Cheap and easy tube power supply

[JKowalski]

Cost - $3

I bought two wall warts from my local thrift store for $0.99. Took em apart - gave me a 120VAC/12.6VAC transformer, 2 amp, a 120VAC/9VAC transformer, 1 amp, a couple of 25V 2200uF caps, and eight rectifier diodes. Put it together, adding in two 400V rated filter caps, and you have a tube power supply, a la the Real McTube.

Please - anyone who sees any problems in this design that could lead to dangerous failures, etc. speak up! I have not found much information on running power transformers back-to-back in this kind of design, save in the McTube, and am largely unaware of any dangers or problems it may lead to in actual use. Don't worry though, I am taking every possible precaution (even to the extreme!) in testing this device 



The construction is very simple. You need a hefty circuit board, the layout is easy. Since the transformers have no mountings, I cut holes for the bottom cores in the board, layed them in, and hot glue gunned the outsides to the board.

Mine, due to the mismatched secondaries, has a B+ voltage of 330VDC+ unloaded. Quite a bit more then I orginally anticipated. The DC is quite good, very little ripple in either line.

The filament voltage is about 15-20VDC right now, I am going to get a hefty 12V regulator to stick in there.

It works well, so far - I have not tested it loaded. There may be problems running it for extended periods of time, as I have not tested for more then a minute or two straight yet.

AS ALWAYS IF YOU DECIDE TO DO THIS TOO BE EXTREMELY CAREFUL! 100-400V IS A LOT! DISCHARGE THOSE CAPS!

[Ripthorn]

How did you determine the amount of ripple in it?  I would be very interested in this if all works well (actually, I am running back to back transformers for a little project I am doing right now).  I think mine might have some ripple and I want to check that out.

[Mike Burgundy]

Did you use the 12.6V tarnsformer as primary, and the 9V as step-up? That's a recipe for trouble!
These things (as RG once pointed out to me) are built for cost-efficiency and won't handle stuff well beyond their ratings - you're feeding 12V into a transformer that's expecting to see 9V - it will saturate and might (edit - probably *will*) destroy itself in time, wether that's short or long (or really short), who knows. I don't see any reason why it wouldn't work the other way round (120V-9V into 12V-120V, should yield 90V)
That exchange is somewhere in the archives...
That the voltages are higher without load is normal - these transformers are rated at their secondary voltage *at specified load* so if you have a 1A 9V secondary, it will put out a higher voltage at lower current draw.

[JKowalski]

How did you determine the amount of ripple in it?  I would be very interested in this if all works well (actually, I am running back to back transformers for a little project I am doing right now).  I think mine might have some ripple and I want to check that out.

Later today I am going to put it on the scope - being at such a high voltage, I cannot do this directly, so I am going to hook a decoupling capacitor to the output with a resistor to ground (high pass filter) with a cutoff well below 60hz. Then I can scope the ripple on the DC line without any really high voltages being near my scope - which can only handle up to 300VDC.

But I could approximate the ripple with my FLUKE multifunction voltmeter, it reads 0Hz and 0.15 mV RMS. That leads me to believe that there is not any substantial ripple on the supply line. I'm not sure though - perhaps that reading means nothing, and it will be terrible when I see it on the scope. I'll report back.

Did you use the 12.6V tarnsformer as primary, and the 9V as step-up? That's a recipe for trouble!
These things (as RG once pointed out to me) are built for cost-efficiency and won't handle stuff well beyond their ratings - you're feeding 12V into a transformer that's expecting to see 9V - it will saturate and might (edit - probably *will*) destroy itself in time, wether that's short or long (or really short), who knows. I don't see any reason why it wouldn't work the other way round (120V-9V into 12V-120V, should yield 90V)
That exchange is somewhere in the archives...
That the voltages are higher without load is normal - these transformers are rated at their secondary voltage *at specified load* so if you have a 1A 9V secondary, it will put out a higher voltage at lower current draw.

