Author Topic: cleaning up modified sine wave  (Read 28607 times)

0 Members and 1 Guest are viewing this topic.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
cleaning up modified sine wave
« on: March 25, 2013, 01:57:25 PM »
Has anyone attempted to clean up a modified sine wave inverter output, and with what results?

Some of the inverters I have put out a waveform that does not resemble in any way a sine wave, except that the fundamental freq is 60hz. Some have a square wave output with about 50% duty cycle, and others' output has so many jagged edges and overshoots that they don't resemble a square wave.

I've looked at RLC shunt line filters (to get rid of the harmonics and noise above 60hz), DC blockers (to set the avg AC output to 0), RMS voltage regulators (to ensure the output is 120 Vrms, or about +/- 170 Vpp), but I haven't tried any of these yet.

Also, I am trying to estimate if the cost of cleaning up the modified sign wave is not going to exceed the cost of buying a true sign wave inverter (except that I already have quite a few of those on hand, and need to put them to some use that needs relatively clean power).

dnix71

  • SuperHero Member
  • ******
  • Posts: 2513
Re: cleaning up modified sine wave
« Reply #1 on: March 25, 2013, 02:49:44 PM »
I have the feeling that if you could reshape the wave you would ruin the PF. I don't see how you could run any RC circuit like that that wouldn't throw off the current angle.

Sine wave inverters nowhere near as cheap as square wave, but they really don't cost that much if you consider the replacement value of sensitive equipment. I won't even own a mod sine inverter. There is no sure way to know if it matters until you plug something in. The voltage peaks on one I owned were 145 or so instead of 171. I borrowed a scope and was suprised just how nasty and square it was.

ChrisOlson

  • SuperHero Member
  • ******
  • Posts: 3642
  • Country: us
Re: cleaning up modified sine wave
« Reply #2 on: March 25, 2013, 03:03:42 PM »
Has anyone attempted to clean up a modified sine wave inverter output, and with what results?

Some of the inverters I have put out a waveform that does not resemble in any way a sine wave, except that the fundamental freq is 60hz.

Either an auto transformer or an induction motor load both do a very respectable job of cleaning up a MSW.  Using an auto transformer to do it, I doubt you could tell the difference between the transformer output and a real pure sine wave.  However, the transformer primary winding will buzz quite loudly on MSW input - AND you will get 120/240 split phase output from an auto transformer if you use a 120V inverter on it, with zero leg balancing problems L1 to N or L2 to N.
--
Chris

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: cleaning up modified sine wave
« Reply #3 on: March 25, 2013, 03:49:39 PM »
thanks for your comments,

I need to look into this auto transformer concept, and also see how the power factor would be affected.

I'll post the results here when I get them.

Also, I wanted to attach some nice MSW inverter oscilloscope pictures for you to see, but got an error on this server for my attachments exceeding 100kB in size (my pics are about 600Kb/picture). They compare MSW output to the superimposed clean 120VACrms grid output. If anyone wants to see them, email me at endeavsys@yahoo.com.

Jack

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: cleaning up modified sine wave
« Reply #4 on: March 25, 2013, 04:14:59 PM »
This is a picture I've had for about four years.   It is the "filter" I use to power chargers etc.  asically it is a RC network two 5 ohm resistors and about .7uF polypropylene MPX capacitor.  That feeds into a corecom like LC network.  This operates on my chest fridge which has its own MSW inverter that turns on and off constantly and is a 120 foot run.  The RC network absorbs spikes from motor shutoff and dampens the line from the square wave.  Mostly just a power strip but I am a strong believer in RC networks that i even put in sine wave equipment.  Another use for a PC power supply case.

Anyway, I see no problem with MSW.  Many electronics probably work better with it.

DamonHD

  • Administrator
  • Super Hero Member Plus
  • *****
  • Posts: 4125
  • Country: gb
    • Earth Notes
Re: cleaning up modified sine wave
« Reply #5 on: March 25, 2013, 06:39:04 PM »
@OH: would that count as a snubber network (in which case it's nice for zapping RFI/spikes anyway), or is your thing meatier?

