Voltage Multipliers (for our tube projects...)

[frequencycentral]

Thanks Mick, I'll order some of those in different values. You appear to be our resident expert on the nixie SMPS - it's good to have your input, and it's given me plenty of inspiration to continue expermenting with it.

I'm also very curious as to what you use your SMPS's to power???

[Mick Bailey]

I use these mainly in hi-fi valve pre-amps. The high voltage gives more headroom and the current capability gives better output swing to drive the power tubes. I started using this idea due to the relatively high cost of suitable power transformers in the UK and to improve signal to noise characteristics. It also meant a real saving on space.

I've also built quite a few pedal-based guitar pre-amps  that are straight lifts of classic amp circuits. These don't tend to work very well at low voltages 'as-is' and often need modifying for (say) 30v operation. When you've got up to 300v to play with you can get the necessary dynamics back to capture at least some of the original sound.

BTW, I first saw these circuits in the '70s in an electronics magazine to power a small-to-medium florescent tube for camping. I still have one of these somewhere as an experiment from my youth, built on a piece of bent aluminium sheet. It worked superbly when powered from a small motorcycle battery. It used a pair of transistors in free-running astable mode instead of a 555 and with no regulation/feedback to limit the voltage. If you disconnected the load the voltage shot up and fried the switching transistor.

Wonder when the SMPS idea was first used for guitar pedals?.....................

[Ice-9]

After a bit of a disater (was decorating house and accidently threw out most of my electronics paperwork) I'm getting back on track , this is the HV SMPS i'm working on at the moment and its a design using the MAX1771, it should supply about 220V  at 50ma.

I have most of the components needed but am waiting on a couple to arrive so i thought i'd start the pcb making. Ther following pics are of the schematic, PCB and PCB overlay. I hope to complete and test the circuit early next week.

The next update to the circuit is going to be a saftey device. If you check the schematic pin 4 of the max chip is tied to ground, but if its switched to Vcc then the chip will shut down preventing the HV getting to the output. I will be putting a push to make switch and a resistor in there so that when the base of the enclosure is removed it will disable the HV. I have a few other things to work on with this but would like to post what i have so far.

I also have the pcb files for printing if anyone wants them i can post them up. The size of the pcb is about 30mm x 35mm

[Ripthorn]

So Mick, Rick and Mick, do you guys use multi turn potentiometers for your VR1?  That seems to be a good idea, but I am wondering if it is worth it.

[Ice-9]

Hi Brian, I have went for the multiturn pot in this for two reasons, firstly i think it will be more accurate to dial in the voltages and secondly they seem a lot solider than the single turn trim.

these are the ones i got  http://uk.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=521+9798

[Mick Bailey]

I use standard cermet pots with no problems at all, the only reason being that I have plenty of them. You can also temporarily wire in a pot, set the on-load voltage then read off the value and replace the pot with a fixed-value voltage divider. In most cases you'll find that the pot is set to its maximum travel anyhow.

[Ice-9]

you could always take out the VR and have the voltage at a set level. In the schematic i have posted above you can keep R3 value at 10K, put a link in for VR1 and adjust the value of R2, a simple equation for this is 

R2= (6.7 * Vout) Kohms

R2=(6.7 * 220)
R2= 1474 or 1M5

So for 300v output, (remember you would need to change the ratings of other components as i have only put in max ratings of 250v for some caps

R2=(6.7 * 300)
R2=2010 or 2M

the full equation is R2=R3 * (vout\1.5)

[brett]

Hi
a couple of comments...
A 555 is a very efficient oscillator for these things.  The bipolar version can source up to 200mA (!). Another option would be an LM386 set up as an oscillator.
The fellow who used an 8 ohm:1 kohm transformer is a genius.  That should give about a 10-fold increase in voltage (square root of 1000/8 minus losses).

An astable 555 will output about 7 V p-p from a 9.5V supply, and that can be boosted to about 60 or 70V p-p by the transformer.  From there you only need 2 charge pumps (voltage "doublers") to get to a usable voltage (somewhere between 120 to 200 V DC at 2 or 3 mA).  I *think* that's enough to run a few 12AX7s.  Of course, you'll want a decent 9V battery (preferably NiMH), as power consumption will be quite high (This is true of all these designs).

