12V pk-to-pk to 220V peak conversion

[samcheetah]

i have this freind who has a small problem in his project.

he has a 12V peak to peak output from an XR2206 IC. the maximum peak is 12V and the minimum peak is 0V and the average value is 6V. he doesnt have a negative supply as the whole thing is going to be powered from a 12V battery and thats why the signal has the whole swing in the positive side. and by the way the output is a sine wave.

now what he wants to do is to convert this signal to a 220V peak signal with average value at 0V. when i first heard this from him alot of ideas poured into my brain. but here comes the worst part. the output should have at least 16A driving capability. do you have any solutions

 

[Oznog]

Easy solution is to find a 220v to 6v transformer and use it backwards, provided the signal is 50-60 hz, with a large nonpolarized cap to filter out the DC component. But you won't get 16 amps from it unless the transformer and cap are huge! Do you really have such a powerful source? I mean, a battery or alternator does not put out the kind of voltage signal you describe. If this is coming off some kind of signal generator I would tap the actual power source on the other side. A 12v DC source would be easy, you'd just look for a 220v inverter.

 

[audioguru]

220V peak is 155.5V RMS. A 2489W audio power amplifier will do the trick to provide 16A at that high voltage. A 3kW into 8 ohms amp will provide 155.5V RMS at up to 19.3A and will get warm enough to keep you nice and warm in winter.

Oh yeah. A 3kW amp operating class-D with 85% efficiency will draw 288A from the 12V battery. A very large battery that will be discharged by 3kW very quickly.

If you want 220V RMS like the mains at 16A then the amplifier will need to be 6kW into 8 ohms, and will draw 580A (!) from the 12V battery. Wires as big as your wrist.

 

[samcheetah]

i think i misunderstood what he told me.

he told me that he has connected a power amplifier to the output of the XR2206 signal generator. and maybe he said that the input to the transformer should be at least 16A. because when the voltage will be stepped up, the current will be stepped down. but if the current is already 16A, stepping it down wouldnt reduce it much.

 

[audioguru]

Hi Sam,
See how confusing it gets when you talk about peak voltage instead of RMS, and current (without saying if it's input or output current) instead of number of Watts.
So your friend is using a power amplifier to drive a step-up transformer. Hopefully the amp has enough power to do the job, the amp is able to drive the very low impedance of the transformer and that the step-up ratio is correct.

Your friend's problem is that the amp isn't powerful enough. He judges its power based on the false misleading claims made in advertisments. An average current of 16A from an amp that is powered by a 13.8V car battery would require an amp rated at 61 real continuous RMS Watts into a 0.24 ohm load.

Most little car amps are advertised at 50W per channel. Their real continuous output power is only 14W per channel into a 4 ohm load.
Your friend should use a high power car amp with a step-up the voltage power supply in it. A class-D sub-woofer amp would do the job without heating too much. Of course, since its output voltage is much higher then the transformer's step-up ratio must be changed to match. :lol:

 

[samcheetah]

let me clear one more thing 8)

he isnt doing a car related audio amplifier project. he is making his own exchange.

now that thats cleared, ill contact him again and ask him what output current and output voltage he needs.

but anyway i want to ask you one thing. if lets suppose he wants 100V peak at the ouput and 2A current driving capability. then what ???

 

[audioguru]

Hi Sam,
A telephone exchange?
If he wants a 90V ringing generator, please don't say he wants a 127.26V peak generator.

In your example, 100V peak is 70.7V RMS. If 2A (peak or RMS?) is fed to the step-up transformer then you can't calculate the transformer ratio unless you know how much output power is required, or how much amplifier power is available into a specified impedance.

 

[samcheetah]

okay, now i know what he wants. he wants to build a UPS

now ill try to remember the signature of Styx

Nothing is impossible.
Once a problem is realised, the rest is just details

and the hardest part is realising the problem.

anyway, i remember a really good discussion on inverters some months back. can someone guide me to it.

 

[audioguru]

The most important thing about inverters and just about any power supply is: HOW MUCH POWER is required by the load and can the battery supply it for the length of time it is needed?

Is a sine-wave really necessary?

