Voltage doubler/quadrupler

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Having a brain fart this morning.

Will the following schematic work or will there be some kind of short?

Background: I use a 120V coil to get 320VDC B+ for an amp. I need 560V for a driver stage, and I'm trying to get the voltage without adding another supply.

Cheers.

Koda
 

Attachments

  • PSU-dbl-quad.png
    PSU-dbl-quad.png
    40.4 KB · Views: 287
Thanks, Mooly. Correct me if I'm wrong, but this circuit is half wave vs full wave? I ask because the 320V output supplies 500mA peaking at 1.5A, while the 640V only needs to put out 20mA. As of right now, I'm getting ready to use a DC boost board for it.
 
It is possible to graft a low-current, half-wave doubler to a so-called full-wave doubler to get a low-current quadrupler together with a relatively high current doubler:

attachment.php


Note that the so-called "full-wave doubler" is a misnomer: it looks like the real thing because the ripple is at 2*Fin, but the individual capacitors have their charge replenished only once per cycle.

As an example, I have included a true full-wave doubler on the right: it requires 8 diodes, and to my best knowledge, it is not possible to reduce that number, which makes it highly unattractive.
 

Attachments

  • 2x4x.png
    2x4x.png
    80.1 KB · Views: 332
Note that you might want to include an antiparallel diode with C1 like Mona/Ketje did, to protect it against possible startup inversions, but many designs dispense with it, and it also depends on the value of C1 relative to other caps in the circuit.

Tube TV supplies of the sixties/early seventies never used one: at most, the only concession was to make C1 bipolar (silicon rectifiers were expensive back then)
 
Administrator
Joined 2007
Paid Member
I ran out of time last night but cobbled this together from the version of the half wave attempt I did earlier. It has to be similar to Elvee's when its all drawn out properly.
 

Attachments

  • Annotation 2019-08-07 080852.png
    Annotation 2019-08-07 080852.png
    128.1 KB · Views: 94
Tiny error around C9? Polarity is backwards?

The load for the quad is 20mA. It will be LCRC filtered to output 550V or so.

It is indeed 3300uF. This amp idles at 480mA... Peaks 1.5A. It also allows such small and less wasteful resistors and still makes a good filter.
 
Last edited:
Administrator
Joined 2007
Paid Member
Polarity is OK :)

C9 and R6 need to connect to the lower winding, otherwise the 630 volt rail would just be the same as the 330 volt one... or so it is in simulation :D

That's great though.
 

Attachments

  • Capture2.JPG
    Capture2.JPG
    223.8 KB · Views: 71
  • Capture1.JPG
    Capture1.JPG
    217.4 KB · Views: 68
Administrator
Joined 2007
Paid Member
Here is the .asc file if you or anyone else wants a play. I'll look in later but I think you should try it with added load corresponding to the currents you will draw. The cap peak currents will be high... you need really good caps for this and also you might find a little added resistance to the caps would cut down the peak currents dramatically and increase the cap life significantly.
 

Attachments

  • Draft1.asc
    3 KB · Views: 28
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.