Paralleling Transformers

Discussion in 'General Electronics Chat' started by jackw19, Jan 24, 2009.

  1. jackw19

    Thread Starter New Member

    Jan 22, 2009
    I have six identical transformers (same mfr, same number) 240v in and (depending on wiring of primary coils) 20-55v out. The secondary consists of two identical--not center tapped--coils. The 20-55v value is per coil, that is the secondary coils are not connected in series. Each transformer weighs 23 pounds. I want to use these transformers in a combined configuration for a rectified welding power source (constant voltage). My questions concern the best way to do this.

    1) I have just assumed that the primaries should be connected in parallel, but what would happen if they were connected in series? Is there any advantage of one method over the other?

    2) Turning to the secondaries, consider first, please, the case of a single transformer. The two secondary coils could be paralleled and the output rectified using a 4 diode FWB. Would it also be possible to connect the end of one coil to the beginning of the other and, using this connection as a center tap, rectify with a 2-diode center-tapped FWB? Paralleling the secondaries seems superior because all the copper is in play every half cycle, but in the center-tapped configuration half the copper gets to rest (cool) every half cycle. Does one configuration have higher amperage capacity? Both would deliver the same voltage, correct?

    3a) However the secondaries of the individual transformers were configured, I would then parallel the outputs of the six transformers.

    3b) If a center-tapped FWB design was used, I assume I'd connect all six center taps, all six dotted secondary ends and all six un-dotted ends. Then on to the two bridge diodes.

    4) Would the mounting of the transformers have any effect on their magnetic relationship and regulation? My plan is to make a rack for them leaving an inch or so of cooling space all around each transformer.

    In terms of cooling several individual transformers might actually be better for a constant voltage power supply than one big transformer. Large transformers in some welding machines, even CV machines, sometimes have air gaps between winding layers for cooling; it seems that this would degrade coupling and regulation. Each of my six have the general shape of a shoe box, with just a hint of winding showing at each end--the windings are buried in core, which makes me think these transformers will have excellent regulation.

    One last question: If I were to use one of the secondary coils in a buck/boost configuration with the primary, that wouldn't compromise the isolation properties of the transformer, would it?

    Thanks in advance.
  2. jackw19

    Thread Starter New Member

    Jan 22, 2009
    Alberto, thanks very much for your reply.

    Why is it necessary to rectify the individual transformer outputs and parallel the DC? It'd be a lot easier just to have one diode bridge.
  3. eblc1388

    Senior Member

    Nov 28, 2008
    Transformer secondary voltage usually don't change by that much.

    To get different output voltage, secondary windings are combined in different configuration rather than changing the primary winding configuration.

    This led me to believe that each transformer has two 120V windings, which should be connected in parallel for your local voltage of 120V.

    If you connect the primary in series, it will not damage the transformer but then you are effectively connecting a 240V transformer to your 120V AC source. You can then get only half the rated VA rating from the transformer if I remembered correctly.
  4. jackw19

    Thread Starter New Member

    Jan 22, 2009
    The transformers are marked as "Signal 1063," which I believe is a non-catalog number. Anyway, I did find a document from a Canadian certification agency on the web that listed:


    240V, 50/60Hz
    Auto: 120V, 2A
    or 136V, 1.8A
    Sec 1: 16V, 25A
    Sec 2: 16V, 25A

    Also, isn't transformer convention for primaries to be labelled H and secondaries X? I've got terminals H1 thru H5 and X1 thru X4, and I can get 16 volts each from X1-X3 and X2-X4. So I think I've got the correct specs.

    With the secondaries in series I can get 32v at 25amps, which is just about right for this MIG welding project. It's a little disappointing that a 23 pound transformer is only rated for 25 amps, but it does have class H insulation and won't be used continuously for welding. With a 60% duty cycle I assume the rating would be significantly higher. And if I parallel the six I've got ...
  5. thingmaker3

    Retired Moderator

    May 16, 2005
    25 Amps at 32V is 875 VA. This requires a bare minimum of 4.73 square inches cross section of core. (This is based on the rule of thumb of 0.61*sqrt(VA).)

    The core runs a little over a quarter pound per cubic inch.

    Also consider the weight of the copper.