Huge heatsinks for convection cooling, a fan or class-D?

I think this thread started with a function generator feeding a power amplifier feeding a step-up transformer. That would make a very hot inverter. :lol:

 

[samcheetah]

i agree, i didnt like the idea that he gave. i havent seen inverters working like this.

actually in this case the sine wave is very very very very (and did i say very) important :lol:

the input is from a 12V car battery and the output required is 220V with maximum of 2A. and the frequency is 50Hz.

the reason he was using a function generator IC was that he had heard that inverters dont produced perfect sinewaves so he thought that using an XR2206 he could generate a perfect sinewave and then somehow amplify the current with an amplifier and then step up the voltage.

im sorry that this thread started from something else and is headed somewhere else.

you can see that the load isnt that big. the main thing is the perfect sinewave. can you give me a nice link?

 

[audioguru]

Hi Sam,
A 220VAC/2A inverter provides 440W.
A stepup transformer has some loss and you don't want the amplifier to clip when the battery voltage drops a little, so perhaps the transformer and amplifier should be rated for 500 real watts. They will be very big and expensive.
A standard amplifier is only about 60% efficient so will heat with about 200W and will need huge heatsinks and a fan.
The total of 500W to the transformer plus 200W of heat equals 700W.
With a 700W load, a 12V battery must supply 58.3A. Therefore a fully charged 100A/hr car battery will last only about 1.5 hours.

A 500W class-D amplifier is up to 90% efficient so will heat with only 50W and will use a reasonable-sized heatsink without a fan. The current from the battery will be 45.8A and it will last about 2 hours. :lol:

Will it do the trick? :?:

 

[samcheetah]

:lol:

leave the signal generator -> power amp -> transformer idea. that was when i didnt know he was making an inverter. now that i know what he wants, lets do it how it should be done.

 

[i_build_stuff]

That inverter will draw 36A from the 12V battery plus more to cover whatever is lost (36 + Ploss / 12)A.

I would use a full-bridge, possibly with a few parallel MOSFETs per switch, to step that up to about 350V, and then a PWM inverter to get the 220VAC out of it. The step-up will obviously be the hard part, as the current will be fairly high; the inverter section, at just 2 amps, won't have to be very special at all--probably just a pair of IGBTs wired to a basic PWM (it will have to have a little dead-time though). Gate drive on the high side might be a little tricky because of the widely variable duty cycle (you can't use a simple pulse transformer in this case). Carrier drive, or a small floating supply, would work.

If I might make a suggestion, though, does the battery side *have* to be 12V? The UPS systems I've seen used 4 smaller 12V batteries instead of one big one, to give 48VDC. If you do this, then your low-voltage side current is only 9 amps, which is much easier to switch.

 

[audioguru]

step that up to about 350V

You have some very good ideas, but the p-p voltage of 220VAC is 622V and the Mosfets will have some loss so you will need even more supply voltage for a PWM sine-wave. :lol:

 

[samcheetah]

that 4 battery idea looks good to me. that would also reduce the power lost, right??? so that seems like a really good idea. can you give me some link about the inverter you are talking about.

i have read some theory about full bridge configuration, and i know about PWM. but can you recommend me a good book for this. im getting interested in this, i think i should also do such a thing, that is if i get enough time :lol:

 

[i_build_stuff]

The inverter isn't really from any specific source--it's basically industry-standard stuff.

Texas Instruments has loads of switching power design information on their website, including all the papers presented at the annual Unitrode power supply design seminars. If you go to https://www.ti.com and search for "power supply design seminar book", a lot of great stuff will come up. Most of it is aimed at designing different types of DC supplies, and the topics every year are somewhat random, but there is an insane amount of info.

You have some very good ideas, but the p-p voltage of 220VAC is 622V and the Mosfets will have some loss so you will need even more supply voltage for a PWM sine-wave. Laughing

Since the output of the step-up supply will be floating (it has a transformer), the PWM section can be full-bridge and still have one output terminal grounded (needing 311VDC to run). This would reduce the necessary blocking voltage of the output transistors, which leads to lower channel resistance and higher eff.