making pure sine waves?

OK, I'm playing with building a battery back up and of course trying to do it as cheap and as difficult as humanly possible. A pure sine wave inverter is big bucks compared to a modified sine wave, so, can the modified wave be made pure?

Some time back Yanche wrote about cleaning up sine waves from a generator. Here's part of what he said: " You likely could clean up the generator output with some inductive filtering. Find two identical junk microwave ovens. Remove the transformers and wire them up secondary to secondary. The transformers will do the filtering..... (some deleted here).....If you try this be careful the voltages can be lethal!"

Basically he's saying wire 110V in from the generator ( or I hope inverter), wire the 2,300 volt of the first to the 2,300 volt of the second, and out of the second transformer you get clean 110V.

So I have a two part question... what if the transformers are not identical, but quite similar? And, will trying this clean up the sine waves from a cheap inverter?

If the transformers are not exact, might it just be that you'd have 110V go in and 114 volts come out?

Here's pics of the tranformers I found. One has 110V in and 2,300 out, the other though, has 110V in and only one wire for the 2,300 coming out, the other leg of the 2,300 goes to the case. Do I just connect one of the wires from the other transformer to the case? And...now the case is "OMG don't touch me!"? The white arrows point to the 2,300V's. There is another wire on the back side of the larger one that can't be seen in the pic.

Fred, the green wire on the unit to the right, is the 2nd wire. It should be grounded to the case, just like the ground wire on the unit on the left, where you have a white arrow pointing.

Let me explain it a little better. The one on the right has two tabs for the high voltage, one we see in the front, the other we can't see that's in the back. The one on the left has one high voltage shown with the arrow, but the other high voltage shown with the exclamation pointed arrow goes to ground. So, if I want to join the high on the left one to the high on the right one, do I ground one of the highs from the one on the right? Then attach the other two wires together? And doesn't that make the chassis high voltage? Or...should I remove the grounded high voltage from the left one and attach it to the unseen tab on the right one, then join the other two together? At least that would seemingly isolate the high voltage from the chassis,

Ok, I was confused with the first explanation. On both of them, I'd hook up the 110vac first, then use a voltmeter to see whats coming out the other side. Better to be safe than sorry.

The high voltage winding on many microwave transformers are referenced (connected) to ground. This is done because in many microwave designs the high voltage is referenced to chassis ground any way and this saves having to make the connection externally.

Freddy, here's how I would wire it. Use the heavier transformer as the input, i.e., the output of the UPS or generator. The high voltage wires then are connected. If one transformer does not have one high voltage lead connected to ground (chassis) at the transformer try to identify which one did go to chassis in the microwave internal wiring. Connect it to ground. Connect the high voltage lead of the input transformer to the high voltage lead of the second (output) transformer.

The reason you want to identify which high output lead went to ground is because depending on the transformer design there may be less quality insulation on one end of the high voltage winding than the other. Every trick to save a micro-penny. You don't need good insulation if it's connected to ground do you?

Be careful. Fuse the primary of the input transformer and be sure all grounds are connected to the safety ground (green wire) of the power cord. Don't attempt to measure the high voltage unless you have special high voltage probes that have arc suppression rings. If you are using an oscilloscope to look at the final output waveform make sure you realize you will be connecting one end of the 120 volt output to the chassis connection of the oscilloscope. Use a X100 attenuation scope probe. Make all connections first, including the oscilloscope leads before powering the transformer. Don't move the scope probe without turning power off first. Allow time for an stray capacitance to discharge after a turn off before touching anything. Be careful.

This is interesting ....

So what you're doing is connecting the 2 secondaries of the transformers together, generator in on one primary, then the other primary is pure sine wave? Do I have that right?

If so that would have saved me BUKOO BUCKS over buying that Yamaha EF6300iSDE .... that Visa bill is gonna be due soon..

Yanche, thanks for the info. I WILL be careful. Thanks for the fuse the incoming 110V tip. I do not have an oscilloscope so won't be trying to measure the high voltages. I'll just measure what comes out....and have that hooked up before I power up. My idea is to power up, look at volts, power down, come back the next day before touching anything. It may be next week before I get to it as I'm going to be gone this weekend.

Just to put it all in one sentence... one side of each high voltage coil will go to ground. The other high voltage wires get hooked to each other. Than, dirty 110 in, clean 110 out.

If it happens that 91V or 138V comes out I will have to hunt down another transformer that is a closer cousin to one of these.

Not having an oscilloscope my idea is to run a motor from the modified sine wave. The motor should growl at me. Then run the motor from the clean side of the transformers. If the motor sings happily I will know it's a clean sine wave. Ought to be a fun project.

