Voltage Multipliers (for our tube projects...)

[blackcorvo]

You have just reinvented the vibrator power supply that was a feature of car radios until the age of the 12-volt plate tubes and 2N176 germanium power transistor.

Wow,,, I've never heard of that

[sundgist]

On the other layout, c4-6-8 are 220µ on the original plan.

Hmmm. You're right. I was following the schematic higher up on that page. The author mentions 2k2uF is plenty at 60v, but as you say the layout lower down shows 220uF.

The multiplier circuit shown (to the right) is repeated as many times as needed. It uses just two diodes and two electrolytic capacitors to make a copy of the input voltage and add it to the original voltage. In this way it will add 12v to give 24v. Build two multiplier stages and the voltage will rise to 36v. The PCBs in the ZIP file go all the way up to +60v. Since the output current capability reduces as the voltage rises, so too does the need for the 10uf capacitors. With 60v, only 2.2uf at 64v working is necessary. This will still fit on the PCB if you get the right component.

I'll try with 10uF as I've got a bag full and see how it performs.

[amptramp]

You have just reinvented the vibrator power supply that was a feature of car radios until the age of the 12-volt plate tubes and 2N176 germanium power transistor.

Wow,,, I've never heard of that

Check out this schematic for a 1940 Buick radio showing a vibrator type power supply that chopped the 6-volt input and transformed it into a reasonabel B+ voltage:

http://www.nostalgiaair.org/PagesByModel/550/M0006550.pdf

There were other car radios that had synchronous vibrators that had a separate set of contacts that acted as the rectifier.

[blackcorvo]

Check out this schematic for a 1940 Buick radio showing a vibrator type power supply that chopped the 6-volt input and transformed it into a reasonabel B+ voltage:

http://www.nostalgiaair.org/PagesByModel/550/M0006550.pdf

There were other car radios that had synchronous vibrators that had a separate set of contacts that acted as the rectifier.

Interesting...

I though of using this because I happen to have a 12v sealed battery, and I wanted to make myself a small portable tube amp, and the smaller, simpler and most efficient power inverter to use, the more "portable" this amp can be.
I'm also not aiming for too much power. Anything around 2w is enough (sometimes I get complains for playing the guitar even when its unplugged, and it is a common strat copy).
That's why im considering the idea of using a 6AS5 for my power tube, a 6BZ7 for the preamp, and maybe adding a solid-state drive channel, or just a semi-dirty booster to drive the 6BZ7.
I have yet to think a little more about it, because right now I'm working with another tube amp project that I want to make into a combo for home practice.
The portable one is more for the case of having a power outage, or travelling, or any situation a battery-powered amp might come in handy.

[iccaros]

Check out this schematic for a 1940 Buick radio showing a vibrator type power supply that chopped the 6-volt input and transformed it into a reasonabel B+ voltage:

http://www.nostalgiaair.org/PagesByModel/550/M0006550.pdf

There were other car radios that had synchronous vibrators that had a separate set of contacts that acted as the rectifier.

Interesting...

I though of using this because I happen to have a 12v sealed battery, and I wanted to make myself a small portable tube amp, and the smaller, simpler and most efficient power inverter to use, the more "portable" this amp can be.
I'm also not aiming for too much power. Anything around 2w is enough (sometimes I get complains for playing the guitar even when its unplugged, and it is a common strat copy).
That's why im considering the idea of using a 6AS5 for my power tube, a 6BZ7 for the preamp, and maybe adding a solid-state drive channel, or just a semi-dirty booster to drive the 6BZ7.
I have yet to think a little more about it, because right now I'm working with another tube amp project that I want to make into a combo for home practice.
The portable one is more for the case of having a power outage, or travelling, or any situation a battery-powered amp might come in handy.

I posted a 24volt tube amp, its .4 watt, but even that is too loud sometime, my neighbors complain after 8pm from the other block.
http://www.diystompboxes.com/smfforum/index.php?topic=91606.0
no voltage multiplier needed.

[blackcorvo]

I posted a 24volt tube amp, its .4 watt, but even that is too loud sometime, my neighbors complain after 8pm from the other block.
http://www.diystompboxes.com/smfforum/index.php?topic=91606.0
no voltage multiplier needed.

Cool, but I can't find these space charged tubes anywhere here in Brazil... and I can't buy anything via e-bay or any other site where I'll have to use paypal, because I don't have a bank account, so I wanted to try something easier for me to get. The idea I had for using the vibrator supply and the 6AS5 for the output are simple. The 6AS5 works at just 150v, it's a pretty cheap tube and not that hard to find. I might try using it at lower voltages, but I really wanted a portable amp that could give me some clean tones without having to lay my head on the speaker to hear it.

