1
\$\begingroup\$

I have a transformer (12 0 12 5 amps) and I am using it for a variable bench power supply. Currently I am using it in full wave rectifier configuration and I'm getting around 40volt at no load after filter capacitors (2*25 volt series) so I was planning to use a Xl4015 based voltage buck converter but its max input is 36 volt only.
Then I thought of using the center tap in half wave configuration and getting around 20 volt at no load. In this case I was planning to use a boost converter but as we know when voltage increases current decreases. That means i will be getting 24 volt at around 2.5 amps. So now my question is is there any way I can modify my transformer to output 12 volts at 10 amp. I have read a couple of articles that mentioned connecting the secondary coils in parallel, something like this https://www.electronics-tutorials.ws/transformer/multiple-winding-transformers.html

Any views on this is appreciated enter image description here

\$\endgroup\$
4
  • \$\begingroup\$ Defined your desired IO specs V,I,P \$\endgroup\$ – Tony Stewart EE75 Aug 25 '20 at 17:10
  • \$\begingroup\$ Nit pickers on board. \$\endgroup\$ – Tony Stewart EE75 Aug 25 '20 at 17:10
  • \$\begingroup\$ It's absolutely OK to parallel the secondary as you show. Note it's ALSO OK to parallel the PRIMARY (your diagram above shows a transformer which is capabile of this). So if you paralleled the above diagram, you'd have 120V on the primary and 6V on the (paralleled) secondaries. It is critical the polarity marks are observed. (The little dots on the transformer). BUT do NOT parallel the secondaries if they are not "identical" windings.... I.e. don't parallel a 5V and 12V winding. \$\endgroup\$ – Kyle B Aug 25 '20 at 17:52
  • 1
    \$\begingroup\$ It's OK to parallel two separate identical secondary windings (with 4 terminals) as shown in the tutorial article. However the original question refers to a 12-0-12 V secondary. If this is a single centre-tapped secondary winding with 3 terminals the 2 12 V terminals must not be connected together as this will short out the secondary winding. \$\endgroup\$ – Graham Nye Aug 26 '20 at 1:23
0
\$\begingroup\$

If you want to modify the transformer in the fashion you describe, the coils will need to be connected in parallel, in the same phase. This means that, if the output only has three connectors, the center tap connection of one of the two output coils will need to be severed from the center tap pin or wire. This may or may not be possible, depending on the design of the transformer.

If the transformer is fully metal cased, it will have to be opened. Once the core and taped windings are exposed, you can try carefully peeling back the tape covering the windings in the area of the center tap wire or lug to see if you can distinguish where the two wires for each half of the coil come together, and then cut one of them.

If the transformer has solder type output lugs or PCB mount legs, you have a better chance that the two wires both go to the connector where you can get to them easily. If the transformer simply has a wire for the center tap coming directly out of the windings of the transformer, then there is a larger chance that the center tap common connection may be buried inside the windings somewhere, making this project likely not worth the trouble. If your transformer is torroidal, then you should have easy access to both sides of the center tap connection.

Once you have isolated one of the output coils, you will need to identify which of the two center tap wires belongs to which output coil of the transformer, and this can be done with a simple continuity test using a standard DMM. Once you have identified which set of now-independent wires belong to which coil, you will need to connect the severed center tap wire to the output of the non-severed coil's non-center tap wire, and connect the output of the severed wire coil's non center-tapped wire to the non-severed coil's center tap output, as per your screenshot.

At this point, the two now-independent coils will be wired in-phase parallel, and you will get half the previous output voltage at double the current capacity.

\$\endgroup\$
10
  • 1
    \$\begingroup\$ To check whether the windings are in phase, you can also use a scope, I think. \$\endgroup\$ – linuxfan says Reinstate Monica Aug 28 '20 at 4:49
  • 1
    \$\begingroup\$ Yes, but it wouldn't be anymore useful than using a volt meter. If you were to try to visualize the waveforms on a two channel oscilloscope connected to each of the two independent windings, then the grounds for the two probes would be effectively connecting the two windings together, and you would either see a full amplitude waveform, or a nearly zero volt waveform, when using the channel 1+2 function on the scope. The same measurement could be done by simply using an AC volt meter... \$\endgroup\$ – Hitek Aug 28 '20 at 5:46
  • 1
    \$\begingroup\$ It is good the idea of the voltmeter. But "the two probe grounds would connect the windings together"... well, they are already connected together... aren't we talking about a 3-tap secondary? The oscilloscope would be used in lieu of the bridge, it seems to me. \$\endgroup\$ – linuxfan says Reinstate Monica Aug 28 '20 at 5:50
  • 1
    \$\begingroup\$ This commentary was all based upon the fact that the output was a single center tapped coil, that had already been separated into two individual coils by cutting one coil wire away from the center tap, thus making the single center-tapped coil into two individual coils, which then may be need to be tested to make sure that the phasing is proper when reconnecting the now-individual coils to each other. Following the instructions I left in the answer should make all of this unnecessary, though, as a single center tapped coil will always have both halves in phase, but wired in series... \$\endgroup\$ – Hitek Aug 28 '20 at 5:56
  • 1
    \$\begingroup\$ @linuxfansaysReinstateMonica - Sorry, forgot to tag you in the above comment; didn't have room left for your long name anyways lol. Having to test the difference between the outside outputs of two center-connected coils would only be useful if the modification separated the two coils, and then they were wired independently with two wires each, and then the wires were somehow mixed up. If the person modifying the transformer remembered which wires went to the center tap and which ones didn't, then none of that would be necessary, as per my methodology in the answer... \$\endgroup\$ – Hitek Aug 28 '20 at 6:06
-1
\$\begingroup\$

Use a series pass FET design to limit above 36V before the caps. You are getting 40% over voltage from noload, but need to store 8T charge time constant at full load to get 10% ripple.

Unregulated ACDC cant utilize the 5A rating because the peak current is 8x so derate 25% Otherwise you need an isolated SMPS ACDC down converter for max power and efficiency.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ If you are a professional, comment otherwise leave it \$\endgroup\$ – Tony Stewart EE75 Aug 25 '20 at 16:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.