Announcement

Collapse
No announcement yet.

Linear PSUs with high output V

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Linear PSUs with high output V

    Hi everyone.
    What could be the reason linear PSU to output higher than its rated voltage?
    I have two 12V psus which output 18vDC ( 16vDC with load ) and 22vDC ( 18vDC with load ).
    Both 470uF filter caps are fine ( capacitance and ESR ).
    On the secondary transformer side I get 14 and 17 vAC.
    Diode bridge rectifiers.

    #2
    Re: Linear PSUs with high output V

    Feedback dead/gone... or pass transistor/mosfet shorted and therefore not reducing voltagge.

    Comment


      #3
      Re: Linear PSUs with high output V

      That is what to expect from linear power supply with just rectifiers and cap with no load.
      The capacitor will be charged to the peak Voltage.
      So output at no load will be (14VAC RMS x 1.41) - 1.4V drops of two diodes = 18.34V.
      You can also calculate the ripple Voltage to based on the capacitance, frequency and current draw.
      https://www.electronics-tutorials.ws/diode/diode_6.html

      https://cdn.badcaps-static.com/pdfs/...c73d63190d.pdf
      Notes: there is error on page 12, it should be 'Note that there are two diodes drops in conduction path(s)'
      Last edited by budm; 05-31-2019, 08:06 PM.
      Never stop learning
      Basic LCD TV and Monitor troubleshooting guides.
      http://www.badcaps.net/forum/showthr...956#post305956

      Voltage Regulator (LDO) testing:
      http://www.badcaps.net/forum/showthr...999#post300999

      Inverter testing using old CFL:
      http://www.badcaps.net/forum/showthr...er+testing+cfl

      Tear down pictures : Hit the ">" Show Albums and stories" on the left side
      http://s807.photobucket.com/user/budm/library/

      TV Factory reset codes listing:
      http://www.badcaps.net/forum/showthread.php?t=24809

      Comment


        #4
        Re: Linear PSUs with high output V

        Yes, need to make a distinction between linear and "unregulated" power supplies.

        Incidentally I've come across unregulated switch mode power supplies too...

        Comment


          #5
          Re: Linear PSUs with high output V

          I'm not that familiar with PSUs, so probably I did not give the correct description.
          Those PSUs consist only from transformer, rectifier and a filter cap.
          The load that I used was a router rated 12v@1A. Both PSUs are rated 12V@0.5-1A.
          So basically this is the best that PSU type can do?
          The funny thing is that one of that PSUs is the original one that came with a device with no internal voltage regulator...
          Last edited by madan1; 06-01-2019, 08:21 AM.

          Comment


            #6
            Re: Linear PSUs with high output V

            Well, it's good to learn something new every day.

            "Linear" implies that there's regulation via a "linear regulator". So in this case your wall wart with transformer, diodes, and filter capacitor has no regulation and "linear" or "linear regulator" would not be accurate.

            Sometimes I wish that these bulk unregulated DC supplies would disappear off the face of the Earth because of the requirements some newer devices have with incoming power (or at least they get relabeled as unregulated). As stated, due to the higher voltage some of these unregulated supplies can pump out, it can do damage to downstream devices if it can't handle it because it's making assumptions of the power supply.

            Comment


              #7
              Re: Linear PSUs with high output V

              https://mcitransformer.com/power-sup...ulated-linear/

              Agree: Unregulated DC supplies is a more fitting term.
              Last edited by budm; 06-01-2019, 01:05 PM.
              Never stop learning
              Basic LCD TV and Monitor troubleshooting guides.
              http://www.badcaps.net/forum/showthr...956#post305956

              Voltage Regulator (LDO) testing:
              http://www.badcaps.net/forum/showthr...999#post300999

              Inverter testing using old CFL:
              http://www.badcaps.net/forum/showthr...er+testing+cfl

              Tear down pictures : Hit the ">" Show Albums and stories" on the left side
              http://s807.photobucket.com/user/budm/library/

              TV Factory reset codes listing:
              http://www.badcaps.net/forum/showthread.php?t=24809

              Comment


                #8
                Re: Linear PSUs with high output V

                With nothing connected to such a power supply, the output of a transformer can be higher by some percentage, up to 5-15% higher.

                As you connect something to this unregulated power supply which consumes power and "loads" the transformer, the transformer output will be fairly close to its specified output voltage.
                So for example, a transformer that's supposed to output 12v AC may output 14-15v AC with nothing connected to it but as soon as you connect something to it, for example a 10-20w 12v incandescent bulb (from a halogen night stand lamp for example) you would see the voltage get closer to 12v AC.

