Super simple -12v source for MBFM

[nickca]

Hi,

I've just completed my Midibox FM, and I have a working solution to the -12v problem that I don't think anyone's tried yet. My MBFM is totally minimal: no control surface, just a core, OPL3 board and bankstick. I happened to have an old laptop power supply that provides seperate, nicely regulated 5v and 12v, and I'm using it to power my MBFM. If your power supply only provides 12v and you have the full control surface, this might not work due to the additional load, but if you have a full control surface and can obtain a power supply like mine, it should work fine since all the extra load will be on the 5v. Anyway, here's the trick:

I used this circuit: 555 Timer negative voltage generator to invert the 12v to a little over -11v, which is fine for opamp purposes. There are many variations of this circuit floating around, I just used this one because I happened to have the parts in my junk box. Every different variation uses different resistor/capacitor variations, but they all work the same way: the 555 converts the DC to an AC wave, and the capacitor/diode network rectifies it to negative DC. The reason I say it might not work with a full control surface and a single 12v supply is the negative voltage decreases (gets closer to 0v) as you add load. Since the load in my minimal MBFM is, well, minimal, I get an output of pretty close to -12v. It works perfectly, and fits on a tiny scrap of perfboard. Much simpler and safer than mucking about with transformers.

If anyone else wants to give this a try, I'd be interested to hear how it works in an MBFM with a full control surface and a single 12VDC supply. It may still work fine, but with the LCD and LEDs, there's a lot of extra current being drawn, so I don't know. It's worth a shot anyway, it takes maybe 10 minutes to wire together out of less than a few bucks' worth of parts. Also, I used a CMOS 555 and 1N4148 signal diodes since it's what I had lying around. If you use an original NE555 and 1N400x (say, 1N4003) diodes, supposedly you'll get less voltage loss.

[stryd_one]

Welcome aboard nickca!

Thanks for sharing your hard work! Unfortunately, there are two problems which spring to mind, that such a design will introduce:

1) Noise from the switch mode PSU

2) A DC shift in the audio from the imbalanced bipolar voltages (mixdown nightmare)

That said this is a simple design... I don't know if I could say I'd recommend it though :(

[nickca]

Well, you could eliminate the imbalance by running the +/- voltages through positive and negative 9v regulators. If I understand correctly, the opamps just need a bipolar supply, not necessarily +/-12v. I actually was going to do that originally, but it worked without them so I omitted them to save space. As for PSU noise, I haven't noticed any.

[stryd_one]

The vregs might smooth out some of the noise too... Yeh, it's there.... Unless you got the first ever non-noisy switchmode I ever heard of ;D

[arumblack]

The reason I say it might not work with a full control surface and a single 12v supply is the negative voltage decreases (gets closer to 0v) as you add load.

Well, all the other CS stuff and LCD, etc... are powered from the 5V supply, so there should be no more load on the -12V line in a box with all the bells and whistles.

That still leaves potential switching noise and Imbalance problems of course.

AB