Hmm. I suppose I should just find another matched 12VAC/120VAC to stick in there instead then, that shouldn't be too hard.

I flipped through a couple pages about core saturation, I see what you mean.

And yes, I expected it to be higher because of the 12v/9v/120v deal - that was pretty much the plan. I did not realize that the transformers were so picky about voltage ratings. Thanks alot! You probably saved me from a heap of trouble!




And a question - Why arent wall warts designed with a voltage regulator in them? It's not like they are expensive, or complicated to incorporate! I never understood this.

[Ripthorn]

Regulated wall warts do have a regulator in them, it's just the cost and space and proper heatsinking that makes it so that most manufacturers don't put them in.

[biggy boy]

For added safety may I suggest a fuse on the 120 volt primary side.
I have made basically the same setup as you have to power my Valvecaster at 157 volts.
I used one of the fuse holders that you mount through a hole in the enclosure for easy access.

Glen

[davent]

AudioXpress has an interesting  article on using AC wall warts in creative ways here. http://www.audioxpress.com/magsdirx/ax/addenda/media/stamler2874.pdf

Duncan's Amps has a very helpful, free, modeling software to help in designing simple linear (unregulated) mains power supplies, typical of tube power supplies. http://www.duncanamps.com/psud2/index.html

dave

[JKowalski]

For added safety may I suggest a fuse on the 120 volt primary side.
I have made basically the same setup as you have to power my Valvecaster at 157 volts.
I used one of the fuse holders that you mount through a hole in the enclosure for easy access.

Glen

Yeah, I have a fuse in line right now, its just to the left of the picture. I will get one of those enclosure mounted ones when I find/build an enclosure for my project.

[earthtonesaudio]

I did not realize that the transformers were so picky about voltage ratings.

Me either.  I thought they were just ratios.

[frank_p]

I posted it to Mike today and it's been a while since it is in my "folders".  I talked to Mac also and I am at the same point again.  There is an article in GEO about transformers:

www.geofex.com/Article_Folders/xformer_des/xformer.htm

Primary inductance
Set up a test circuit with a high power transistor or MOSFET switch to switch 12VDC into the series connection of the transformer primary and an accurately known resistor to ground. Parallel the series resistor/primary with a clamp diode that is normally reverse biased. Drive the switch with a variable duty-cycle pulse generator and slowly increase the duty cycle from very tiny, watching the voltage across the resistor and inductor with an oscilloscope. The voltage across the resistor, which is proportional to the current through the inductor, will ramp up linearly when the switch is on and ramp down linearly through the diode when the switch is off.  Ensure you never use a duty cycle of 50% or more, as that does not let the inductor current go to zero before it is started back up again. In this case, the inductor current will ramp into saturation and burn something out.

What could be a good clamp diode and a convenient mosfet transistor to wire up what R.G. is talking about in the end of the paper ?  Anyone have a cue ?

Thanks.

fhp

PS:  I post this because I also have the Mc Tube in seperate pieces in my bins and it was a project I wanted to do but I was not quite sure if the conceptor of the project was knowlege(able) about the two transformers setup of this particular project.   And since (indeed) it's a high voltage project, how can we be confident about transformers issues in that particular project ?

Sorry: Thanks again.

[solderman]

Hi
I'm building a high voltage tube amp and this type of PS was one of my options. I bought 3 different radioshack type of tyrannies, one 240/12x2V 2A and one 240/12x1 1A and one 240/9v 2A. And starts to connect them in different ways to See what happens. Before that I red a couple of article earlier mentioned here. The result was OK until I added load on the high voltage output. Then the "new" secondare/primary got hot ans started to smell. If I remember correctly it the 240-12-12-240 was the worst case. I loaded down them with normal 240V lightbulbs from 15W and upwards. With 100W theoretical ~ 0,41A but in practice with losses probably over 1A the trannies got so hot I diddent dear to continue.