Rgds

Damon
Podcast: https://www.earth.org.uk/SECTION_podcast.html

@DamonHD@mastodon.social

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: cleaning up modified sine wave
« Reply #6 on: March 26, 2013, 12:20:28 AM »
i would take a look at the snubbers on the primary side of the transformer as well, and how the mosfets are driven.

i've seen some inverters where they drive the fets directly from an lm358 opamp.. lmao, and the snubber is a diode, electrolytic cap, and a 5 watt resistor on each side of the push-pull inverter.
the electrolytic caps dry out, the fets light on fire when you drive a motor or other inductive load (and don't even think about driving a heavilly capacitive load)
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: cleaning up modified sine wave
« Reply #7 on: March 26, 2013, 04:12:09 AM »
Most MSW inverters use a H bridge output.   Internally to each FET is a protective zener that is used to snub a reflective spike back to the + or - supply rail.  FET manufacturers recomend using an external diode if the circuit has to absorb a lot of spikes otherwise derate the FET power.  Inverter mfg get by without the external diode but I believe this is the cause of some inverter failures.

I used to work for a company that manufactured among other things RC networks ,  I worked with a lot of customers solving their problems with these devices, some of them very custom.  Problems started to abound with hte use of inverter drives and the demise of the light bulb.  They were throwing a lot of noise on the power line and light bulbs that provided resistive dampening for the noise were dissipearing. 

Motor and inverter manufacturers don't like it when the distance between the two goes over 100 feet.  This is when standing waves can appear, stacking up voltage and breaking down motor insulation.  Motor inverters likely have faster rise times than 60hz inverters and there are more of them with chop rates of 10K or more.  Still, the principle is the same, a house has long lengths of wire.  Send out a pulse and it will reflect back.  I have a TDR that works on this principle.   If you have a 1,000 foot burried cable and it shorts or breaks someplaqce I can tell you where to dig within five feet by looking at the reflective wave.

All capacitors are not the same.   I said I used MPX polypropylene capacitors.  These have low internal resistance and good frequency characteristics.  Polyester makes cheap suits and even worse capacitors.  Too bad they are the most common capacitor material.  The dissipation characteristics of these are bad at even audio frequencies.   We sold three phase RC suppressors to control arcing when a contactor opened.  We had customers with an inverter for another motor on the same power line.  They would call up and tell us the boxes were getting hot.  The capacitors were gettinjg hot not the resistors.  That is what can happen with polyester. 

So, that is why I always put a RC network on inverters.  Maybe it does something.  First thing to try if you have noise,  Just use a good cap.


« Last Edit: March 26, 2013, 04:32:24 AM by OperaHouse »

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: cleaning up modified sine wave
« Reply #8 on: March 26, 2013, 01:36:43 PM »
For any of these solutions (auto-transformer, RC networks, etc) it would be nice to see some sort of before-and-after performance measure. For example, oscilloscope pictures of distorted waveforms before the solution is applied, and cleaned up waveforms after the solution is applied, if you can provide any. This would give us incentive to spend time and $$$ on that solution, or not.

Regarding the autotransformer, Variac etc make some units, seem to be optimized for power line conditioning, also capable of adjusting the output voltage.

Without going too deep into theory, the MSW smoothing effect seems to be the low-pass inductor/winding characteristic, which here may be able to filter our strong MSW harmonics and noise beyond 60 hz.

The fact that there is no isolation between input and output doesn't seem to matter, because I don't have any isolation to begin with.

A failure in the winding insulation should only have impact on the inverter, and would not endanger the load because I am using the auto-transformer as slight step-up and not step-down.

Also good benefit, if I have several taps in the secondary (or continuously variable secondary), then I can use it to regulate upward the RMS output voltage to the desired value with no-load, or make it load-dependent.

One thing I don't yet fully understand, there may be a phase difference between the voltage and the current at the auto-transformer output that powers the load (just like in any inductor with some resistance in the winding). This degraded power factor would cause reactive power to develop in the auto-transformer winding, which would be sent back to the inverter and not to the load. How much power loss would there be, and what this would do to the inverter, I don't know.
Seems like we are dealing here with a case where the power source itself, and not the load, is causing the reactive power component. Perhaps some capacitive power factor correction would be needed at the auto-transformer output to take care of this, and make this solution usable ??? Any comments?

Regarding the RC networks, I assume this would be R=5 ohm and C=0.7 uF elements connected in series. Then I am not exactly sure, two RC series circuits would be connected from each of the hot and neutral outputs of the MSW inverter to ground, or one RC series circuit connected between the hot and the neutral?
Which way is it?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: cleaning up modified sine wave
« Reply #9 on: March 26, 2013, 02:14:06 PM »
an auto transformer isn't going to clean up the output, only make it look more clean.
Ideally a transformer stores no reactive energy at all, so when we're talking about 500-2Kw isolation transformers, what is really happening is when you load the inverter through the transformer, at the full rated load of the transformer, the transformer appears not as a TX but rather a 5% reactor. (plus a 5% resistor) and the capacitance of the coils is not significant. (it is probably the iron loss of the transformer that is actually cleaning up the waveform--by absorbing the harmonics)

if you want to put an LC filter on the output, go right ahead, but to get any descent voltage regulation out of it, you will have to build resonant traps for each of the harmonics.. and the harmonic power generated by the inverter--180Hz and above-- is going to be recycled as conduction losses.