[iccaros]

you can do that simply, I am building a small amp in to a snoopy lunch box with a 3 watt 6" speaker.  you can take Ricks Superfly and exchange the 6111 with 12ax7 or 12au7's. The 12au7's work well at 120 volts.. and you can get a watt or more, which with an efficient speaker a 1 watt amp can be as loud as a 10 watt amp with inefficiency speakers.

[jaapie]

I'm about to try out the 555 Nixie SMPS tomorrow and was wondering if breadboarding is advisable or not with high voltages like this. From what I can see from a bit of googling (and inferences from this thread) it should be relatively safe (at least, as safe as a circuit with such voltages can be). I just wanted to make sure I (probably) won't inadvertently blow up or fry something when I fire it up on the breadboard.

Also, in the Nixie SMPS pdf file, it recommends not using a battery as the 12v supply when doing the initial smoke test. I'm planning on using the SMPS with a 12v SLA battery to power a Superfly. I've got a radio shack adjustable power supply (the white kind with the reversable connector tip), would that be acceptable for the smoke test? I was also thinking about trying a 9v battery, since it can only supply so much current... I'd really hate to wind up with exploded battery guts all over me if at all possible, though.

[jaapie]

So I breadboarded the Nixie supply today and it's not quite working. When I first attach the battery, the voltage shoots up to 80-90v, but immediately falls to around 4v, where it stays until I disconnect the battery. Then it slowly drops until it reaches zero or it's discharged. The guide mentions to check the 555 with a logic probe to make sure it's oscillating... I guess that means I'm off to find or build one to check that out. The inductor I'm using is a 100uH RF choke from radio shack; I built it from the schematic in the pdf file but using the components in rick's layout.
I checked the battery before and after testing it a few times; It was just over 10v before testing and about 9.75 afterwards, so it's not draining the battery too fast. Would I get better results using a wallwart power supply?

[iccaros]

So I breadboarded the Nixie supply today and it's not quite working. When I first attach the battery, the voltage shoots up to 80-90v, but immediately falls to around 4v, where it stays until I disconnect the battery. Then it slowly drops until it reaches zero or it's discharged. The guide mentions to check the 555 with a logic probe to make sure it's oscillating... I guess that means I'm off to find or build one to check that out. The inductor I'm using is a 100uH RF choke from radio shack; I built it from the schematic in the pdf file but using the components in rick's layout.
I checked the battery before and after testing it a few times; It was just over 10v before testing and about 9.75 afterwards, so it's not draining the battery too fast. Would I get better results using a wallwart power supply?

In the next week I will attempt to build Ricks voltage multiplier so I will see if I can duplicate your issue and report back, I have the same inductor, just waiting on the transistors to start.

[jaapie]

I rebuilt the circuit a few times and it still wasn't working right, so i built this little logic probe http://www.swansontec.com/sprobe.html to check the output of the 555 (I used an LM358 for the opamp). When I check the output with the logic probe, only the "high" LED lights. I switched out the 2n2 cap to change the output to an audible frequency and not only could i hear the output when connected to a speaker, but both LEDs lit up (which is what I'd expect if it was oscillating). I changed the value back to 2n2 and tried a few different caps to make sure the one i'd been using wasnt bad, but every 2n2 I've tried won't light up both LEDs.

Since the circuit oscillates at lower frequencies with the exact same components (except for that cap), I'm pretty certain that the timer SHOULD be oscillating with the 2n2 caps-- but I have no way of telling (I don't have access to an oscope at the moment). The most reasonable explanations that I can think of are:

1. The timer IS oscillating, but with with the 2n2 cap, the output frequency is too fast for the LM358 to track, so it only lights the HIGH led

2. Maybe the breadboard's capacitance (or a bad/wrong value component) is screwing with that particular cap in such a way that lower values of caps (like the 2n2) are more affected than higher values

3. All my 2n2 caps are bad.

FWIW, All the caps I've tried so far are the little green mylar ones. Would electro or ceramic work better?
Also, since the problem appeared to be coming from the oscillator, I only have the 555 part of the SMPS on the breadboard right now (U1, R1, R2, R3, C1 and C2 on the schematic Rick posted on the first page)

[frequencycentral]

1. The timer IS oscillating, but with with the 2n2 cap, the output frequency is too fast for the LM358 to track, so it only lights the HIGH led

Yeah, the whole point of a SMPS oscillator is for it to be operating above human hearing. Not sure of the freq of this one, but certainly higher than 20,000kHz. I doubt the LM384 (or your eyes if it did) would track that.

