Pure sine wave inverter

[Fused]

Hello everybody.
I bought a  24v 1500/3000 watt "pure sine wave inverter" from China. Not a good idea but it was all I could scrape together.
I have no scope to test it, to be sure it is actually pure sine wave.
After reading all the issues that can develop from modified sine wave Im concerned that I may not actually have a pure sine wave inverter.
I tested volts and get 115v ac at the 3 prong outlets. Is there any tests a person can do without special equipment to see if it actually puts out a clean sine wave? I cant afford to replace refrigerator compressor to find out its actually modified sine. Any series of tests that I can do to make myself comfortable with plugging in sensitive equipment? I know Im asking a lot, but we have some brilliant people on this board. Any help is appreciated.
I have a picture of the insides if any one interested.
Fused

[DanG]

Put a couple of clocks on it, old-style motorized and a newer digital LED display type on it for 24 hours and see if the times match.

[Fused]

Thanks Dan.
All my clocks are battery operated. Time to hit up the yard sales.

Cool idea. Ill try it.

Fused

[ghurd]

It is late and I am tired...

Does anyone think this is a good or bad idea?

Put the 120VAC output to a 400V bridge rectifier.
Hang a small 250 or 400V cap on the bridge output.
Check the cap voltage.

Seems like MSW would only reach about 140VDC, and PSW would reach about 165VDC.

Or maybe just use a meter.
PSW should read 110~125VAC output.
Many MSW inverters read 95~100VAC output, and I have tested them as low as 90VAC.
G-

[joestue]

i will reply again in the morning if that don't work.

just make sure you have a few watts of load on it, as the switching spikes may slowly charge up the cap.

[Fused]

I dont have those parts to test it that way Ghurd, but I do get 115v ac at the outlets.


Fused

[Fused]

Does looking at its guts help in deciding this?



Thanks everybody.

Fused

[dnix71]

If you have a small desk fan or tabletop oscillating fan, try starting it from a your sine inverter or the grid. Then try the unknown inverter.

Motors with fine windings like that do not like square waves. They'll hum and run slow. You hear and see the difference. That's why I don't even want a mod sine inverter, it's a gamble getting some things to run correctly. I also want idiot proof so if I leave something running it won't fail or catch fire while I'm at work.

[Fused]

These are the kind of tests I can handle. 

Still looking for that plug in clock.


Thanks

[BrianSmith]

You can probably find an old oscilloscope on fleabay for $50 with shipping ... maybe.  The other thing is to ask around at work, etc for someone with an oscilloscope laying around in the attic....

It would be good to measure the output with it loaded with a pretty good load to see how it looks as well as unloaded.

Good Luck

[OperaHouse]

That inductor on the right is a good indicator that it is sine wave.  Some of the other suggestions seem about as good as throwing chicken bones on the inverter and reading them.  MSW gets a bad rap.   I can't think of any reason that I would ever buy a sine wave inverter.   Those things with capacitor drops can be corrected for normal operation.   I do put RC networks on anything inductive to absorb the transients and prevent corona effects in the windings.

[ghurd]

I did the bridge and cap test, with a current limiting resistor.
Peak rectified DC voltage.
The grid and MSW both read about the same.
Interesting, but the MSW was actually a couple volts higher, so I retested.  2nd test was 1V lower.  3rd test was within a half a volt.
They are close enough to call them the same, about 165VDC.

The inverters open VAC is 96.5V.

If anyone does it themselves, they should use a current limiting resistor.
G-

[Fused]

Thanks OperaHouse.
Any body have a drawing of this?

Interesting stuff.

Thanks

Fused

[Xan]

I just plug my soldering gun into whatever I need to test. It's one of those types with the inbuilt transformer. Pull the trigger and hold it up near my ear. If it's a nice sine wave it'll be a clean hum. If it's modified, it'll make a raspy buzzing sound. Works well. Quick and dirty. Good idea to know where the business end is pointing though.

