz harmonics will contain 90% of the power associated with harmonics.
    My guess is that you will need a filter that drop at least 80db at 80Hz or
    90Hz to effectively clip the first harmonic. *cringe*

    I will not bother disscussing the cost of equipment you will need to find to
    be able to develop you filter.

    IMO it is just not worth the time and effort to develop a filter unless you
    want to go in business with such a product. I am not convinced that it would
    give a competitive advantage over a true sine inverter.

    Good luck with your power supply, those are always fun to tinker around

  5. boB

    boB Guest

    If you made the DeWalt charger work with a modified square wave, then
    you did very good I would say ! Those usually break when plugged into
    one of those kind of inverters...

    Good luck on filtering though. You may be able to reduce the
    harmonics, but you will significantly decrease the power level of the
    inverter as well. It just doesn't work too well. We tried doing
    just that years ago while at Trace. Don't want to discourage you
    though. You just might make it work !
    Keep us posted !

  6. Jeff

    Jeff Guest

    The newer (really light weight) ones run off of a triac (as far as I know),
    which uses the firing angle with a AC sine wave to regulate voltage and
    current. A modified square wave is not a good wave to try and use a firing
    angle method of control. Older ones may have been transformer based, which
    would have little problem handling a modified sine wave)
    Having designed a small 350 W sine wave inverter, I'm also going to say good
    luck on filtering a square/modified sine wave. You can reduce the harmonics,
    but to filter it to a sine under all conditions at high power levels is a
    dream that's not likely to come true. To get a good, high efficiency sine
    wave, under changing conditions, you need to synthesize it with PWM (there
    are analog ways to do it, but at a large efficiency loss).

    It probably prefers a square/modified wave! It is a switch mode power supply
    after all!
  7. Nelson

    Nelson Guest

    Mind telling us what you tried? I'm not trying to make a sine wave out
    of a sow's ear, but I'd like at least to get the harmonics out of the
    broadcast band someday! The inverter works well in all other respects,
    even, to my amazement, variable speed drills and an old one-horse induction
    This is a Trace DR3624. The extent of the harmonics surprised me,
    especially since it seems (I could be wrong) to have a transformer output.
    I put an oscilloscope on it once, and the spike on the leading edge was
    several times taller than the waveform.
  8. Jeff

    Jeff Guest

    That does not seem right - maybe a snubber network inside the unit died?
  9. Thanks for all the intelligent commentary on the details of trapping the non
    sinuosoidal bits of "modified sine waves". I gather that it is not likely
    feasible to improve the output much.

    Of course the other option for charging cordless tools are the 12V
    chargers - I bought one that recharges my Bosch 24V tool batteries quite
    nicely off of 12V battery power.

    However I haven't seen them yet for the new Makita 24V batteries.

    Any reccomendations for a small (300W) 12V pure sinewave inverter? How
    about the bright yellow GO-Power SW300? This looks like a nice unit (priced
    about $229)

  10. George Ghio

    George Ghio Guest

    Get a battery drill of between 7 an d 12 volts. Chuck the battery. Stick
    a lead on it. Run off the 12V batteries you are taking with you on your

    My drills will run for two weeks or more on an 18 ah 12v battery.

  11. Or the probe compensation was not adjusted.

    When a measurement seems too far out of whack, don't forget to
    question the instrument.

  12. I don't have the original post so forgive me for commenting here.

    If this something like the switch-mode Makita charger I bought with my
    last drill, it will actually be much happier with this inverter than
    the old transformer-based chargers were. Just like typical computer
    power supplies etc. - not only do they not care about sine vs modified
    sine, they would be just as happy with 110VDC.

    I know the DeWalt's are notorious for going up in smoke though.

    Probably unnecessary.
    They're intended to take out high frequency interference, whereas the
    distortion components of a modified square wave inverter go way down
    to the third harmonic. They will do *something* but probably not what
    you'd hoped.
    I've often thought the same thing, but the only way I could see
    pulling it off would involve LC filtering and a step-up transformer,
    since any useful amount of filtering would reduce the RMS voltage
    If you value your time, you'd be better off to purchase!
    If you want to experiment, use two chokes, with one choke from each
    output terminal to one side of the load, and the cap directly across
    the load.
    Remove components of the input waveform and you reduce the RMS value
    of the output, yep.
    Although modified sine has significantly reduced harmonic distortion
    compared to a square wave, the first significant harmonic is the third
    (180Hz assuming 60Hz output) so I suppose the ideal filter would be a
    brick wall at around 100Hz. :)

    You can certainly take the edges off with a more realistic filter, but
    if you do enough filtering to be reasonably effective, you're going to
    run up against the voltage problem.

    Now, what I'd REALLY like to know is the nature of a Trace device that
    came and went several years ago, called the COSINE. I have it listed
    as a "new" item in a supplier catalogue from 1997. "Use it with
    modified sine wave inverters and get pure sine output," it says. No
    power rating quoted, around US$350 at the time, and about 10lbs.
    Intended for critical devices (like your filtered power bar idea) it
    did not claim to globally "convert" the output of the inverter.

    What could it have been but a whacking LC filter with a voltage boost?

    From time to time I've mentioned this gizmo but it's hard to find
    anyone who had even heard of the thing. Seems to me one fellow did
    perk up once and mentioned power handling in the 600W range, but that
    could be an invented memory.

    But these days you *can* get a 600W sine inverter for that kind of

  13. Ecnerwal

    Ecnerwal Guest

    I've often guessed that a big honking isolation transformer might do a
    good job of converting MSW to SW, but I'm a good 12 years or so past the
    last place where I had easy access to big honking isolation transformers
    and oscilloscopes without spending a good bit of my own money. I'd guess
    you'd want a transformer rated a good bit above where you'd be running
    it, as the MSW input would tend to heat the transformer more than clean
    SW input.

    Naturally, there would be the loss of the transformer to cope with,
    making a true SW inverter a better bet for systems which need minimal
    loss / maximal conservation.
  14. boB

    boB Guest

    Ahhh.. the old Co-Sine unit that Milt Rice (the engineer that
    designed way back when. Turns out it didn't get the go ahead because
    it was a bit too expensive to build. There were a few of them sent
    out for Beta testing as I remember.

    What it was, was a non-isolated sine wave inverter that got its power
    from the rectified output from a square wave, modified square wave,
    or sine wave 120 VAC output, and re-inverted that.
    It was a cool idea though. Milt was very smart and had lots of great

    Milt was one of the engineers that used to work at Dynamote,
    (Seattle), and designed their Brutus sine wave inverter. (bought by
    Vanner). Milt died in an experimental aircraft accident soon after
    the Co-sine was designed. He loved flying and experimenting with
    his Thunder Gull.

    Hope all had a great Christmas.
    Have a great New Year too !
  15. ptaylor

    ptaylor Guest

    Actually,most AT/ATX style PC power supplies have a cap right across the
    input jack,most of the time,I just junked a supply for parts,and snipped
    a .33uf cap off the back of the IEC (?) jack.Sometimes they are located
    on the board aswell,along with some other caps.I'm not sure about laptop
    supplies,but I'd assume the switching type (almost all are) would have
    this aswell.

    I wonder if it is hard on MSW inverters,interesting..
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day