Voltage doubler circuits (not charge pump ICs) suprise... advice?

[Morocotopo]

Hi people, i´m making a chip based amp, already have the pre, I´m using a LM1875 for power. Already fired it up, works good, I have a 24V 1A transformer, but I´m not satisfied with the output power, so I´m thinking about testing it with at the maximum voltage the chip can take (60V), to see if that increases output power. Getting another transformer means more money to spend, and also the one I have barely gets warm, so I thought "let´s use a voltage doubler, I´ll get 48V at 500 mA" How intelligent I am...
So I tried this:

Works, BUT...
with 4.7 uF 63V non polarized caps and 1N4007 diodes, I get 66V DC!!! What the...?
When I tried 1 uF caps, I got 32V. Anything higher than that gets 66V. The transformer puts out 23 Volts AC. The V output is not connected to anything.
What am I doing wrong? Will the voltage change when I connect it to a load (amp)? Also, how do you choose the caps and diodes values? Any formulas? If anyone is familiar with this type of circuit and can give me a hand, I´ll appreciate it.

Thanks

[John Lyons]

The load will drop down the voltage most likely.

john

[Paul Perry (Frostwave)]

Yeah, the 'peak' of the sine wave from a 24v transformer is 24 x 1.4, that's about 33V.. double it & there is your 66V. 

[Ice-9]

I've also been looking at getting a higher voltage to drive a tube preamp. does anyone have any ideas on how to get 200 - 300 volts from a 16vac power adapter.

I have a blackstar pedal that runs from a 16vac power adapter and claims to get 300v to the valve circuits. I don't really want to strip the pedal down to find out how its done so so some pointers in the right direction would be great, I don't want to hi-jack this thread (think the topic fits this thread) Cheers

[R.G.]

Do a Google search on "%^&*croft-Walton".

[Ice-9]

Thanks R.G. I have seen read about the "%^&*croft-Walton" circuit but the problem here is the component count of capacitors and diodes i would need for each stage. I want to be able to fit the entire circuit into a pedal. i will take this blackstar pedal to bits and see if i can figure out how they have done it. other ideas on voltage multipliers would be very welcome.

[slacker]

You can do this using transformers, you use say a 240 -> 12 volt transformer to get 12 volts AC then you feed this into the 12 volt side of another similar transformer and you get ~240volts AC out of the side that would normally connect to the mains. Rectify this and that will get you in the region of 300volts DC.

The Real McTube pedal does this, unfortunately the only schematic I could find is this one from Paul Marossy's site which is a word document http://www.diyguitarist.com/Schematics/RealMcTubeII.doc

I think you should find some more info if you search for something like "back to back transformers".

[Gus]

Ice-9

  Sometimes people use a transformer.  Google Mctube.

EDIT must have been posting at the same time.

[earthtonesaudio]

%^&*roft-Walton multipliers are very cool but they discharge very quickly.  You'd need a mammoth storage cap to keep the output voltage constant under load... Unless you find the "dynamic voltage sag" thing appealing.  I think I would.   

[R.G.]

Thanks R.G. I have seen read about the "%^&*croft-Walton" circuit but the problem here is the component count of capacitors and diodes i would need for each stage. I want to be able to fit the entire circuit into a pedal. i will take this blackstar pedal to bits and see if i can figure out how they have done it. other ideas on voltage multipliers would be very welcome.

OK. Google "Flyback switching power supply".

You can up-vert to any practical degree if you load up an inductor and then let it discharge into a diode/cap/load. The old Civil Defense radiation meters up-verted two D cells (3V) into 300V for the ionization chamber this way. Camera flashes work this way.

And frankly, simply deciding that something has to fit in a previously-decided size of pedal is a recipe for some effort at least, and some misery at worst. The effort needed to put things in a fixed sized enclosure goes up exponentially as you approach completely full. Then making it WORK correctly when it's stuffed into a fixed size box goes up exponentially too. Tube grids are notorious for picking up hum and stray capacitance feedback.

There is no harm in setting yourself a task to be done, even if it's only barely possible. But ask first if what you want is the exercise or the finished goods to work well and easily, and how much of your time and effort you're willing to spend to get that.

[Morocotopo]

Oh, I see Paul, this circuit actually has a (half wave?) rectifier in it, so that´s where the extra V comes from. Thanks.
What do you think, will V be lower when I connect the load? I don´t want to fry my amp... (chip´s max V= 60V)
What cap values are recommnded? and what Diodes?

Thanks all

[Minion]

The Voltage doubler won"t work very well for an powering an amp...The amp draws too much current for the doubler to handle , The voltage will sag really badly even with say 10,000uF caps ,and there will be a lot of ripple on the supply lines and you will probably get a bad hum...also 1n400X series diodes aren"t suitable either as they have a max current of 1a and that amp will draw more than 1a of current....

Voltage doublers/triplers are really only good for low current applications , I have used them to produce 48v DC  from 15v AC for produceing phantom power for mic preamps ,and i had to use 3 x 1000uF caps to get the ripple and noise down to an acceptable level and phantom power only draws about 14ma per channel so trying to suck a couple amps of current from it would be next to impossible.....


I recomend just getting a higher voltage transformer or bridgeing two lm1875 chips .....


Cheers

[Ice-9]

Thanks for the info all
@RG i checked up on flyback psu and also took the blackstar pedal to bits and i'm sure thats what it is using. (ua3843 chip and mosfet) After abit of reading this flyback circuit can be simplified with what i believe to be called TOPSwitch flyback. I must do a lot of reading on this as my psu knowledge is pretty limited. I will do some research and may start a new thread as this is a bit off topic for so i apologise to  Morocotopo who started this thread.

[Morocotopo]

Ice-9, no need to apologise, sometimes topics shoot in a different direction than originally intended. That´s cool, usually those are the most interesting ones to read.
Minion: so you think it won´t work? Well, I have a plan B! see my topic on combining transformers. 
Might try it nevertheless, just to see what happens. Thanks.

[Minion]

You don"t even need to combine the transformers....If you have 2 identical 24v transformers you can use one of them for the Positive supply and one fore the negitive supply....I believe your LM1875 is set up for a single supply but converting it to a dual supply would be fairly easy and advantagous as you can omit the output cap on the LM1875 amp....

The only problem I can see is that your voltages are going to be too high for a LM1875 as 2x24 ac will come pretty close to 2x32v DC.....

I still believe your best bet is to get a 2 x 18v 50vA AC transformer and run the LM1875 off of a dual supply , this way you have the correct voltages pluss you can run the preamp off of the Positive Rail.....

I actually have a Mini guitar amp design (Schematic and PCB) posted here on this forum that uses the LM1875 (and TDA2030/2040/2050) and has it"s own on board preamp with different selectable diode clippers and a volume , Gain and Tone controll ,and the power supply is allready non the same PCB with the Pre and power amp...The whole amp actually takes up less than 4in x 4in....

If you do a search for it you will find it....

Cheers