Phase two of the project is to wire up the little relay that I got today so when the power fails, the inverter will kick in automatically. I'll charge the batteries manually. They are gel batteries and hold their charge a long, long time if left to sit.

Thanks!

Fred

Nice! I'll be watching your progress. I never thought of using a transformer in reverse -- makes total sense.

My china genset smoked 2 ballasts on me when one leg spiked up over 130V. I knew something bad was gonna happen when the lights got SUPER bright white! After reading about that, I think there was an internal neutral reference issue ... although the unit worked fine if you kept both legs balanced. If I drew too much from one leg, the other leg would hit some equipment-frying voltage levels!

Freddy, your "one sentence" is correct.

Just for everyone to understand what we are trying to do here. Inexpensive UPS systems and "inverter style generators" do not produce a pure sine wave output. They have a stepped staircase like waveform. There is a sharp change in voltage from one step to another. This can cause all sorts of harmful effects, particularly on sensitive electronics. It also harms motors. The harmful effect is because a square wave is really the sum of many, many sine waves. It turns out all the odd harmonics of the fundamental frequency. So a 60 Hz square wave is really, 60 + 60x3=180 + 60x5=3000 + 60x7=4200 + etc. ... Where each of the frequencies are a sine wave. The amplitude of each harmonic sine wave is lower than the previous frequency, but when all added up it becomes a square wave. Now these higher frequencies can cause all kinds of problems in motors. What you hear is a hum, but it's really a motor trying to run at two or more speeds at the same time. Can't do that, so it gets hot and left long enough it turns to toast.

The back to back transformer idea is a cheap way to add some inductance to filter out the higher frequencies. The transformers will see the higher frequencies and change the high frequencies into heat without any damage (hopefully). The output of the second transformer should be close to a clean sine wave. For this idea to work with higher power generators you will need some large transformers, the size in electric arc welders. It's not without risk. Those sharp spikes in the square wave can puncture the transformer's wire insulation. After all it just enamel varnish. You get a puncture and "puff" quick toast.

Always and education here.

Thanks

Yanche wrote:The back to back transformer idea is a cheap way to add some inductance to filter out the higher frequencies. The transformers will see the higher frequencies and change the high frequencies into heat without any damage (hopefully).

Yanche, all that heat comes at a price. Do you know what the typical Loss-rate would be for this setup? I'd guessitmate 10% ? I doubt Freedy's xformers could handle many amps, probably just enough to run the fans on a stoker? 500w or so? As you stated, much larger ones would be needed needed to filter a generator.
Freddy, why do you think a pure sinewave inverter is "So expensive"? I bought my 300/600 True Sine wave Inverter off of ebay for $103.
Maybe the danger aspect of hooking your xformers together is giving you the high BP

The amount of power loss would depend on the actual square wave input. If it were a 60 Hz fundamental square wave the loss would be high, perhaps the 10% you suggest. But most non-sine wave UPS systems are stepped square waves, so the loss should be a lot less.

On to Freddy ... Freddy is a tinkerer. He's my long distance technician, I make a suggestion, he tries it and reports back. Fun for both of us.

Freddy, get that BP under control. Decades of medical evidence show controlling BP increases life span. Get yourself a cuff type BP instrument and take your own BP several times a day. Keep a record. You will see what activities and foods change your BP.

It's not the cost of the inverter, it's the challenge of "do it as cheap and as difficult as humanly possible". The transformers were free. I do think I may need to find a different one though. To look at them one has many more windings than the other. They may not be close enough cousins to work. I'm wondering if I can do a lower voltage test.... Maybe input 24 volts & see what happens? But...all is on hold for now... headed out for the weekend. A 24 hour vacation. Yippeeee!

Yaa, I'm guessing this wouldn't push many amps. I'm thinking just the stoker and burner motors. Any loss? I wouldn''t care if it were 50% loss....it's just batteries charged from the windmill.

If it were me I'd just take 'em outside so I don't burn the house down, then run all kinds of voltage starting low, & working my way up -- just to see what's gonna happen. Could see how hot they get with an infared thermometer .... If they turn into toaster elements outside, no harm no foul!

Is there a cheaper device out there besides a scope that can measure if your actually getting 60Hz or not?

SMITTY wrote:Is there a cheaper device out there besides a scope that can measure if your actually getting 60Hz or not?

No, because what you are really interested in is the shape of the waveform not just it's frequency, i.e., is it a sine wave or not? Other instruments that measure the quality of the waveform (harmonic distortion, crest factor, etc.) are more expensive than an oscilloscope.