I'm getting too much voltage out of my AC/DC converter

Discussion in 'Power Electronics' started by Ed. M., Jul 23, 2019.

  1. Ed. M.

    Thread Starter New Member

    Nov 8, 2017
    21
    1
    Hi,

    Please look at the circuit below. I'm using a GBPC1208W bridge rectifier. The resistor and capacitor values are shown (C1 = 500uF, R1 = 1K, R2 = 300):

    acdcconverter.PNG

    My input AC voltage is 17V and I'm trying to get 12V DC out at M1. When I simulate this in PartSim it get pretty much exactly what I want. However, when I breadboard this circuit, my output DC voltage is the same as my input voltage. Any idea how this could be?

    Thanks.
    - Ed.
     
  2. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    9,933
    2,397
    AFAIK, Spice simulators always use peak voltage rather than RMS voltage for their SINE source.
    If 17V is the RMS voltage and PartSim says 12V at VMON it's telling lies. If 17V is the peak voltage then PartSim is correct.
    Peak voltage = RMS voltage x √2 for a sine wave.

    Edit: As soon as you draw additional load current from the VMON point the voltage will drop, because of R2.
     
  3. crutschow

    Expert

    Mar 14, 2008
    22,492
    6,588
    PartSim is showing the correct voltage at M1 for a Vac peak of 17V.

    As Alec noted, If you want the simulator input to be 17Vrms, than you need to have its peak voltage be √2 * 17V = 24Vpk.

    For 12Vdc output, you will then need to increase the R2/R1 ratio.
     
  4. Ed. M.

    Thread Starter New Member

    Nov 8, 2017
    21
    1
    Thanks for the replies.

    Interesting. Right now I'm using a variac through an isolation transformer. The variac is plugged in to my US 120V/60Hz outlet and is set at 17-18VAC. When I measure the voltage across the inputs to the bridge rectifier my multimeter is showing 17VAC. When I measure the voltage across R1 I'm seeing 17V DC. What am I doing wrong and how do I get the voltage down to 12V DC? Change R1 to be 1K? My goal is to plug the DC output line into Vin to power my Arduino.
     
  5. crutschow

    Expert

    Mar 14, 2008
    22,492
    6,588
    The R2-R1 resistor ratio is 1/1.3 = .769 which is close to the ratio been peak and RMS for a sine-wave, so that's why you are seeing 17Vdc for 17Vrms-ac.

    One factor you need to account for is the forward drop of the diodes, which is about 0.8V for junction diodes at your circuit load, or 1.6V for both diodes in the bridge (since two diodes are always in series).

    If you want 12Vdc for 17Vrms-ac then you need to reduce the voltage by 12 / (17*1.4 - 1.6) = 0.534.
    Changing R2 to 850Ω with R1=1kΩ will achieve that (as long as you draw no current from the output of the voltage divider).
     
  6. AlbertHall

    AAC Fanatic!

    Jun 4, 2014
    7,928
    1,955
    But if you want the arduino voltage to be 12V when it is drawing current then you should ditch the resistors and a 12V regulator instead (7812 for up to 1A).
     
  7. crutschow

    Expert

    Mar 14, 2008
    22,492
    6,588
    Yes, if you want the 12V to power an Arduino, the use a regulator as Albert mentioned.
    Either a 7812 or the common LM317 adjusted to 12V output should work fine for that purpose.
    They both need 2-3V headroom, so the input to the regulator from the rectifier bridge should be a minimum of 15Vdc (including any ripple).
    A 12.6Vac transformer with a bridge rectifier should give sufficient voltage for the regulators.
     
    Last edited: Jul 23, 2019
  8. crutschow

    Expert

    Mar 14, 2008
    22,492
    6,588
    Below is the simulation of a 12.6Vac transformer with an LM317 regulator set to generate a 12Vdc output.
    The output load shown is 500mA.

    Note that the transformer RMS current rating must be least 1.7 times the maximum DC output current (due to the high peak transformer current generated by the diode-capacitor supply, causing a high RMS current value).

    upload_2019-7-23_19-46-24.png
     
  9. Ed. M.

    Thread Starter New Member

    Nov 8, 2017
    21
    1
    Thanks this looks really good. Couple of questions:
    1. I'll have between 17-18V AC input to the bridge rectifier. Do I need to make any adjustments for that? Will the regulator get too hot?
    2. If I use a 7812, what changes would I need? Datasheet suggests using a 0.22uF input capacitor. I shouldn't need output resistors.
    3. Do I need a 2000uF input capacitor or will me 500uF one be sufficient?
    4. I assume from your diagram that the output capacitor is ceramic and the input one is electrolytic. Does it matter?

    Thanks again!
     