I notice that the first design used 220uF caps.  That seems excessive and expensive, especially for the last stages where the current is only a mA or two.  Above 50 V, my calculations indicate that you only need 5 to 10uF of storage. You'll lose a tiny amount of efficiency, but save $$$.
cheers

NB the frequency response of those audio transformers is very limited (due to their small core, capacitances, etc).  It will need to run at about 1kHz to be effective.  Unfortuantely, that'll spray some electro-magnetic noise around at a mid-audio frequency.  Some shielding, especially over the transformer, might be called for.

[Johan]

did you guys ever consider something like this?

it's the little driver that you need for backlighting on displayes. with 5 volt in ( the regulator is soldered on the back in the pic) I get 70Volt AC out. efter rectideing I get 100volt dc..
the teleplug is for size reference
j

[Ripthorn]

Okay, I am way interested in that.  Is the max input voltage 5V?  And where would you go about getting one and roughly what does it cost?

[frequencycentral]

Okay, I am way interested in that.  Is the max input voltage 5V?  And where would you go about getting one and roughly what does it cost?

+1! Tell us more.....

[Ice-9]

+ another 1, defo tell more, how much current can it deliver ?

[Renegadrian]

It seems we're all interested...Yeah, tell us more!!! 

[Renegadrian]

THIS ONE seems pretty easy to build...

[Johan]

Okay, I am way interested in that.  Is the max input voltage 5V?  And where would you go about getting one and roughly what does it cost?

+1! Tell us more.....

you can find them at most of the big electronics component companies..for example www.elfa.se search for part number 75-559-49  or electroluminence panel voltage converter. they usualy want 3-5volt dc in and you buy them according to how much current you need. often you find them under opto/display/backlight in the catalogues.
cost is around 10 euro
j

[brett]

Hi
about a year ago I looked at luminescent panel drivers.
All of the reasonably priced ones had firly low current capacity.  *maybe* enough to drive a 12AX7 stright into a JFET buffer with a couple of megohms of impedance, but maybe not, too.
For a while (since 2003) there's been a 12V wallwart-powered 12AX7 pre-amp kit available here in Oz.  http://www.siliconchip.com.au/cms/A_30804/article.html

I bought the article way back, and if I recall correctly, it used a DC to AC converter chip and a step-up transformer (switching?) so that the final size of the box was quite small.

I still think you could do a lot with an astable 555, 1k:8 transformer and a bridge (about 100V DC without a charge pump stage).  All in about 2 inches by one inch of PCB.

cheers

[Renegadrian]

http://www.tube-tester.com/sites/nixie/nww/nixie-watch.htm

[Ice-9]

Wow, this thread has turned up some wonderful ideas and links

That is one well wound inductor, It would probabally take longer for me to wind that than it would to learn to play the entire collectoion of beatles song.

That inductor is a piece of art in its own right.

[Ice-9]

Iv'e been working on this PSU now for a while and the rest of the components have arrived so i have made up the PSU. I'm powering the lot from a wall wart AC adapter and it should work with an input anywhere from 9v-18v,  I'm getting 230v out of it as the picture shows.
This is a PSU for a project i'm working on and has +/- 15v DC to power opamp circuits, 12v Dc at 1A to power tube heaters and 230v Dc to power the valve plates all built in, The PCB also houses the footswitch led and jacks. I will do some testing under load and may try to get about 270v on the output by changing the feedback potential divider top resistor to about 2M instead of the 1M5 i have there now. (I will do the maths first).

Anyway here are some pics now showing what i have so far.


A pic from the top of the PCB showing most components

With meter connected showing the 230v output

Showing the bottom of PCB with inductor and PWM chip

Showing the inductor and stuff up a bit closer.

[frequencycentral]

Mick that looks excellent, and that SMD MAX1771 is tiny! The inductor looks like it will handle the ma too.