In stead i did whats forbidden and and used 4x 1N1007 to rectifi the wall 240V and that got me 308V DC and yes i 0,25 A used fuses on both conductors all the time PLEAS DO NOT TRY THIS ITS VERY DANGERUS
. this works but I don't feal safe with it so Nixie supply and normal PS supply for the filament is the road i,m going down at the moment.


//Solderman

[frequencycentral]

I never really saw the logic of the back-to-back tranny approach. I've built quite a few submini tube amps running at 12v, using charge pumps or nixie SMPS's to get HV, which I think are a geat solution if you want to keep things physically small. When I eventually get round to building a full size tube amp I'm planning to use a suitably spec'd and specifically designed PT. I can see the attraction of the back-to-back trannies re. cost, it just seems to me that getting the correct parts outweighs saving a few £s/$s. Just my 2c.

[frank_p]

I can see the attraction of the back-to-back trannies re. cost, it just seems to me that getting the correct parts outweighs saving a few £s/$s. Just my 2c.

Thanks for confirming my thoughts Rick.

[MetalGuy]

If small transformers for tubes power supplies are available from well known online store in Europe only for 10 Euro the piece I think this problem is already solved.

Pri: 230 V
Sec1: 200 V @ 0,015 A
Sec2: 6,3 V @ 0,6 A
Leistung / Power: 7 VA
Grösse / Size 50 * 40 * 35 mm

[frequencycentral]

If small transformers for tubes power supplies are available from well known online store in Europe only for 10 Euro the piece I think this problem is already solved.

Pri: 230 V
Sec1: 200 V @ 0,015 A
Sec2: 6,3 V @ 0,6 A
Leistung / Power: 7 VA
Grösse / Size 50 * 40 * 35 mm

Cool, and the source? Sec1 could be a little higher - maybe they have a range of specs?

[davent]

Hi
I'm building a high voltage tube amp and this type of PS was one of my options. I bought 3 different radioshack type of tyrannies, one 240/12x2V 2A and one 240/12x1 1A and one 240/9v 2A. And starts to connect them in different ways to See what happens. Before that I red a couple of article earlier mentioned here. The result was OK until I added load on the high voltage output. Then the "new" secondare/primary got hot ans started to smell. If I remember correctly it the 240-12-12-240 was the worst case. I loaded down them with normal 240V lightbulbs from 15W and upwards. With 100W theoretical ~ 0,41A but in practice with losses probably over 1A the trannies got so hot I diddent dear to continue.

//Solderman

Hello,

I'm assuming that you have the 240-12-12-240 transformer wired so the 12v's are paralleled and the 240's are paralleled, the 12's are being fed 12v from the 240/12x1 1A transformer. According to the AudioXpress article, to calculate the current draw through your 12v primary it would be;  0.41A x  240/12 = 8.2A. So your backwards transformer would need to be 12v x 8.2A = 98.4VA in order to  handle your current rather then the 12v x 2A = 24VA transformer you have on hand. In the  example in the article, designing  a two tube, 6v6/12ax7 amp they used  a current draw of 50ma through the primary, so using the formula you get 0.96A pulled through the 12V primary , the 24VA transformer would be plenty big enough. If you're just doing a preamp with a couple triodes the current draw is going to be only a few milliamps. These numbers are all excluding the heater current draw as it doesn't get pulled through the 12v primary.  In TUT5 in a chapter dealing with building an amp with dumpster finds the author calculates a backwards 12VA transformer as being big enough to power a 6v6/12ax7 amp.

Take care,
dave

[MetalGuy]

Cool, and the source? Sec1 could be a little higher - maybe they have a range of specs?

http://www.tube-town.net/ttstore/product_info.php/info/p2358_Tube-Town-Powertransformer-EI-230---200---6-3-7-VA.html

Check out also this one:

http://www.tube-town.net/ttstore/product_info.php/info/p2192_Torodial-30VA-Preamp.html

[davent]

Another article i just came across on using back to back transformers for tube projects, also offers a kit for the build. http://www.electricwestern.com/diytubes.html

dave