Connect with us

Troublesome powersupply

Discussion in 'General Electronics Discussion' started by tarry, Jun 26, 2014.

Scroll to continue with content
  1. tarry

    tarry

    6
    0
    Jun 26, 2014
    Hello all!

    I have built a 3A power supply from a kit (documentation here: http://www.electrokit.com/productFile/download/383), and for small loads it works OK. However, I am having trouble getting it to deliver even 1A without a drop in voltage. The configuration is a bit special, in that I am not using its built-in rectifier bridge, but instead feed it 13.8 V DC from another power supply that I have. (The ultimate goal of the project is to have 10 student sub-supplies fed by a main 13.8 volt 30 A DC power supply). I have checked the feeder supply; its output voltage does not drop while the load on the "child" supply approaches 1A. Only the child supply voltage drops.

    There are 5 1-ohm resistors in parallel in the child supply between vcc from the rectifying bridge and the collector of the main power transistor. These resistors will - if power-supply knowledge serves - have a small voltage across them, which increases as current through the main transistor increases, and which will trigger voltage limiting as the power approaches allowable maximum for the device.

    The thing is, voltage cutting kicks in too early. As said earlier, the power supply should be good for 3A, so dropping the voltage even before 1A seems a bit over-cautious. Would it be a good idea to solder in a few more 1 ohm resistors in parallel with those already in place? Advice and/or suggestions are very welcome.
     
  2. Arouse1973

    Arouse1973 Adam

    5,164
    1,080
    Dec 18, 2013
    What voltage is it dropping to? Where are you connecting the d.c input to. You need to connect it after the bridge rectifier.
    Adam
     
  3. Ehsan

    Ehsan

    100
    1
    Jun 12, 2014
    What is exactly the purchased transformer rating (VI) ?
     
  4. davenn

    davenn Moderator

    13,560
    1,856
    Sep 5, 2009
    Presumably it was part of the original kit. But at this stage not a concern, as he's experimenting powering it from another DC PSU

    to Tarry .... is that all the documentation that came with the kit ?
    as in there isn't even a parts list telling you/us what the component values are ?

    it looks like its based around the 723 reg chip, a VERY OLD chip, haven't used one of those in 20 years and never did like them

    cheers
    Dave
     
    Ehsan likes this.
  5. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,362
    2,758
    Jan 21, 2010
    Remove a couple of the 1 ohm resistors and see if the voltage drop occurs at a lower current. Alternatively, measure the voltage across these resistors when the voltage begins to drop.

    I agree that the 723 is an ancient, and difficult to use chip in relation to modern regulators. In addition they're really easy to damage. I last used one about 35 years ago.
     
  6. tarry

    tarry

    6
    0
    Jun 26, 2014
    Thanks for the responses so far. I did connect it before the rectifier, which seemingly costs me around 0.7 V, but that is a constant voltage drop, independent of the load, isn't it? Yes, this is all the documentation. I suppose I could look up the component values on the web. The resistor and capacitor values are printed on the PCB; the other components can be discerned by their looks. I will take a photograph of an unassembled PCB later today, and post it here.

    I do not use a transformer (other than that which is part of the feeder supply). The kit does not come with a transformer either, this is an external component which will have to be purchased separately except as stated, I am ultimately going to use a 30 A DC power supply as a power source.

    The voltage is dropping from about 12 V to 10 V when the load is between 0.8 and 0.9 A

    The input voltage from the feeder supply is 13.8 V, in order to simulate the constant-voltage power supply that will ultimately be used.
     
  7. davenn

    davenn Moderator

    13,560
    1,856
    Sep 5, 2009
    well if its a bridge rectif. as indicated, the drop will be through 2 diodes so around 1.4V
    doing that is OK .... it solves any issues of accidentally reversing the input DC supply, as the output from the bridge will still have correct polarity


    D
     
  8. OLIVE2222

    OLIVE2222

    690
    25
    Oct 2, 2011
    The whole setup will not allow a wide voltage range for the "child supplies". I don't remember exactly the (indeed old) LM723 specs but Vin should probably be around 3 volts higher than Vout.
    From 13.8v less 1.4 of the diodes less 3 volts you will have a Max output of less than 10v. Even if you only add 1 diode at the input you will not be be able to reach the popular 12V.
     
  9. davenn

    davenn Moderator

    13,560
    1,856
    Sep 5, 2009

    Yup exactly and per the datasheet, the input needs to be 3V higher than the required output

    so with that in mind, tarry, can you supply it with around 16 - 19V input ?

    cheers
    Dave
     
  10. davenn

    davenn Moderator

    13,560
    1,856
    Sep 5, 2009
    Tarry, .....

    you would be much better off scrapping that PSU and all its unnecessary complexity and going for
    a bunch of cheap boost/buck converter style of power supply modules ... like one of these ....

    http://www.ebay.com.au/itm/4-38V-to...Electrical_Test_Equipment&hash=item417ad78e20

    or one of these with a built in volt meter...

    http://www.ebay.com.au/itm/DC-LM259...Electrical_Test_Equipment&hash=item27e11300d5

    you are going to have much less stress sorting it all out for a very minimal cost

    cheers
    Dave
     
    Last edited: Jun 28, 2014
  11. tarry

    tarry

    6
    0
    Jun 26, 2014
    Good comments here - thanks for ideas and information. What I cannot understand is what you are saying about the bridge rectifier sending the power through 2 diodes. As far as I can see, it passes through only one. A bridge rectifier is often pictured as a diamond with AC entering at opposite corners, passing its power alternatingly to each adjacent corner where DC is provided. With DC entering as input, this is not changed, other than that only 2 of the diodes will ever have current passing through them, and the other 2 will always be blocking (unless you reverse polarity on the inputs at any time)
     
  12. Arouse1973

    Arouse1973 Adam

    5,164
    1,080
    Dec 18, 2013
    Yes I agree Olive2222 . I would remove the diodes also. The output transistor will also play a major part in the reduction of the output voltage. The output transistor will have a base emitter voltage drop which will vary with output current.

    If the OP could supply the part numbers of the components we could work it out properly.
    Adam
     
  13. Arouse1973

    Arouse1973 Adam

    5,164
    1,080
    Dec 18, 2013
    Depends where you have connected the input. If you just connected the d.c supply where the transformer used to be then you will have 2 diode drops. You might have just connected the 0V to the 0V of the circuit board and the positive to anode of the input diode on the bridge, then you would have only one diode drop.
    Adam
     
    KrisBlueNZ likes this.
  14. tarry

    tarry

    6
    0
    Jun 26, 2014
    Here is the pristine PCB with - hopefully - the required component information. I added a mirrored image of the backside.
     

    Attached Files:

  15. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
  16. Arouse1973

    Arouse1973 Adam

    5,164
    1,080
    Dec 18, 2013
    Wow Kris, nice one
     
  17. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    Thanks Adam... the schematic was in the PDF linked in the first post; I just added the values.
     
  18. Arouse1973

    Arouse1973 Adam

    5,164
    1,080
    Dec 18, 2013
    Good work mate. I hate circuits without any values, I am sort of lost without them.
     
  19. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    Same here. This design had it kind of backwards - the schematic has circuit references but no component values (should have both), and the PCB overlay has component values but no circuit references (should have references only)!
     
  20. davenn

    davenn Moderator

    13,560
    1,856
    Sep 5, 2009

    You contradicted yourself and answered your own question in the same post !! :)

    That's what we said 2 diodes :)

    Dave
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-