Also, I *think* it has an uneven pulse width duty cycle, so that would make sense that you're seeing only the 'greater' part of the pulse width - ie only lighting one LED.

[sundgist]

Not sure of the freq of this one, but certainly higher than 20,000kHz.

Too many zeroes or extra k? 


With my build, (I used 150uH inductor I had to hand) the frequency is about 45kHz from 230v down to 110v but then drops steadily to 18kHz as the voltage is turned down to 90v.

Can't help with where your issue might be, orientation/pinout of irf740 and bc547 perhaps?

[frequencycentral]

Not sure of the freq of this one, but certainly higher than 20,000kHz.

Too many zeroes or extra k? 

.....well, you knew what I meant. 

[jaapie]

I remembered that I've got a friend around the corner with a scope, so I dropped by to check out the oscillator output-- it IS oscillating after all. I guess I'd either built it wrong the first few tries, or my problem was somewhere else.

Yeah, the whole point of a SMPS oscillator is for it to be operating above human hearing. Not sure of the freq of this one, but certainly higher than 20,000kHz. I doubt the LM384 (or your eyes if it did) would track that.

Also, I *think* it has an uneven pulse width duty cycle, so that would make sense that you're seeing only the 'greater' part of the pulse width - ie only lighting one LED.

According to the math, the circuit should oscillate around 31kHz, which the scope verified.

So, back to the breadboard to see if I can get the rest of it working now. Thanks for the help!

[sundgist]

According to the math, the circuit should oscillate around 31kHz, which the scope verified.

This just confused me for a bit so bear with me through this brain fart.

I've measured the ripple at the output which is 45kHz. Which is also mentioned in the notes here as the oscillator frequency. http://www.ledsales.com.au/kits/nixie_supply.pdf

My question is.... How does 31kHz become 45kHz??

I may now be making a few assumptions.

You've measured the output of the 555 at 31kHz which going by the values is right if the 555 is wired using only R3 (1k), R2
(10k) and C2 (2n2). Which I believe is how you measured 31kHz.
R1(56k) and bc547's collector on the control pin would be affecting this ratio to get it up to 45kHz, or in my case at 90v, down to 18kHz.
I havn't got a datasheet handy. Would this change the frequency or the duty %age?

Am I thinking this through right?

All replies appreciated though I am off to sleep right now......

I remembered that I've got a friend around the corner with a scope, so I dropped by to check out the oscillator output-- it IS oscillating after all. I guess I'd either built it wrong the first few tries, or my problem was somewhere else.

I suppose you could rebuild the whole circuit and take it back to your friend's and see what is happening.

[jaapie]

I've measured the ripple at the output which is 45kHz. Which is also mentioned in the notes here as the oscillator frequency. http://www.ledsales.com.au/kits/nixie_supply.pdf

You've measured the output of the 555 at 31kHz which going by the values is right if the 555 is wired using only R3 (1k), R2
(10k) and C2 (2n2). Which I believe is how you measured 31kHz.
R1(56k) and bc547's collector on the control pin would be affecting this ratio to get it up to 45kHz, or in my case at 90v, down to 18kHz.
I havn't got a datasheet handy. Would this change the frequency or the duty %age?

I'm pretty sure that's what's going on. I was a little confused because of the discrepancy between what I'd measured and the spec'd frequency, but since I'd taken most of the rest of the circuit away I just figured that the rest of the components would cause the frequency to change when they were added back in. The only troubleshooting tip in the notes said to check the output of the 555 if the circuit wasn't working...now that I know it's oscillating I'll hook up the rest and see what's happening.

[jaapie]

I was too busy with school to work on it again until yesterday morning, but after a couple false starts I was getting 200-300v without a load! When it's powered up it starts putting out 230-340, which gradually drops to 200-300v over a minute or so. Now I gotta solder it all up!

[Renegadrian]

tres utile et interessante si vous parlez francais, mon amis!!!

(j'ai etudiè francais à l'ècole, beaucoup d'annèes sont passè...)

-.-

http://techniguitare.com/forum/realisation-montage/alimentation-decoupage-12v-300v-sg3525-t7791.html

[iccaros]

tres utile et interessante si vous parlez francais, mon amis!!!

(j'ai etudiè francais à l'ècole, beaucoup d'annèes sont passè...)

-.-

http://techniguitare.com/forum/realisation-montage/alimentation-decoupage-12v-300v-sg3525-t7791.html

Google Chrome's Translator works well in this case