in theory, you don't need snubbers on full bridge or half bridge inverters. the mosfet's DV/DT is not what causes it to fail.*
what causes MSW inverters to blow up is improper circuit design.. and the Leakage inductance of the transformer when the inverter is configured as a push-pull inverter --(two mosfets source tied to ground, both drains go to opposing sides of a center tapped transformer)

this is how most <1KW MSW inverters are configured.. because its cheaper for some reason... even though, you need 71% less copper on the low voltage side of the transformer to configure it as a half bridge or full bridge.

voltage source inverters require low inductance power supplies. batteries do not count as low inductance.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: cleaning up modified sine wave
« Reply #10 on: March 26, 2013, 04:53:34 PM »
So I fired up an inverter and put it through a transformer and as I expected a perfect staircase.  A 5 ohm and 1uf BOSCH HSFM capacitor with ESR of .61 accross does nothing to change the waveform except getting rid of a little transition glitch.  Added power consumption was 1.9W from 12V current draw.  I had a IC MWR series 4.7uF with ESR of .22 and  that bairly drew any extra current and just buzzed.  Proves my point about capacitors.  Anyway no pictures.  This was an open converter out of the case and had things electrically isolated with a sheet of paper.  Happend to bump it with an elbow and shorted something. Dead.  Now I have something to work on.
« Last Edit: March 26, 2013, 05:02:25 PM by OperaHouse »

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: cleaning up modified sine wave
« Reply #11 on: March 26, 2013, 08:38:27 PM »
there is a resonant push-pull leakage inductance snubber shown in this document:
http://homepages.eee.strath.ac.uk/~bwwilliams/Book/Chapter%209old.pdf
not sure if this link works without registration.
http://project.eee.strath.ac.uk/textbook/website/chapters/Chapter%209.pdf
page 15.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: cleaning up modified sine wave
« Reply #12 on: March 27, 2013, 08:19:39 AM »
And Chinese Apricots are going to be a glut on the market this year.

ChrisOlson

  • SuperHero Member
  • ******
  • Posts: 3642
  • Country: us
Re: cleaning up modified sine wave
« Reply #13 on: March 27, 2013, 09:14:26 AM »
So I fired up an inverter and put it through a transformer and as I expected a perfect staircase.

When we had our SW Plus inverter with the 6 kVA auto transformer on it for step-up 240V split phase I had a scope on the setup some time back.  The SW Plus is a pure sine wave inverter and it forms its sine wave with 27 to 52 steps, depending on voltage and load.  The roughness in the sine wave is visible on the scope.

The 180 degree out-of-phase sine wave from the auto transformer L4 output is perfect, OTOH.
--
Chris

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: cleaning up modified sine wave
« Reply #14 on: March 27, 2013, 09:15:48 AM »
Modified sine invertyers are really modified square wave and it is far from easy to clean up the weaveform into anything resembling a sine wave and still keep something like the original rating.

You may be able to make some attachment to clean things up enough to work some of those critical chargers for power tools etc but even then it is not easy to do much more than round off the sharp corners.


Those modified sine inverters are intollerant of capacitive loads so watch your step adding shunt capacitors.

Joestue is absolutely right, we built early no break inverters using thyristors with quasi square output. The filter circuit was horrendous with a resonant tank circuit, series acceptor at fundamental and 3rd and 5th harmonic traps. It worked because the current fed thyristor inverters were near indistructable and were happy with reactive loads.

Trying to adapt this to a mosfet inverter fed from a capacitive source would probably be impossible.

Basically modified sine works for most things well enough, for the applications where it doesn't work it would be better to start with a sine inverter in the first place.

Early sine inverters used ferro resonant transformers like the constant voltage transformer and it worked well enough but the wound components cost too much. Don't try to use the ferro resonant devices to clean up a modified sine H bridge, it's a different technology.

Flux

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: cleaning up modified sine wave
« Reply #15 on: March 30, 2013, 12:00:48 PM »
Well, it looks like I have some options lined up, after this fruitful discussion.

I'll probably play with the auto-transformer and the RC networks at some time in the future.
Will post the results/pictures here if/when I get them.

thanks, Jack