Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Transformer for power supply

  1. Nov 16, 2011 #1
    Building a simple linear power supply for a project and need to spec out a transformer. I have no idea how they're built and used in the industry. I was instead handed a 120VAC input with a 24VAC output and 2A max step down transformer. I am content with it for the moment.

    But I'm looking to make my power supply universal. The question is, are there step down transformers available that can handle 100-240VAC 50-60HZ inputs? Or is this feature exclusive to switching power supplies?
  2. jcsd
  3. Nov 17, 2011 #2


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    yes there are transformers available BUT they need to be manually switched from one voltage range to another. Maybe you have seen gear, particularly test equip for the international market, that has a voltage selector switch on the back ?

    Basically, the primary side of the transformer has 2 windings on it, usually rated at ~ 110-120VAC. When switched in 110-120VAC mode just 1 winding is used, if 220-240VAC input is required, then the 2 windings are switched into series.

    I have never seen an auto switching transformer system as described above.

    In a switch mode PSU (SMPS) the stable output voltages are primarily determined by the feedback between the output and the input circuits. ie. as the input voltage rises, the DC output will also want to rise, but that will generate a higher error voltage in the feedback circuit back to the HV DC stages of the primary side.

    remember in a SMPS the incoming AC voltage 120, 240 etc is rectified to DC and then switched at ~ 10kHz or more. I'm not totally up with how the feedback works but I suspect its a form of PWM to control the switching of the primary osc. changing its duty cycle will control the voltage generated in the primary winding of the transformer.

  4. Nov 17, 2011 #3
    A transformer that can handle 240 volts in can also handle 120 volts. But then the 24 volt output would drop to 12 volts. You probably want a constant output voltage. But that will require a non-linear circuit to achieve.

    If you are going to use a transformer with a range of inputs, you need to not only make sure it can handle the highest voltage, but make sure it can handle the lowest frequency. If it can handle 50 Hz, it can handle 60 Hz. But if it is a 60 Hz design, when used on 50 Hz the voltage limit for saturation goes down by the same amount. So a transformer that can handle 240 volts at 60 Hz can only handle 200 volts at 50 Hz. Likewise, a transformer that can handle 230 volts at 50 Hz can probably handle 277 volts at 60 Hz (provided all other aspects can deal with that).

    A switched transformer with 2 windings should put the 2 windings in parallel when in the lower voltage mode. That would allow it to handle the higher current that needs to be drawn from the lower voltage to keep the same VA rating.

    Switch mode power supplies can generally handle a wide voltage range and produce the same output in DC by changing the switching durations as needed to keep a nearly constant voltage on the DC side.
  5. Nov 17, 2011 #4
    I don't know if any actually do this, but they could use a pair of switchers in opposite directions as the rectification bridge. Just switch one way in the up cycle, and the other way in the down cycle. I did see a circuit design once where multi-way switching was done like this to convert directly between 50 Hz and 60 Hz in three phase (this being possible because three phase has a constant level of available power). I suspect variable speed motor drives can do even more interesting stuff. It's also how I envisioned designing an inverter to convert DC to AC (at any desired frequency), with the switch timing varying to create the sine wave.
  6. Nov 17, 2011 #5


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    which part ? the rectification part or the feedback part ?

    The rectification part is pretty standard amongst SMPS units.
    As I said with the feed back I have only read a bit here and there and not totally conversant but PWM seems to be used in some designs.

    It really a subject I would like to know a lot more about. I started TV servicing in the mid - late 1970's when SMPS supplies were starting to become the rage ... in New Zealand and Australia we have the Philips K9 range of TV's, AWA's, Thorn's are a few others. All with their own take on SMPS design.
    The Philips K9 and KTV were amongst the easiest to work on... I remember well the days on using a 240V 100W lightglobe as a load for testing the 155VDC main rail from the supplies.
    And although I had a good handle on repairing them, I never really got into the theory of operation too deeply :)

  7. Nov 21, 2011 #6
    Did a little bit more reading on SMPS, complicated power electronics stuff which I haven't study too deeply in. Otherwise the small transformers they use is one of the neatest feature along with variable inputs. But it left me wondering, don't these use a chip for feedback control that requires power from elsewhere?

    Also what's a proper and safe way to construct and test a power supply design? Can everything be hooked up to a breadboard?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook