AbouT the PSU II simulator & the "Cockeyed Bridge"

[Mosquito]

Hi everyone
My question is very simple,
It is correct to simulate a SS diode full bridge rectifier, followed by a CLC filtering, get the desired B+, and then convert to the "cockeyed bridge" topology, while expecting the same Vout and current capability?
Thanks in advance
J

[Mosquito]

I mean ,of course, the same ratings, but with the added benefits of reduced SS diode glitches in both directions: to the Xformer windings and at the filter out

[kevinkr]

D5 is serving no useful purpose and can be removed, (unless blocking dc on the secondary winding when power is removed is a big deal?) otherwise it is identical in operation to the more conventional format. This doesn't IMHO confer any real or theoretical advantages over the more conventional approach as far as I can see. Just a different way of doing the same thing.

[Mosquito]

Thanks Kevin. The claim is "where the new rectification scheme will help is in reducing the amount of SS switching noise that gets into the power trafo and, via filament windings, into the signal path."
I was concerned about the problem of modelling the PSU as if it where a normal bridge rectifier in PSU designer II. Re thinking the whole thing , It seems to me that this is effectively the case. But you need a center tapped Xformer..
Probably under 400VDC you can use only two UF4007 and one Schottky..

[kevinkr]

Originally Posted by Mosquito

Thanks Kevin. The claim is "where the new rectification scheme will help is in reducing the amount of SS switching noise that gets into the power trafo and, via filament windings, into the signal path."
I was concerned about the problem of modelling the PSU as if it where a normal bridge rectifier in PSU designer II. Re thinking the whole thing , It seems to me that this is effectively the case. But you need a center tapped Xformer..
Probably under 400VDC you can use only two UF4007 and one Schottky..

I can't see how it does that, it is after all electrically equivalent. The rectified current pulses along with any rectifier induced mis-behavior are still there in the high voltage secondary.

I'd focus on using fast, soft recovery diodes, proper snubbing, and if critical use a separate filament transformer. They cross contaminate each other anyway if dc filament supplies are in use.

[Mosquito]

Mmm .. probably the solution by Geek is the best (simple) option , when your Xformer is not center tapped..
(post 4 and 8)
FREDs on PSU

[Eli Duttman]

Since I advocate the "cockeyed bridge", I (naturally) take exception to the thinking that the diode on the CT serves no useful purpose. Make that diode "noiseless", either Schottky or vacuum, and PN diode switching noise does not sneak into the B+ rail. Hop over to AA and search for VoltSecond's posts on the 5th diode in combination with a true bridge setup.

The OP asked about simulation. In fact, the "cockeyed bridge" is a FWCT setup and should be modeled as such.

In theory, the PIV rating of the "extra" diode is of no consequence, as reverse biasing never occurs. In practice, having a substantial PIV there is a good idea, as lots of things can go wrong. For instance, one fellow installed a Schottky backwards. Fortunately, he had a decent PIV part and it did not fail.

[Mosquito]

Thanks Eli, I'll search AAsylum for VS posts.
I will try different types of PSU's in a bredboard the following weeks. SS, tubes, Doublers, to gain practical experience in such a fundamental circuit part. Me and my friends, a huge Variac and a Fluke true RMS meter..

[Merlinb]

Originally Posted by Eli Duttman

Since I advocate the "cockeyed bridge", I (naturally) take exception to the thinking that the diode on the CT serves no useful purpose.

But what's special about the cockeye in particular? That extra, fast diode works the same with any rectifier configuration. Also, if you placed it in the ground side of the circuit then it wouldn't even need a high Vrrm rating.

[TheGimp]

For the two UF4007 diodes in series, wouldn't one want two shunt caps, one across each diode instead of one across both diodes?

D5 should never see any voltage greater than the output voltage, and it will only see that at the instant that power is turned off.

Shunt D5 with a cap of equal value to the two caps shunting the two diodes in series and it will only seee B+/3.

It's PIV rating can therefore be equal to B+/3, but is best kept at B+ or greater.

If D1-4 are already UF seried diodes, I don't see the advantage of an additional schottky in series with them.

If D5 being a Schottky would allow D1-4 to be 1N4007s instead of UF4007s, then I could see an advantage.