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

Converting 24vac to +5vDC (fail!)

  1. Jan 14, 2012 #1
    As a newbie, I've completed my first reasonably sophisticated design and have a working board...sort of. The power supply I designed 'bricks' the board (although my power LED glows brightly :) ). If I hand-wire in a known good +5v/GND source (isolating my on-board power supply) the rest of my board works perfectly.

    I need to drive the Atmega328 CPU based (originally Arduino) logic off +5v derived originally from a 24vac input source. Here is the simple schematic (including part numbers, values, etc.) for the power portion of my schematic (data sheet links below).

    PowerCircuit.jpg

    Actual voltage readings off my board are as follows...

    +24v: +24.31v
    +9v: +7.86v
    +5v: +4.30v

    The low +9v and +5v values confuse me. The data sheets (links below) often show "compensation capacitors", which I excluded out of ignorance (and the fact that they are referred to as "optional" in the MC7805CT data sheet).

    Any suggestions? I am so close to getting the project working, any help would be greatly appreciated.

    TIA,
    Mitch

    Here are links to the datasheets for the three significant parts.

    DF01M Bridge Rectifier
    LM7809ACT 9v Regulator
    MC7805CT 5v Regulator
     
  2. jcsd
  3. Jan 14, 2012 #2
    I suspect that I'm over-driving the +9v regulator. I think I misinterpreted the "Absolute Maximum Ratings" section of the data sheet to mean all variations can all accept 35v. If I look at the section specifically for the LM7809, I see a VI=15v.

    Could that cause the low output voltage?

    What would happen if you apply 24vdc to a voltage regulator rated for 15v VI.

    -Mitch
     
  4. Jan 14, 2012 #3
    Ahhh! (horrified scream)

    You *must* put a large capacitor right after the bridge! Use 100uF or larger electrolytic with the proper voltage rating. Without that you don't have steady DC power.

    Then, you *must* add the small compensation caps too. Without those the devices may oscillate.

    The absolute max is above 30 like you first thought.
     
    Last edited: Jan 14, 2012
  5. Jan 14, 2012 #4

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    you did connect the - side of bridge to the two regulator pin 2's, of course?

    more likely - where is your filter cap?

    since the AC line goes to zero every half cycle, so will bridge output.

    so the first regulator can only regulate during intervals when bridge output is greater than 9V , second one only when it is presented with > 5v.

    easy check for this is switch your DMM to AC volts.
    Output of each regulator should contain only a couple millivolts of AC, at most.
    Output of bridge should be in tens of millivolts AC.
    If you see like hundreds of millivolts you need to add filter caps.
    usually a few hundred microfarads of aluminum electrolytic is fine. should cost less than a dollar each.

    look up National's appnote "Introduction to power supplies".
    the filter cap acts like a surge tank reservoir on your well pump - to hold voltage up between AC line peaks.

    old jim
     
  6. Jan 14, 2012 #5
    Exactly the sort of posts I was hoping for. Thank you.

    I added a 100uF capacitor rated at 25v and my 4.3v, which is supposed to be 5v went to 4.98v. Then about 7 seconds later the capacitor exploded...violently...and loudly. Hmm...I knew I was at the limit since I'm at ~24.31vDC. I don't have a 100uF rated at higher than 25v so I just tried a 470uF rated at 35v. It works too, but I'm a little scared now to leave it plugged in for long.

    Any recommendation on the proper voltage rating on that capacitor?

    I will also add the compensation caps as well.

    Thanks so much.

    -Mitch
     
  7. Jan 14, 2012 #6
    See my responses below (inline).

    Thanks Jim.

     
  8. Jan 14, 2012 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    i see antiphon got it while i was typing. did a better job, too.

    your cap that blew was across the bridge, right?

    it shoulda charged to 1.414(that's sqroot of 2) X transformer voltage
    your 24 reading was unfiltered, it should increase by 41% when filtered.
    that's because it charges to AC peak.

    it is good practice to use electrolytics around 80% of their rating

    so for your 24 X 1.414 = 33.9 volts, a 42.9 volt rated cap would be perfect . so get a 50. the 35 will probably last for months or years if kept cool, but do it right.
     
  9. Jan 14, 2012 #8

    vk6kro

    User Avatar
    Science Advisor

    The maximum input voltage for that 9 volt regulator is 35 volts, too, so this design is a little too close to that, at 33.9 volts. You could be fairly certain that a small power surge will eventually destroy it.

    There is a regulator that can stand these sorts of voltages reliably. It is the TL783.
    This can cope with 125 volts input to output difference and give output voltages around 100 volts maximum.

    This is a great chip, but this circuit is really inefficient. Only about 15 % of the power used actually gets to the output because of the large voltage drop being achieved through heat loss.

    You can get BUCK regulators that are better than 80% efficient and deliver high currents if needed.
    These are ideally purchased as a complete module as the design is full of traps.

    Here is one for $5 delivered, (but there are lots of them on Ebay)
    http://www.ebay.com.au/itm/LM2596-D...534?pt=LH_DefaultDomain_0&hash=item2a155c5ad6
    This one will work up to 40 volts input and give up to 3 amps output.

    Another way would be to look for a small transformer that gave about 8 volts out and use that as your power source. A bridge would give about 10 volts out and you could run the 5 volt regulator off that.
     
  10. Jan 17, 2012 #9
    I like the idea of the TL783, and I've ordered a handful of them. I do have a couple of questions as I await their arrival...

    1) Based on the output voltage adjustment formula in the TL783 data sheet, if I use R1=100Ω and R2=470Ω that should give me a VO=+7.125 volts. Did I interpret the formula correctly?
    2) Can I assume I will still need a 5v regulator after the TL783? I'm basing this assumption on the fact that it will be tricky to hit +5v on the head with adjustment resistors.

    Thanks again for your postings.
    -Mitch
     
  11. Jan 17, 2012 #10

    vk6kro

    User Avatar
    Science Advisor

    There is a range of reference voltages given in the data sheet, so it would be necessary to make one of the resistors variable. You can get 10 turn potentiometers and these are great for setting voltages precisely.

    You would get about the voltage you calculated although my data sheet shows the median reference voltage as 1.27 volts so I got 7.239 volts.

    You might need to mount the regulator on a heatsink and this will require an insulator the mounting tab is not grounded. These are available.

    According to the data sheet, there should be at least 15 mA flowing from the regulator, so you can use the divider resistors to consume that current.

    You could just use the 783 to get 5 volts. The only reason you might use two regulators would be to share the dissipation if it became too much for one regulator.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Converting 24vac to +5vDC (fail!)
  1. 240AC to 5VDC help (Replies: 61)

Loading...