one trafo->doubler+rectifier paralel?

Status
Not open for further replies.
schematic
 

Attachments

  • tr.jpg
    tr.jpg
    28.1 KB · Views: 476
Try this one instead (schematic attached)

For caps C1,2,4,5 use maybe 50V rated caps, and for C3,6 use 70V. Use higher values if possible/affordable. Polarity is on schematic.
 

Attachments

  • 01.gif
    01.gif
    6.7 KB · Views: 436
darkfenriz, your schematic is so complicated that I confused myself looking at it 🙂 Another proposed idea has the best diode quantity but disadvantage of half-period rectification on both +/-30V and +/-60V (and 50Hz ripple is more difficult to filter out than 100Hz), besides, you noticed it, separate windings. Why not using classical variant, see attachment. there you will have 100Hz ripple at +/-30V thus suitable for high power, and 50Hz ripple doubler at +/-60V. Capacitors C1, C2, C3, C5 work at 30V, and C4, C6 at 60V. Hope I've been helpful.
 

Attachments

  • rectifier.gif
    rectifier.gif
    4.5 KB · Views: 347
another idea 🙂

darkfenriz, I was thinking that maybe you need this supply for G-class amplifier, so high power would be needed from both +/-30V and +/-60V. Then I made some modification to that schematic so now you can have full-wave rectification for all voltages 🙂 , see attached image.
[and off this topic] - is that you who proposed somewhere at this forum to hear the difference between original audio file and IMD-deliberately distorted (with DSP processing) copy. I was interested in this and then could't find that thread again. If that were you then may I send you my audio file for such processing?
 

Attachments

  • rectifier 2.gif
    rectifier 2.gif
    5.8 KB · Views: 283
darkfenriz said:
hi
I have a 2x21V transformer and I am wondering if I could use a doubler and simpe rectifier in paralel to get -30V.0V.30V and
-60V.0V.60V

help please!!!
For smaller currents it's OK to use some voltage doubler but not for a power amp.

A voltage doubler is a really hard application for the caps and the transformer and it produces lot's of unnecessary reactive power => only heat.

My advice is to get a suitable transformer and make a conventional power supply.
 
To peranders: your statement is absolutely valid for half-wave rectification circuit which is usual case of voltage doubler rectifier like my first posted schematic in this thread. I mentioned that it's not suitable for high power application eg. amplifier output stage, this is widely known fact.

My second schematic here represents full-wave rectification for all output voltages. I think that transformer operation in this case is absolutely the same as in usual single-bridge rectifier, certainly with a transformer momentary and average currents doubled in regard to high rails +/-60V comparing to +/-30V. You can try to simulate transformer operation in such circuit and reveal this.

I just reviewed again the first schematic by darkfenriz and found it is the same full-wave rectifier with the exception of higher losses - 3 diodes in series for each half-wave instead of two possible.
Anyway I agree that these circuits are too complicated and good to use only in case when one cannot have the needed secondary windings.
 
Status
Not open for further replies.