Not quite one of those 'sine wave inverter' jobs...

[Blueteeth]

Hi,

I realise I post semi-random questions, rather than specifics about jobs, and this is no different, but I have googled this one and only come up with high power 'DC-AC', 12V to mains converters, often dangerous and poorly engineered ones .

I have a guitar multi effects that uses an AC 9V (2.2A) power supply brick (essentially, just a transformer and fuse in a box). Its rather big and heavy so as a challenge - as opposed to something that 'needs' to be done - I thought I would attempt to build my own converter allowing it to run off a DC input with a voltage range, so I can use any old SMPS wall-wart with sufficient current capability (and voltage output).

Now, often AC devices that use SMPS just rectify the input straight away, so will run off DC, or an ugly squarish wave at a higher frequency than mains. But this unit has an interesting power supply on board - it uses a voltage multiplier for ~60V for a vacuum tube, and capacitor/diode doublers going to voltage regulators, both come straight off the 9V AC input. So, DC is out (don't want to modify the internals). Given its an audio device, even with its internal filtering I'm cautious about audio-band noise being introduced to it, soo.. whilst I don't need a 'pure sine wave', something without massive harmonics would be nice.

Would a PIC loaded with a sine-wave table, using PWM to drive a H-bridge be sufficient? Its so much easier to just 'buy another transformer' but its the size and weight I'm trying to avoid. And because its only 9V (I assume thats RMS) I think I can get away without a transformer, just a large filter/common mode choke on the output to suppress the PWM switching frequency.

If anyone has actually built a DC-AC inverter (any power) I would appreciate your input. It does look like it can get incredibly complicated but given the low voltage, relatively low current, and slack sine-wave requirements, I'm hoping I could get something with parts I already have (swimming in parts here..). And then there is the question of regulation. Any idea's?

 

[audioguru]

Why does your effects box use such a high current? Does it have a heater or many high power lights in it?
Oh, it uses old fashioned vacuum tubes.

If the effects box used opamps then a little 9V battery could power it for weeks.

 

[Blueteeth]

Yo, well its digital, and quite old. I have a full schem which digitech provided, as its almost 20 years old I guess they don't care about IP anymore.

It does run a single preamp tube (filiment n all), main DSP processor (running at a shocking 5v) and a hell of a lot of analogue. In fact, much of the tone comes from the analogue, the digital part just provides delay/pitch-based effects. The manual says '12w' as the power consumption, so I imagine the PSU was just overspecced. I will have to measure the current draw over a few minutes of use to get the actual figure, but I'm thinking 1.5A tops. I only just dug this out after gigging with it for 7 years in my youth, and almost all of the switches/mechanics were replaced, but surprisingly the backup battery that stores settings is still going strong.

I was thinking something like this:
**broken link removed**

With a lower voltage supply of.. 1.414*9 + a bit = 15V. I honestly do not know what (if any) feedback I would need, as it'll be direct drive. That could of course be a bad thing, as isolation is always handy - both terminals of the input have 47n caps to the chasis, which is connected to the system ground via 0-ohm resistors. I haven't found many small 1:1 transformers rated for 50Hz

As I said, its probably just not worth it, as mains transformers are fairly ubiquitous still. Just trying to 'standardise' power requirements for guitar equipment mostly. Thought it would be an interesting project though!