12v DC to 12v AC 50Hz

[lw395]

Ok,bit late to this game but...

OI you lot, quit harping on about 240V (yeah, I know its easy... lazy buggers). 12V 50Hz is audio!. Think sine wave generator and audio amp.

and another thing... without upping the volts, you cannot make 12Vac from 12Vdc! (its an RMS thing, innit?) That may well be why the lamp barfed at 12Vdc.
So, you need a 12V -> 24V buck converter and the audio gubbins run from that, all done as jdb says "dead bug style"

edit add: use 24V as these buck converters are common vehicle kit.

Nearly right.
12 to 24 would be a boost converter.
Does the 12V ac need to be isolated? does it need to be referenced to the DC 0V?

You can make 12V AC from 12V DC using a bridge audio amplifier. But the output will be two 6V sine waves of opposite phase.
If you need DC isolation, you can put capacitors in series but they will be big.
Or you could use a transformer to isolate.

 

[jdc]

Ok,bit late to this game but...

OI you lot, quit harping on about 240V (yeah, I know its easy... lazy buggers). 12V 50Hz is audio!. Think sine wave generator and audio amp.

and another thing... without upping the volts, you cannot make 12Vac from 12Vdc! (its an RMS thing, innit?) That may well be why the lamp barfed at 12Vdc.
So, you need a 12V -> 24V buck converter and the audio gubbins run from that, all done as jdb says "dead bug style"

edit add: use 24V as these buck converters are common vehicle kit.

I agree that converting to mains will be much more complex than need be. Re: the rms thing, a sq wave does have rms = peak. But I doubt that rms is relevant; LEDs have a forward Voltage of about 3V, and take DC, so threse's got to be some converter and current limiting circuit. This won't be like a heater, ie dependent on power alone. It would be good to get a spec, but I'd certainly try a sq wave first.

 

[JohnGC]

1. How clean a sinewave must the 12V AC be? Could it be a square wave (lots of odd harmonics, but depending on the application may not be an issue).
2. What frequency? Although they say 50Hz, isn't always always essential. Have you any idea why it needs to be 50Hz?

Of course you may not have a spec, but it's possible that the light bulb, if a halogen, is not at all fussy. I have some halogens in the bathromm and kitchen that run off 12V AC, but it's from a small 'transformer' box which actually delivers a square-wave at several kHz.

PS: Did you try DC? Not a daft question: it's hard to envisage an electronic design for a bulb which would mind. A 50Hz transformer is a big and heavy thing, so almost nothing these days uses one. Instead almost all devices convert to DC immediately on input, and then convert again, if need be, with a 'switch-mode' circuit running at some tens or even hundreds of kHz. These circuits will accept sq waves up to some kHz, and in fact many will accept DC at their inputs.

It sounds as though the LED lamp uses a reactive element (capacitor) to limit the current rather than a resistor. Reactance of a capacitor is inversely proportional to frequency; so if you use a higher frequency the voltage will need reducing for the same current.

 

[Lon nan Gruagach]

Nearly right.
12 to 24 would be a boost converter.
Does the 12V ac need to be isolated? does it need to be referenced to the DC 0V?

You can make 12V AC from 12V DC using a bridge audio amplifier. But the output will be two 6V sine waves of opposite phase.
If you need DC isolation, you can put capacitors in series but they will be big.
Or you could use a transformer to isolate.

That'd be 12V peak, not rms as is usually spec'd for sine wave

I agree that converting to mains will be much more complex than need be. Re: the rms thing, a sq wave does have rms = peak. But I doubt that rms is relevant; LEDs have a forward Voltage of about 3V, and take DC, so threse's got to be some converter and current limiting circuit. This won't be like a heater, ie dependent on power alone. It would be good to get a spec, but I'd certainly try a sq wave first.

My bet (and its a bloody good one) is that the lamp (half wave)rectifies the incoming and charges a big cap.... giving ~14 and a bit volts. with dc the best it can get is 11.3... not going to cut the mustard. Although the "dc makes it flicker and die" ends to imply capacitative coupling... mebby theres a charge pump in there in which case square wave will do and any frequency will do too(ish)