A new PSU (SMPS like) for vacuum tubes

[gtudoran]

Hey guys,
Today i had an idea, it wasn't new but well ... let's say it was very obvious. I head many complains about the fact that SMPS is used in vacuum tube designs, so i thought ... what if i can use another trick to design a HV PSU that will use a transformer and no HF switching.
The idea behind that was: yes you can use back to back transformers but that will push you to use AC only power supply, witch ... let's be serious is not so common in guitar/bass world. The main PSU for pedals and gadgets are 9V-DC and 12V-DC... but then how can you use a transformer for HV in DC volatge... well you can't do that, only if, you simulate a 50Hz mains freq.
An so this little idea was born: use a 555 to have ~~50Hz and a switching element  (FET) to drive an ordinary iron core transformer, so no fancy inductors, no fancy HF cores, just ordinary iron core and ordinary transformer (according to your power supply you can use a 9V or a 12V transformer).
So that is it... common rectifier bridge (no high speed diodes)... and a classic filtering cell...

The schematic is here: https://dl.dropboxusercontent.com/u/15275178/Vacuum%20Tube%20PSU%2012V/VTPSU.pdf

Egale file is here: https://dl.dropboxusercontent.com/u/15275178/Vacuum%20Tube%20PSU%2012V/VTPSU.sch

You can use your own parts footprint. I've build it on a test board and is looking very nice... what can i say, a good alternative to a SMPS PSU where you want to use a common DC voltage and not a much more uncommon AC voltage.

Build it and tell me what you think.

Best regards,
Gabriel Tudoran
Analog Sound

PS: Watch you fingers HV inside.

[tubegeek]

Gabriel - what waveform will be going into the transformer? I would expect all sort of radiated noise from a squarish wave and - since stompbox enclosures are so small and require high-impedance inputs - I would expect trouble keeping it out of the audio circuit.

Just a guess, haven't tried the circuit of course, but a concern....

[Perrow]

Input AC of a lower voltage than the secondary winding of your transformer, has been done and reportedly works well.

[Gus]

Not new
http://en.wikipedia.org/wiki/Power_inverter

[SISKO]

+1 at what tubegeek said

[gtudoran]

@tubegeek and SISKO i will post some reports after i will supply a real preamp with it and i will get back to you

@Perrow you could use a 220v-15v transformer and supply the whole circuit with 12v

@Gus i didn't say it was new, i was showing another way of getting high voltage... just another way for those who don't like boost SMPS


Best regards,
Gabriel Tudoran
Analog Sound

PS: i'm waiting for my new O-scope to arrive and then i will take a more in depth look at the output of the SMPS

[psychedelicfish]

The reason that most SMPS designs operate at high frequency is that efficiency increases, and the size required decreases. I honestly don't see the problem with high frequency stuff, as you only really need "fancy" parts if you're trying to squeeze as much efficiency as you can from it.

[gtudoran]

I don't have any problem with HF was just an experiment witch i was thinking to share with you 

Best regards,
Gabriel Tudoran
Analog Sound

[Seljer]

Someone here might be able to correct me, but I believe you've designed a forward converter, but running at low frequency and without the output choke. And you're probably running to poor transformer well into saturation.

Slightly nicer to the well-being of the transformer would be to get use a center tapped transformer and two transistors so you could get a push pull configuration.


The benefit of running high frequency, is smaller transformers and filter components (at twice the frequency you can halve the capacitance for the same amount of ripple) as well as switching well above the audible range so it doesn't cause any trouble that you're able to hear.


edit: or instead of using the mosfet there, hook it up to the 555 directly through a big-ass capacitor, that'll get rid of DC component

[Gus]

I posted the link as hint
Look up modified sine wave inverters for design ideas.

[PRR]

Lots of police and taxi radios were powered by vibrators and motor-generators.

[R.G.]

I think:

- Using high frequencies for switching power supplies has two major reasons in this milleau; it makes the reactive components to transfer and filter the unavoidable noise of switching much smaller, and it moves the "power line frequency" above audio, or other boundary, if that matters.

- If you're going to use a power frequency that's still in the audio band, you may as well move it up to 500Hz or 5kHz to get the 10x and 100x respectively increase in the power capability of the magtnetics. You're still going to be fighting noise at the power line frequency, but the filtering will be easier and the standard iron components work even better at higher frequencies if they're not too high.

- Your circuit as shown may have BAD problems with iron saturation unless you ensure that the "on" time of the power switch is noticeably less than 50%.

- Your circuit as shown is a single phase forward converter. It produces power into the output caps through the bridge rectifier only during the "on" time of the power switch. There may be some minor power transfer from the leakage inductances during the "off" time.

- The power you can get out is severely limited by the two 47nF caps before the bridge rectifiers. I'm not sure what these were intended to do, but they are in series with the output current flow. Since caps cannot conduct DC, and since the majority of the power flow happens when the power switch is on, one of these caps will charge up with a DC offset, trying to balance the charge that is being fed through them into the bridge rectifier.

The concept is fine, but the details need some work.

[amptramp]

Lots of police and taxi radios were powered by vibrators and motor-generators.

A dynamotor motor-generator would look badass on a pedalboard!  They run off 12 VDC but they might need several amps.