[Flux]

If you have a cheap multimeter and it reads 115v then it is fairly sinusoidal.  If you have access to a true rms meter and it reads 115v on that and the cheap average reading meter (scaled rms) then it will be sine. If there is a significant discrepancy between the cheap meter and a true rms then it will be modified square( modified sine if you must call it that),

The trick with the modified square inverters to make them work on most loads is to make the peak and the rms values near that of a sine inverter, if you use a square wave of 160v peak then it will run things like radios as the dc supplied to the radio circuit will be the same as peak rectified sine wave. Now this would burn out an incandescent light bulb as the rms would be 160v, but if you restrict the square wave to pulses of half width by delaying the start and finishing early, the rms value comes to about 115v.

If you measure ac with a true rms meter then it will be near 115v , if you rectify and measure peak it will be about 160v as for a true sine, but if you use a cheap ac multimeter on the ac range it will be way off ( probably nearer 90v).

The cheap ac meter will measure average, but be scaled to give the equivalent rms voltage on a true sine wave and on any other waveform it will be way out.

Another test is to feed the inverter via a low voltage transformer (say 2v) to a reasonable bass loudspeaker, use a pot if necessary to avoid overloading the speaker. The difference between a true sine and a modified sine will be very obvious. 

The snag here is that many cheap sine inverters are not filtered desperately well and you will also hear higher frequency ripple so just listen to the 60hz bass note. The modified wave inverter will sound as though the speaker is grossly over loaded no matter how much you reduce the input.

In the end there is still the element of gamble unless it is a true sine inverter with about 3 times the capability of starting the fridge motor. Many modified square inverters will run fridges ok as long as they have enough reserve to start them, although the motor may run a bit hotter. A true sine inverter will run it properly but unless it is way over size it may still fail to start it.

My experience is that washing machines ( speed controlled dc motor ones )and laptop power supplies are the ones that come to grief on modified wave inverters.

The other devices that won't work are drills and grinders and jigsaws fitted with a variable speed drive, if you have one of these and it runs then you can be fairly sure you have a sine inverter.

Flux

[Flux]

Reading all the other posts it looks as though I have summarized a lot of the ideas.

Ghurd hit the nail on the head in that a normal (not true rms meter) will read low on msw. With inverters tending to nearer 120v no load then 95v would immediately suggest msw.

Xan's transformer saturation noise is roughly equivalent to my loudspeaker idea, both need a certain measure of experience and are better in direct comparison unless you have experience.

Measuring peak or true rms is pointless, the things are designed to be equivalent in these quantities. I wouldn't buy a clock, some synchronous motor clocks and some digital ones will work on msw, depends entirely on the clock design, some won't run some run at harmonic frequencies and run fast.

Another factor is that generally magnetic ballast fluorescent lamps run badly on msw but again you need direct comparison as a big msw inverter will run it in some fashion.

If your meter is not true rms and it reads 115v then it should be sine.

Flux

[OperaHouse]

I have a collection of cheap dead inverters and I remember them all being around 140V DC.  That means the pulse width just needs to be a little wider to get the same power. I made note of this because many of them I use to power electronics directly with DC instead of repairing the H bridge.  So there can be some variance.  I have heard of people with specialty lamps that have electronic ballast which don't work well with MSW because the peak voltage is lower.

[Flux]

Getting away from the original question, but I have noticed 2 schemes used with these msw inverters, The first ones I met regulated the dc converter to 320v (UK 230v supply), then used a fixed pwm on the H bridge.

Most of the present ones seem to use a stiff but unregulated dc converter and use a variable pwm on the H bridge. These reach the 320v dc at about 14v battery in so at nominal battery volts the dc is low about  280 - 290v.

A friend has one of these inverters and his gas cooker igniter sparks fast when the batteries are floating and slowly when the batteries are low. The igniter spark rate must be related to the peak volts.

I think the change of scheme may be due to the problem of regulating the dc converter and maintaining stability on light or no load not an easy requirement.  These things are much of a compromise but work well enough for most loads.

The voltage regulation of many small capacitor excited generators is no better and some loads object to them.

Flux