![]() ![]() |
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
124.189.0.93
OK, so a standard rectifier is 4x diodes in a diamond shape ... with:
* 2 diodes in series on the top part of the diamond - and 2 diodes in series on the bottom
* with cathodes facing forwards to the '+ve' output
* and anodes facing backwards to the 'ground' output
* with the transformer secondaries connected top and bottom, at the halfway point between each pair of series-connected diodes.
When I check the diodes before installing them (with my Peak DCA75), as well as telling me the diode is good (or bad) ... the meter shows a voltage reading (Vf).
So it seemed to me that it would be best to select diodes so that voltages matched.
But should I match diodes so that:
* the two between the AC (transformer) inputs to the diode bridge and the '+ve' point add to the same number as the two diodes between the 'ground' point and the AC inputs?
* or the two on one side of the diamond, between the ground point and '+ve' DC point, match the two on the other side of the diamond?
Follow Ups:
I wouldn't worry about voltage drops across the diodes. IN today's world, they are pretty well matched during the manufacturing process.
With a CT transformer secondary, who's to say how close the CT actually is between between the two "hots" - in a transformer it ISN'T equal number of turns on either side. And, all the the rectified AC does is (typically) charge the filter cap(s) - to a voltage probably a LOT higher than the final B+/- rails AFTER downstream regulation.
Of maybe more importance - at least to a lot of "designer audiophiles" are the artifacts of the diodes switching characteristics - spikes/harmonics/whatever. That's why different classes of rectifiers have been developed. Look that/those up and prepare to get lost!
In fact, after a recommendation from someone ... I have stopped using 4x BYV26C diodes in a 'diamond' for rectification and now use a Shindengen Bridge Rectifier (D6SBN20-7000).
This enables me to get the dual output voltages to within 0.1v of each other.
Hi
Not sure what your meaning with the two secondaries part, can you draw that circuit? There should only be two connections to the bridge from a transformer. Often a there is center tap or series connected secondaries with the center point being grounded, then the bridge circuit produces equal + and - voltages with respect to ground (that center tap point)
The "on" voltage for a normal diode is around .6V but they should all be very similar. Matching isn't necessary and a way to look at it is the "on" direction it conducts with a low effective Resistance (but also a .6V threshold) but in the other direction it's an open circuit.
Sure, Tom.
![]()
This is to produce a single, +ve DC rail - so either one of the transformer secondaries can be attached to the two 'AC' connectors ... or both can, in parallel.
Some info I have subsequently received suggests that, if I want to get anal then - due to the diodes' "forward voltage" - it would be best to have the sum of the Vfs of the 2 diodes on the 'top half' of the diamond bridge equal to the sum of the Vfs of the 2 diodes on the 'bottom half' of the diamond.
Yes, the configuration where a center tapped transformer or two secondaries in series to make a CT is what it sounded maybe like. That config has the ground at the center tap and the Bridge rectifiers output is now plus and minus Voltage with respect to ground.
Note sure a tiny difference in on threshold makes any difference once one has more than a couple volts being rectified. Are you making a bridge out of separate diodes?
There diodes that have a lower threshold V than generic silicon diodes if that matters, look for "shottkey" diodes
Thanks, Tom,
Yes, I am making a bridge out of 4 separate diodes.
End result is 24/25v DC.
I will look up Shottky diodes.
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: