Rectifier Noise

[hesamadman]

I just had a thought and wanted some opinions. I hear that bridge style rectifiers are the way to go with SS. The rectifier setup I have in my el84 30 watt is setup like this.

Leg one of PT sees two 1N4007 in series the other leg sees two 1n4007 in series and then the end of those are tied together at the end combining both legs.

Would it be worth changing to bridge rectifier. Is there even going to be a difference? Wondering if this isn't an efficient way, if any AC ripple may be getting through any possible way.

[sluckey]

You can't use a bridge to replace that conventional FW rectifier.  (Well you can, if you don't connect the negative terminal of the bridge to ANYTHING. You would only be using half the bridge, only two diodes.)

[AZJimC]

That sounds like the full wave rectifier, made via a center tapped transformer winding, and diodes that take each of the opposite phase legs (because the center is the ground) and combine them into pulsing DC at 120hz.

A bridge rectifier would not work with that type of hookup. Let's say you've got a transformer with 300v-0v-300v winding. The 0v gets grounded, and each leg then swings, simplifying a bit, +300v to -300v to +300v etc... 60 times per second, or 60hz. This would be 300 x 1.41 = 423v DC when rectified.

Now, lets decide to use a bridge instead. First we have to unground that center tap, or it will become a short. Then hook up the bridge, with it's negative pair of diodes meeting at ground, and it's positive diodes to B+, CAP's, etc.  Now, we have 300+300v or 600Vac, from one end of winding to the other. The bridge takes that 600v and does the same thing your present rectifier does, but now producing 600 X 1.41, or 846v DC.

SO you see, it's two different designs, rather incompatible.

Jim

[AZJimC]

....as far as your noise problem, it's not the job of the rectifier to take the ripple out of the DC, it only chops off the negative half of the AC swing. The pulsing DC is then sent to large value capacitors to smooth the pulsing left by the diodes, or bridge, or rectifier tube.

[hesamadman]

You can't use a bridge to replace that conventional FW rectifier.  (Well you can, if you don't connect the negative terminal of the bridge to ANYTHING. You would only be using half the bridge, only two diodes.)

Ok. So I would only use a bridge rectifier with a PT with no center tap correct? Basically each type of rectifier has its place?

....as far as your noise problem, it's not the job of the rectifier to take the ripple out of the DC, it only chops off the negative half of the AC swing. The pulsing DC is then sent to large value capacitors to smooth the pulsing left by the diodes, or bridge, or rectifier tube.

My noise problem could very well be unrelated. But for my learning purposes I want to throw this out there and see what you can tell me. If a rectifier was 100% efficient in removing the negative part of the AC waveform, there would be no need for filter caps right? As there could be no way of having any AC ripple. Or is there more to it then that?

[Willabe]

....as far as your noise problem, it's not the job of the rectifier to take the ripple out of the DC, it only chops off the negative half of the AC swing. The pulsing DC is then sent to large value capacitors to smooth the pulsing left by the diodes, or bridge, or rectifier tube.

But for my learning purposes I want to throw this out there and see what you can tell me. If a rectifier was 100% efficient in removing the negative part of the AC waveform, there would be no need for filter caps right? As there could be no way of having any AC ripple. Or is there more to it then that?

No.

He just told you, read it again.


                      Brad        

[hesamadman]

I see what he said. I didn't understand how it could be pulsing until I drew a sign wave in my head. Was a very stupid question now that I did this. Lots of positive humps still left in the power supply. Someone can kick me if they want.