Voltage doubler/increaser

[svstee]

OK, after dinking around with different voltages in a couple different pedals, I have come to the stunning and original revelation that starved plate tube pedals do better with higher voltage than 9v, and discrete JFET stuff sounds good juiced sometimes as well.

Anyway, I have a lot of stuff on my board that sounds good at 9v, what I want to do is build a pedal sized unit that has outputs for, ideally, 12, 18, and 27 volts. I can have one of each voltage and daisy chain the outputs to those pedals that need it. I'm not opposed to using transformers, but I would rather not.

Any suggestions, ideas, variations, or changes from my ideal are welcome, aka, if it is simple to build just a doubling type thing, tell me.

[Ripthorn]

Check out chips like the MAX1044.  Should be just what you need.

[R.G.]

Please, folks. Go read all of Geofex.

In this one case, the one that shows how to make voltage doublers, triplers, etc.

[Cliff Schecht]

You can use a CD4049, a couple of capacitors and a handful of diodes to do the trick. Make the CD4049 into a square wave oscillator and feed it into a %^&*croft-Walter type generator (as seen here): http://en.wikipedia.org/wiki/%^&*croft-Walton_generator

This exact circuit is used in the PAiA Tube Mic Pre to much success, but it's really only good for low current applications. Here's a schematic (upper-right hand corner):http://www.paia.com/prodimages/fantusch.gif

[frequencycentral]

Well, I like MAX, as you know. But last week Adriano (Renegadrian) posted this link in the Tube Boost/Overdrive thread: http://w1.859.telia.com/~u85920178/power/vconv_01.htm

[Valoosj]

http://www.diystompboxes.com/smfforum/index.php?topic=71657.0
That's what I got from Geofex. Don't know if it can drive a neovibe for instance.

[R.G.]

http://www.diystompboxes.com/smfforum/index.php?topic=71657.0
That's what I got from Geofex. Don't know if it can drive a neovibe for instance.

That's one of the things I was referring to. It probably will not drive a Neovibe, largely because of the high current use of the light bulb. One multiplier section sags with load as though it has an 80 ohm resistor in series with it. But you can parallel them, and in that case you could get more.

There are many other drivers for voltage multiplier diode-capacitor ladders. But losses get high when you get beyond a few stages.

Tube circuits are generally benign because a 12AX7 stage (for instance) needs only about 1ma or so per section. These low currents don't sag a multiplier power supply so badly.

[esdiezy28]

I just hopped on the Max1044 bandwagon, it works! I have a Valvecaster with 12V regulated for my heaters and 21V going to my plate resistors, opened up alot of headroom and definely added some more punch to the Valvecaster as opposed to when I ran the plates off of 12V. Our man "Frequency Central" has a charge pump design called "Murder One", I used Max1044 stage to get my extra voltage there. If I can get it to work after de-bugging a valvecaster for a month, anyone can. Best of luck!

[frequencycentral]

Our man "Frequency Central" has a charge pump design called "Murder One", I used Max1044 stage to get my extra voltage there.

Yeah - but I got the MAX idea from R.G. at GEOFEX - so all credit to him.

Murder One: http://www.diystompboxes.com/smfforum/index.php?topic=73222.0

[kurtlives]

A transformer(s) with a series of rectification, filtering and a series of regulators would seem like the thing to do.

A charge pump would be good, but aren't those MAX1044s limited to the current they can put out. I don't belive they could power up a tube's filaments.

[frequencycentral]

A transformer(s) with a series of rectification, filtering and a series of regulators would seem like the thing to do.

A charge pump would be good, but aren't those MAX1044s limited to the current they can put out. I don't belive they could power up a tube's filaments.

They put out 20ma. Enough for quite a few tube's plates. The heaters/filaments still just get 6.3 volts / 12.6 volts direct from the power supply - not the charge pump.

[petemoore]

  A Spyder type transformer with a couple or few higher than 9vac taps'd be cool !

[R.G.]

A charge pump would be good, but aren't those MAX1044s limited to the current they can put out. I don't belive they could power up a tube's filaments.

They can be paralleled to any output current; it just gets expensive and impractical after a while. The equivalent load on a 12AX7's filaments is 12.6V/.15A = 84 ohms. With a nominal 80 ohm impedance from one charge converter, you need to get to about 8 ohms to have the converters not halve the available voltage. That's at least ten converters in parallel, at about $3 each, $30. Cheaper to buy a transformer.

Anything can be done in many ways. For instance, gold can be refined from sea water. It just costs more than the gold is worth if you do it that way. Finding the most elegant and practical way is a journey of discovery.

[earthtonesaudio]

My one and only experiment involving voltage multiplication:

It worked, but was severely loaded down even by the impedance of the voltmeter, so I couldn't get an accurate idea of the output.

The one in Cliff's schematic looks like it could supply a LOT more current than the one I posted above. 

You can use a CD4049, a couple of capacitors and a handful of diodes to do the trick. Make the CD4049 into a square wave oscillator and feed it into a %^&*croft-Walter type generator (as seen here): http://en.wikipedia.org/wiki/%^&*croft-Walton_generator

This exact circuit is used in the PAiA Tube Mic Pre to much success, but it's really only good for low current applications. Here's a schematic (upper-right hand corner):http://www.paia.com/prodimages/fantusch.gif

[Cliff Schecht]

I'd think so as well. You could improve yours (or whomever that schematic belongs to) by using the classic astable multivibrator shown here:
http://www.davidbridgen.com/images/astable.gif

That Paia one can source enough current to power a 10 mA large diaphram condenser mic and both sides of that 12AX7, but the tubes are only biased at 12 V @ ~130 uA. I don't have a figure for the max amount of current that particular design can supply, but I can guess it's in the 20-30 mA range. Build it and test it yourselves!

[cpm]

Posted by: Cliff Schecht
You can use a CD4049, a couple of capacitors and a handful of diodes to do the trick. Make the CD4049 into a square wave oscillator and feed it into a %^&*croft-Walter type generator (as seen here): http://en.wikipedia.org/wiki/%^&*croft-Walton_generator

well, this censoring is plain stupid

[earthtonesaudio]

I almost spat beer out on the keyboard after reading that, thanks.