A little rectifier theory...
Jan 6, 2008 at 9:57 AM Thread Starter Post #1 of 11

Nisbeth

Headphoneus Supremus
Joined
Oct 20, 2001
Posts
1,828
Likes
23
Let's consider this, slightly flawed, rectifier circuit with the connections for AC and DC mixed up. Assuming this was a completed, soldered PCB, would it be possible to get it to work correctly but simply flipping some of the diodes around? And if yes, which diodes would have to be flipped so that the polarity on C1 is correct?


/U.

PS: Now, this is of course a purely hypothetical situation, naturally I would never do anything as stupid as this in real life
rolleyes.gif
biggrin.gif

 
Jan 6, 2008 at 10:27 AM Post #3 of 11
Quote:

Originally Posted by Nisbeth /img/forum/go_quote.gif
Let's consider this, slightly flawed, rectifier circuit with the connections for AC and DC mixed up. Assuming this was a completed, soldered PCB, would it be possible to get it to work correctly but simply flipping some of the diodes around? And if yes, which diodes would have to be flipped so that the polarity on C1 is correct?


/U.

PS: Now, this is of course a purely hypothetical situation, naturally I would never do anything as stupid as this in real life
rolleyes.gif
biggrin.gif



Very funny question. Just flip D1 and D3 then it will work. Next time you draw a schematic renember to make the rectifier bridge connection right.
wink.gif


Med venlig hilsen/Best regards,

Torben
 
Jan 6, 2008 at 10:42 PM Post #7 of 11
jsmith: Most "audiophiles" will argue that discrete rectifiers can be made to perform better than a single-chip solution by using high-speed and/or soft-recovery diodes. I consider this nearly irrelevant for audio as these high-spec diodes are designed for much more demanding use but I still use discrete bridges in some projects - if you already have the diodes the only additional expense is a bit of board space (and the time spent troubleshooting it when it doesn't work of course
rolleyes.gif
)


Roee: The caps across each diode are so-called snubber-caps which supposedly cancel out high-frequency noise generated when the diodes turn on and off. Again, this is probably not really necessary for audio but it is considered "good design practice" in many other applications and so I've included it. In applications where it really matters the exact value of the capacitor is calculated to cancel out a specific frequency (and a series resistor is normally added as well), but for these applications the caps used are typically 1-10nF.


/U.
 
Jan 6, 2008 at 10:44 PM Post #8 of 11
Quote:

Originally Posted by Roee2807 /img/forum/go_quote.gif
Why put capacitors parralel to each diode?
isn't one cap parrallel to the load is enough ?



They remove noise added by the bridge, up to about 100pF is a good value here and it mostly is used for regular diodes because Schottky diodes don't make almost any noise anyway.
 
Jan 7, 2008 at 12:38 AM Post #9 of 11
Quote:

Originally Posted by jsmithepa /img/forum/go_quote.gif
Ppl build discrete rectifiers still? I switched to one-chip rectifier long time ago.


how much do hexfred one-chip rectifiers cost?

nuff said.
 
Jan 7, 2008 at 2:07 PM Post #11 of 11
Quote:

Originally Posted by jsmithepa /img/forum/go_quote.gif
Here's a buck.59 1.4amp unit from Radio Shack.

Those of u who prefer ALL discrete, be sure yr soldering skill is perfect else bad solder joints may do away with your "improved" sonic discrete quality.
wink.gif



it uses fast recovery, quiet switching diodes? all you have shown is that there are CHEAP parts available from rat-shack amongst other places.

regarding soldering skill: building a discrete rectifier takes 8 solder joints on a PCB. compared to 4 for a door-slammer diode system i'd take the risk....
 

Users who are viewing this thread

Back
Top