1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

Parallel Rectifiers?

Discussion in 'd.i.y.' started by Mike Hanson, Nov 23, 2022.

  1. martin clark

    martin clark pinko bodger

    No, just twice the number of windings.
    You abs do not need twice the VA.
  2. Avon

    Avon pfm Member

    Again I don't understand. If you don't use half of the windings, then you'd still get the same power output as when you used them all?
  3. jpk

    jpk pfm Member

    Hence I made the simulations: to watch the currents flow...

    My diagrams are meant for different transformers, so the first diagram uses one transformer with 2 centre tapped windings, the second diagram is for one transformer with one centre tapped winding.
  4. martin clark

    martin clark pinko bodger

    Yes - that is how a transformer works: a transformer is an incredibly simple, but not intuitive device: ultimately, you can draw current up the total VA design limit indefinitely from any secondary winding or distributed across them - the limit is the flux linkage between pri:sec. ; plus the losses you incur in the winding and core (which are - in comparison, for obvs reasons - designed to be rather small) - AND, providing only that the secondary winding(s) used, can safely deliver the total current demanded to deliver that power [i.e. without overheating long term]

    so 'unused' / under-used' windings represent no issue, since they are no load whatsoever: it is only the total load on the output side that matters.

    Ex. 1: If I have some weird trafo - oh, lets say, 1x24Vac, 1x 12Vac , 1x 6.3Vac, and I only draw from one or two windings - the total capacity of the transformer is available, on those windings: providing only that they are up to the current that represents.

    Ex.2: say we have a notionally 100% efficient, 100VA transformer, with, oh 100Vac, 10Vac and 1Vac windings:
    • if I only use the 100Vac winding I can load that to the full 100Va - notionally, 1A rms. wiring dissipation proportional to 1A = I^2 = 1unit x R(winding)
    • if I only use the 10Vac winding I can load that to the full 100Va - notionally, 10A rms. But better hope the secondary wire gauge windings are up to that ..likely not. wiring dissipation proportional to 10A = I^2 = 100unit x R(winding)
    • if I only used the 1 Vac winding I can, perhaps, load that to the full 100Va - notionally, 100A rms! Except obvs unless that winding is nice wide fat & flat copper-strip that wont last any time. wiring dissipation proportional to 100A = I^2 = 10000unit x R(winding)
    So we can load the secondaries up to the VA limit in any combo we choose, up to each secondary winding's heat disspataion limit - with no penalty: If I only use it to draw 10Vac at 0.1A for some modest purpose - well, the transformer only draws c 1w (+ inherent core losses) from the wall; the rest of the VA capacity is not used; nor inhibited.
    Last edited: Nov 25, 2022
  5. jpk

    jpk pfm Member

    Thanks for the excellent description! So basically the max VA is dependent on the iron and not on the windings provided all wires are thick enough...?
    Mynamemynaim and martin clark like this.
  6. martin clark

    martin clark pinko bodger

    Exactly that jpk - the transformer is a device capable of constant power transfer capacity - when within its rating overall.
    • If you aren't using all the available VA - it doesn't matter;
    • If you want to lean more-heavily on one winding for a given reason, while using any of the others (within total VA capacity) proportionally-less - no issue, whatsoever, short of grossly abusing that secondary winding's capacity: remember the resistive heating in that winding, and hookup wire, is proportional to I^2...

    - and a few other niceties, such as tolerance to v mild continuous, and gross, but vv short-lived overloads.
    mega lord, Avon and jpk like this.
  7. Avon

    Avon pfm Member

    Wow … Thanks Martin! I’m now un-ignoramus-ed.

    I dug myself into a hole there. But bring us back to the subject of this thread. Why wouldn’t Naim just use a single centre tap and one bridge for the olive NAP250? That’d be cheaper than what they did use. Is it one size fits all, as Pete MB&D thinks? As colasblue said upthread, there must be a reason for this. JV was an experimenter and he didn’t “merely” follow logic.

    We need another experimenter. Please remind me, who started this thread?
  8. jpk

    jpk pfm Member

    Thanks for the confirmation, Martin!

    I did an experiment with my headline clone (it needs only one positive supply rail) on the same trafo with 2 secondary windings: the windings in parallel with 4 diodes vs. 2 diodes from centre tapped secondaries. The latter sounded better, very easy to hear in a few seconds, I was surprised! The schematic shows how I did that:


    With the centre tapped configuration only the two diodes in the 12 o'clock and 3 o'clock positions were populated (and as the headine clone does not have a negative rail the lower cap was not populated at all).

    But again I am not sure - in this case the amp can draw only half of the current, because only half of the time the diodes conduct, right...?
  9. colasblue

    colasblue pfm Member

    A thought (which may or may not be relevant) is that with the two secondaries arrangement, if you correctly arrange the secondaries, you can have +Vout and -Vout driven by the same part of the incoming AC waveform at the same time. You can't do that with any of the other options.

    That would have implications for how (the timing of) any distortion or asymmetry of the incoming AC would be reflected in the rectified outputs.
    martin clark likes this.
  10. Dowser

    Dowser Learning to bodge again..

    I did same years back with a dual-rail DIY ALWSR based Naim PSU - just adding a second transformer and only using 2 diode junctions was an easy audible improvement.

    I’d love to have confirmation of why - the above from @colasblue , or no diode junction in return path, or what?

    Maybe interesting, but trying the same arrangement when adding a dedicated xer/PSU for the 5v rail of 7220 in a Naim CDI kills the 7220 instantly (I tried twice just to make sure :D). I assume it does not like having 0v earth return connected to coils of transformer…but have not since tried a full-wave bridge setup instead to confirm.
  11. jpk

    jpk pfm Member

    Oh, good to know, I am going to recap/mod my CDI as well. The 317 for the 7220 gets quite hot, is that normal? (I might replace the 7220 with a PMD100 plus some glue logic and a better clock.)

    Sorry for being off topic, but I would be thankful to know what else to watch out in regard to the CDI...
  12. MJS

    MJS Technical Tinkerer

    Regarding the CDI - just offboard the analogue supply and apply Martin's CD2 tweaks to the regulators around the digital section. That's CDS territory.

    A Hicap can be very simply separated into two electrically separate +24v supplies. Wire them in series and you've got a dual polarity Hicap you can wire into a CDI or CD3 with an appropriate socket.
    booja30 and jpk like this.

Share This Page


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice