generate A/C from a D/C power supply?

[rodcastler]

Hi all, I need 9V AC at 1.5A to test an old C64 that runs on AC. I only have DC power on my bench and I was thinking of some sort of simple circuit made of tansistors driven by a function generator and get the required AC out of my DC bench supply. Is there a well-known way to achieve this?

Yes, I can easily get an AC transformer, and yes, I can potentially skip the recification on the machine and power it with DC. None of those options appeal to me because I don't get to learn what I'm abot to learn!

[electr_peter]

1) Get a sin wave for some sort of function generator. Build powerful buffer/amplifier from op-amps or transistors.
or
2) buy a lab AC power supply ($$$)
or
3) just use a transformer with a required voltage. There are myriad of older type AC/DC adapters with transformers. Some of those adapters are of AC/AC type - no modification is needed.

EDIT: Do you really need an AC supply? Some devices formally require AC supply, but use internal rectification anyway - supply them with DC and they won't complain.

[Skimask]

1) Find a 9v dc adapter
2) Remove rectifier
3) Profit???

[dannyf]

Is there a well-known way to achieve this?

Oscillator + a power amplifier.

[YU2]

Is there a well-known way to achieve this?

Oscillator + a power amplifier.

Can audio amplifiers be used for that?

[dannyf]

Sure, if the ac you desire is within the band of the audio amplifier.

[codeboy2k]

EDIT: Do you really need an AC supply? Some devices formally require AC supply, but use internal rectification anyway - supply them with DC and they won't complain.

Take a look at this thread. https://www.eevblog.com/forum/projects/commodore-64-replacement-psu-project/

Especially look at the schematic of the C64 just  inside the connector.  The C64 needs 5VDC and 9VAC, but the 9VAC is used to derive +12VDC and +5VDC internally. There's a part of the schematic there that's cut off, but I think it uses the AC frequency to feed the RTC. It might also use it to generate -12VDC somewhere on the board, but i read that this is only used for the RS232 ports.

for the 60Hz line frequency to feed the RTC, you can hack up an external 555 timer for that if you need to.

You can run the whole thing off DC, and you don't need to go to extremes generating AC for it.

EDIT: you might need to short pin 1 and 2 of the 7812 (remove the cap C90) and you don't need 9VDC UNREG if you don't use cassette tape.
EDIT2: there's some rewiring involved inside the unit, the point is you don't really need the AC at all.

[IanB]

Hi all, I need 9V AC at 1.5A to test an old C64 that runs on AC. I only have DC power on my bench and I was thinking of some sort of simple circuit made of tansistors driven by a function generator and get the required AC out of my DC bench supply. Is there a well-known way to achieve this?

Well now, 9 V at 1.5 A is about 14 W, and it corresponds to a load impedance of 6 ohms.

So if you have an audio amplifier that can output 20-25 W into 4 ohms you can feed it with a 50 Hz sine wave from a signal generator and turn up the volume until you get the 9 V output you need.

However, this is way more complicated than getting a 9 V transformer (or perhaps getting a 12 V transformer and adjusting the input voltage with a variac).

Overall, converting AC to DC is "simple". Converting DC to AC is "complicated".

[bugs]

http://www.banggood.com/buy/Class-d-amplifier.html
At 9v 1.5a sounds like you should be able to use a class D audio amplifier to have a reasonable efficiency.
Make sure you check for transients at startup and shutdown. 

[rodcastler]

At 9v 1.5a sounds like you should be able to use a class D audio amplifier

Very interesting. I didn't think of audio amplifiers. I assume these will perform only within the audible range. Will they work at 50Hz? and how far up would they go on the higher end?

Now you all made me think of this as a permanent source of amplified A/C for octopus component testing and such type of requirements.

And just being curious: How about square or triangle waves? Will a class D audio amplifier work with any shape of wave?

Thank you all for your replies!

[IanB]

Audio amplifiers are going to have a design bandwidth, for example 20 Hz to 20 kHz. Frequencies outside this range are going to get rolled off.

In particular, square and triangle waves will have frequency components outside the range of the amplifier, so they will not be preserved on output.

[torr032]

Audio amplifiers are going to have a design bandwidth, for example 20 Hz to 20 kHz. Frequencies outside this range are going to get rolled off.

In particular, square and triangle waves will have frequency components outside the range of the amplifier, so they will not be preserved on output.

This is really interesting, actually i never thought about it. Volume knob would be the voltage regulator, but how would I control the current flow?

[suicidaleggroll]

This is really interesting, actually i never thought about it. Volume knob would be the voltage regulator, but how would I control the current flow?

You don't.  You control the voltage, the load controls the current.  If you're looking for a current limiter like you'd find in a lab supply, there won't be one.

[Seekonk]

Do a search on ebay for self oscillating half bridge.  Think of a power 555.

[oldway]

Make a quase sine wave with two sincronised square wave inverters, one running at 3x mains frequency (180 or 150 Hz) and 3.2Vpeak and another running at mains frequency (60 or 50 Hz) and 9.6Vpeak.
The secundaries of the two transformers must be connected in serie, so you will have a quase sine wave of 12.8V peak and 9Vrms.
 

[bugs]

if you want to make a 50/60 Hz square/triangle/etc that should not be a problem. Remember that your favorite music may use them, you can usually hear the difference quite clearly :-)
Even the 20th harmonic is only around 1kHz.