3 rails power supply question

[Deepdark]

Hi

I'm drawing a positive, negative and 48V power supply. I plan using a LM317HV for the 48V rail. other rails are +15V and -15V and a 18V dual coil toroidal, readily available everywhere. But, how to manage the 48V rail? Looking at the exemple below, we have a 20V-0-20V transformer (with the Toroidal, we just strap the lead we would use for a serie connection, and send it to the ground. So we effectively have 20V each side. Each side will passed through the rectifier, each their turn. So once rectified, we got, in the exemple below, 20v x 1.414 = 29v. Far enought for the 15v rail, but not enought for the 48v rail. I just don't understand the doubgler circuit below. Thanks

 

[Deepdark]

I should have said, explain like i'm five

 

[JohnRoberts]

See Dick run...

This shematic seems above a 5 yo comprehension level.

That doubler is using full wave rectification which is unusual since 48V rails are usually low current so just half wave rectified (less parts).

C3 and C4 are doing the same thing just opposite polarity of the mains waveform. So I'll just describe one.

When the transformer winding connected to the bottom of C3 swings negative (-1V), the top of C3 is held to 0V by the diode in BR2 connected to the center tap (0V), charging bottom of C3 to -1V (for +1V across that capacitor). Then when the same transformer winding pushes positive, C3 that was charged up to 1V during the negative half cycle now has its bottom pushed up to +1V, this makes the top of C3 +2V charging C11 to +2V through the rectifier diode. 

C4 does the same thing for the other polarity of the mains signal swing. These parallel doublers deliver 2x the current of one, but like I said these are generally low current PS.

Another stage could be stacked in series to make 3X the voltage if needed. C3 is pumping current into C11 at the mains frequency, so size of that capacitor directly affects output current of 48V supply, but these are generally only several mA draw.

JR

 

[Deepdark]

Thanks John! Explain like I'm 10, then  8)

Seriously, I revised a schematic with another type of doubler/tripler connestion. So the trick of the first schematic is to stock the voltage into C11.W hen we are on a negative cycle, its charge the bottom of the cap (don't get it about the bottom being charges at -1 and made it +1 at the top)  and when we're on a positive cycle, it,s charge the bottom positively (ie, +1 at the botom so we got +2 at the top, same principal as previously??) we basically doubled and discharge this voltage into C11. Right?.

In the revised schematic, I separate the ground from the bipolar rails from the 48v rail. Is it ok to do this? I want to separate them and connect my phantom 0v to XLR pin 1. But I'm not sure if it's do-able since I share the same power transformer/center tap. And about the doubling/tripling circuit, I guess the trick is to stock voltage into both 1000UF (the ones in parallel to the 22K résistors) Right? For the tripler, we're passing through another series of 1000uf cap in the bipolar rails. 

 

[Deepdark]

Just tried something else to be able to use 1 coil /rails. Let me knwo if this one would work, and also, Should I be better off to make a tripler instead of the doubler for the phantom power rail?

 

[Deepdark]

hmmm that is weird. I received 2 tonofication on my mail telling me there was 2 reply to this post, but nothing new here...weird 

 

[mjrippe]

hmmm that is weird. I received 2 tonofication on my mail telling me there was 2 reply to this post, but nothing new here...weird 

Those were YOUR two replies, LOL!

 

[Deepdark]

HAHA if only, their was one from Ian and the other from another member, can't remember teh name. Pretty funny lol

 

[Whoops]

Each side will passed through the rectifier, each their turn. So once rectified, we got, in the exemple below, 20v x 1.414 = 29v. Far enought for the 15v rail, but not enought for the 48v rail. I just don't understand the doubgler circuit below. Thanks

Check again how the Bridge rectifier for the 48V rail is connected to the transformer secondaries, it's not connected in the same way as the +15V or - 15v rails.

In this case I guess the math should be:
40v x 1.414 = 56.56VDC