Hi, are there any benefits in using two bridges on a duel secondary transformer or connect the secondaries to form a centre tap and use one bridge? This will be for the supply for a class A amplifier.
I have checked other threads on this subject but can't find a definitive answere.
Best Regards
Alan
I have checked other threads on this subject but can't find a definitive answere.
Best Regards
Alan
Indeed.
There are also more subtle advantages: for example, it creates a permanent, galvanic connection between the output GND and the transformer's copper, a way to minimize common-mode perturbations, and it minimizes the amount of diode switchings.
For industrially-produced transformers, the turn number of two nominally identical windings is going to be exact to the turn.
If it isn't the case, even a center-tap rectification will pose problems, as it will introduce a parasitic 1*Fmains component superimposed on the 2*Fmains ripple
I discuss the topic in this post, and later posts in the thread may be instructive as well:
https://www.diyaudio.com/forums/power-supplies/143539-look-lm317-lm337-regulators-7.html#post5104900
https://www.diyaudio.com/forums/power-supplies/143539-look-lm317-lm337-regulators-7.html#post5104900
You are right. But, transformers (and inductors) is a topic many engineers/technicians never become really comfortable with. It is electro-physics, not straight forward logic. Therefore, we have to be patient and sometimes repeat ourselves. Every time more may get the drift.
You come from an area and tradition where teaching in physics appears strong judged by me from some members of the same nationality as yours. Further to the west, computer technology has been the big employer for decades on the account of electro-physics and analog electronics.
When we meet on this forum, our differences are one of the challenges to tackle.
No it does not have to: all three bridges see the same current, the difference is that the two separate bridges will see half the voltage.
Nelson Pass started the use of two bridges, he must have good reasons...
getting rid of the transformer center tap is one....
for an amplifier that is capable of over 1000 watts, a loss of 20 watts using two bridges may not matter that much...
besides high current rectifiers have gone down in prices these days....any diy'er can afford....
getting rid of the transformer center tap is one....
for an amplifier that is capable of over 1000 watts, a loss of 20 watts using two bridges may not matter that much...
besides high current rectifiers have gone down in prices these days....any diy'er can afford....
You are right, but I can tell you, that, at present time, unfortunately, this tradition has lost.
Novadays, even students of electro-technology universities (or departments of technical univers.) are not good at phisics (chemistry and so on).
Thay don't see anything interesting in technical sciences, may be, bacause, TV and smartphones (Iphones) had been invented yet (
) and they don't need anything more.
Last edited:
i first saw the use of bifilar method of winding secondaries in the GAS Ampzilla amps in the 80's, and the use of square cross section copper wires...
bifilar method exactly located the center tap that previous method of one coil atop another, the bifilar wound will have identical dc resistances and so voltage drops are identical, while the "one coil atop another" will have secondary terminal voltages become unbalanced with current draw, and so the center tap wanders off too, and this is not a good thing...
that is why a bifilar traffo makes sense to me...
and getting rid of the center tap by using two bridges totally eliminates the traffo secondary in the equation...
diode bridges are dirt cheap nowadays....
bifilar method exactly located the center tap that previous method of one coil atop another, the bifilar wound will have identical dc resistances and so voltage drops are identical, while the "one coil atop another" will have secondary terminal voltages become unbalanced with current draw, and so the center tap wanders off too, and this is not a good thing...
that is why a bifilar traffo makes sense to me...
and getting rid of the center tap by using two bridges totally eliminates the traffo secondary in the equation...
diode bridges are dirt cheap nowadays....
I don't exactly catch what you're saying. For a two-rail DC supply you either need a CT secondary or two separate secondaries. At least here in Germany the price for both options is the same if a custom wound Xformer is ordered. And I'm pretty sure, too, that in a toriodal trafo it is much easier for the winding shop to provide separate termial leads for two secondaries instead of connecting them and soldering a CT wire.
Best regards!
the ct in a split dc rail psu is just a reference point....
i have had reports from some builders that even without the ct, the amp works, bleeders of equal value provided the virtual ground, dividing the rails in half...
i make my own traffos and i can definitely say bifilar winding does not take anymore space than traditional one on top of the other winding method...
it is not hard to imagine two strands of wire wound at the same time will have the same length and therefore same dc resistance and terminal voltage....
since the 80's bifilar is my winding secondaries method of choice...
I thought the CT is usually connected to the reference point.TonyTecson said:
Leaving out the CT wire will work provided that the amp draws the same current from each rail on average. This may be approximately true in many cases. It is still bad practice.
It seems to me that bifilar winding means that each turn has more air space associated with it so the winding occupies more space. Of course, if you can carefully align things so the wire of one turn fits into the air space of the previous turn then you lose less space.
Two strands of wire wound at the same time side by side will be slightly shorter than the same wire wound bifilar. The difference depends on the twist length, and will be small for long twists.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.