                Here's some basics...

                When you rectify an AC voltage with a bridge rectifier, you get a voltage with a peak that's equal to Vdc peak = sqrt(2) x Vac - 2 x (voltage drop on one diode in the bridge rectifier)

                So if the transformer outputs 12v AC and you have a typical bridge rectifier with around 0.7v voltage drop on each diode inside the rectifier, then the peak DC voltage will be

                Vdc peak = 1.4142 x 12v - 2 x 0.7 = 15.5v DC

                This is peak dc voltage, the bridge rectifier only rectifies the AC voltage to DC, but the output is not smooth, 100-120 times a second the output goes between 0v and that peak voltage of around 15.5v
                To make the voltage smoother, a capacitor is added, which helps keep the voltage above a minimum level all the time.

                A formula that approximates the capacitance required to keep voltage above some minimum level is this :

                C = Current (A) / [ 2 x AC Frequency x ( Vdc peak - Vdc minimum desired) ]

                so in our fictional example, let's say we want minimum 11.5v and maximum 1A of current and we're in Bulgaria which has 50 Hz as mains frequency and our peak DC voltage is 15.5v so we put these numbers in the formula and we get:

                C = 1A / 2 x 50 x (15.5 - 11.5) = 1 / 100 x 4 = 1/400 = 0.0025 Farads or 2500 uF

                So we need at least 2500uF to keep the voltage always above 11.5v.

                If the capacitor is this value or higher, at 1A the output voltage will hover between 11.5v or a bit more... but not get close to that peak of 15.5v because the capacitor rarely has a chance to fill up close to that 15.5v value, it's constantly topping up and discharging.

                At lower currents, like let's say 0.5A, the capacitor will get a chance to periodically charge up to some higher level, so you could see for example the output voltage hover between 11.5v and 14v - over time, the average may be somewhere around 12.5-13v

                At very low currents, the capacitor will always have time to fully charge and you'll pretty much always see an output voltage close to that peak DC voltage.

                Knowing this, you can guess what happens with a much lower capacitance, like 470 uF you have...
                At low current values, this amount of capacitance is big enough that the capacitor can charge up with energy and you'll have on the output a high voltage, close to the peak dc voltage.
                The more current increases, the capacitor will have a harder time and you'll see the voltage fluctuate between a low value and a high value. At 1A, you may see something like 10v..14v up to 100 times a second.

                This is acceptable for a device that has voltage regulators inside which take this voltage and convert it to lower voltages like 3.3v or 5v - as long as the voltage is within a reasonable range, the regulators inside don't care.

                It's also acceptable for LED lamps, as they'll have led drivers that control the brightness and current going through the leds so they won't be damaged by this fluctuating voltage.

                However, if you need something that must have a very smooth and stable voltage on the output, it's best to add a regulator that would make sure the voltage is as stable as possible.

                The easiest is a linear regulator.
                Linear regulators take an input voltage (that must be higher than the output voltage by some amount) and automatically monitors the output current and voltage to keep it as steady as possible. The difference between input and output is thrown away as heat.

                So for example, taking a basic 7812 linear regulator... such a regulator must have an input voltage that's at least 1v higher than the output in order to output a steady 12v.

                So you must design choose that capacitor size so that the minimum voltage will always be above 13v or so... let's go with 13.5v to be sure...
                So C = 1A / 2 x 50 x (15.5v - 13.5v) = 1 / 200 = 5000 uF ... so let's say you use a 4700 uF capacitor which will guarantee a minimum above 13v.

                Now you can add the 7812 linear regulator and you get exactly 12v on the output all the time.

                For example let's say

                Comment


                  #9
                  Re: Linear PSUs with high output V

                  "Vdc peak = 1.4142 x 12v - 2 x 0.7 = 15.5v DC" without the filter cap, the output Voltage be average DC output.
                  https://www.electronics-tutorials.ws/diode/diode_6.html
                  Never stop learning
                  Basic LCD TV and Monitor troubleshooting guides.
                  http://www.badcaps.net/forum/showthr...956#post305956

                  Voltage Regulator (LDO) testing:
                  http://www.badcaps.net/forum/showthr...999#post300999

                  Inverter testing using old CFL:
                  http://www.badcaps.net/forum/showthr...er+testing+cfl

                  Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                  http://s807.photobucket.com/user/budm/library/

                  TV Factory reset codes listing:
                  http://www.badcaps.net/forum/showthread.php?t=24809

                  Comment

                  Working...
                  X