    Last edited: Jul 24, 2019
  10. crutschow

    Expert

    Mar 14, 2008
    22,492
    6,588
    Depends upon the load current.
    The input capacitor is fine.
    Yes, you don't need output resistors.
    That value determines the ripple voltage, which is also determined by the load current.
    Can be either.
     
  11. SamR

    Active Member

    Mar 19, 2019
    710
    250
    I was hoping someone would introduce a regulator.:D
     
  12. be80be

    AAC Fanatic!

    Jul 5, 2008
    1,847
    373
    Forget all that if your making a supply for a arduino don't use 12 volt if you read the tech spec
    it shows 7 - 12 the problem is at 12 the regulator gets hot I would stay around 9 volts dc
    there is no real heatsink to handle 12 volts.
    And some of the clones are even worst at handling the heat from dropping 12 volts to 5
    If you look at Input Output Differential you'll see you should really not go above 3 to 4 volts above
    the 5 volt output 2 would be ideal 3 or 4 don't use above 500 mA of power or the regulator will get hot.
    But then the arduino spec sheet tells you not to use VIN to supply more then a total of 500 mA of power.
    But just watch youtube and someone show one spitting out a amp but they don't show you the uno they burnt up LOL
     
    Last edited: Jul 24, 2019
  13. SamR

    Active Member

    Mar 19, 2019
    710
    250
    One of the nice features of the 317 is its adjustable output voltage and it doesn't have to be very precise because the Arduinos regulator is going to reduce it to 3.3 and 5V anyway. But yes not to "strain" the Arduinos regulator 7-9V is fine. So there are both fixed and the 317 variable to choose from. Or just get close by the resistor output method and let Arduino's regulator sweat the details. Listen to crutschow he is a good tutor.
     
  14. Ed. M.

    Thread Starter New Member

    Nov 8, 2017
    21
    1
    Well I've fouled something up. I breadboarded the circuit as follows:
    - C1 = 500uF
    - R1 = 1200 Ohm
    - R2 = 180 Ohm
    - C2 = 1uF
    My breadboard circuit output was showing ~9V DC which is perfect. I then soldered the components onto my PCB, hooked everything up, and now I'm not getting any output voltage. I've used my DMM to check all the connections and they appear to be correct. My output ground wire is connected to R1 and C2. C1 is holding a charge in the 40-45V range after I disconnect (I don't know if that voltage is expected or not) so I'm assuming the bridge is working. I don't know how to test the regulator.

    Any advice would be appreciated. Thanks as always; you all are really helpful.
     
  15. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    9,933
    2,397
    :eek: The Absolute Maximum input/output differential for the LM317 is specified as 40V, so it's likely the IC is fried if you're getting no output.
     
  16. Ed. M.

    Thread Starter New Member

    Nov 8, 2017
    21
    1
    This is probably a dumb question, but is there any way to determine what I could have done wrong? I don't want to just put another LM317 into my existing PCB and fry it too. The differential should only be 8V or so (17-9).
     
  17. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    9,933
    2,397
    Yes. Find out why you are getting such a high voltage at the input of the IC. Sounds like the variac is wrongly adjusted or the transformer is totally unsuitable and/or of poor quality and is putting out too high a voltage under low/zero load.
     
  18. SamR

    Active Member

    Mar 19, 2019
    710
    250
    Double-check the pinout for the 317. Those are pretty sturdy devices so I wouldn't expect it to be cooked. Also double check your ckt against the one CrutSchow gave you. The one you posted earlier was incorrect.
     
  19. KeepItSimpleStupid

    AAC Fanatic!

    Mar 4, 2014
    3,442
    632
    I don;t know how many times I've posted this http://www.hammondmfg.com/pdf/5c007.pdf link. Here is yet another. it shows the relationships of various AC to DC power supplies.

    it also might be time to mention is that most cheap handheld meters are average responding RMS reading for only a sine wave input within a certain frequency range. Many times, you need the RMS value which would be the same DC voltage that would dissipate the same power for a fixed resistor. the mathematical sine function used in EE is a form of v(t) = Vp*sin(ωt+θ); ω=2πf and is known as the Radian frequency. The sine fimtion varies from -1 to1; Vp is the 0 to peak value. RMS is known as Root Mean Square. It matematically flips the voltage so it's always positive and does a mathematical average. The result is always positive.
     
  20. Ed. M.

    Thread Starter New Member

    Nov 8, 2017
    21
    1
    It Works!! I double checked everything and my ground was connected correctly (missing a link between my capacitor negative and R1/C2. Now I'm getting a rock steady 9.7V DC out which hopefully will make my Arduino happy. Thanks everyone!
     
    SamR likes this.
Loading...