LM317 Question

Discussion in 'General Electronics Chat' started by Dr.killjoy, Jul 12, 2013.

  1. Dr.killjoy

    Thread Starter Well-Known Member

    Apr 28, 2013
    1,190
    156
    What is the purpose of the Tantalum capacitors in a lm317 circuit and does it matter what these values are ??????



    Thanks
     
  2. Shagas

    Active Member

    May 13, 2013
    802
    74
    which lm317 circuit?
     
    Dr.killjoy likes this.
  3. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    Have a look at this schematic that Bill_Marsden has drawn:

    [​IMG]

    The small capacitors at the input and output are there to prevent the LM317 from oscillating.

    Bertus
     
    Dr.killjoy likes this.
  4. Dr.killjoy

    Thread Starter Well-Known Member

    Apr 28, 2013
    1,190
    156
    Does the values matter or can I use any cap I have laying around ???
     
  5. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    In the standard configuration, there are values between 0.1 and 1 μF used at the output.
    At the input there is always a 0.1 μF capacitor.

    Bertus
     
    Dr.killjoy likes this.
  6. Dr.killjoy

    Thread Starter Well-Known Member

    Apr 28, 2013
    1,190
    156
    What the signification of using that certain value???
     
  7. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    The explanation can get highly technical - and a bit dry unless you're a mathematician, you can easily get lost in the theoretical maze of filters and poles!

    OTOH; most manufacturers publish reasonably clear application notes that explain the placement of decoupling capacitors, generally; the o.1uF caps should be as close to the regulator as possible - some regulator types have cautions with using multilayer ceramic chip capacitors - so *DO* read the appnote.
     
    Dr.killjoy likes this.
  8. LvW

    Active Member

    Jun 13, 2013
    674
    100
    For my opinion, the only purpose of C3 is to ensure (or improve) stability properties of the whole circuit. In this regard, the internal loss resistance of the capacitor is also important (the quality of the cap should be not "too good")
    Don`t forget, there is a feedback loop via R1 and R2.
     
    Dr.killjoy likes this.
  9. Dr.killjoy

    Thread Starter Well-Known Member

    Apr 28, 2013
    1,190
    156
    Thank you everyone for the help


    Personalian field I want to thank you for your input and it really help out alot .....
    http://www.youtube.com/watch?v=UW_XFGGTh0I
     
  10. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    If using a pot like in the circuit in post #3, it should have a capacitor from the ADJ pin to ground too.
     
    PackratKing likes this.
  11. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    Another point worth checking on the appnote - I've a dim & distant memory something like a cap on the Adj pin compromises ripple rejection.

    But I haven't used 3-terminal regulators for quite some time - like I said, *DO* read the appnote.
     
  12. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    No it's the opposite. Having a cap from Vadj to Gnd fixes the voltage at Vadj, and the result is higher closed loop Voltage gain (Vripple:1.25v), so it gives improvements in ripple rejection compared to the normal voltage gain from the resistor divider (Vripple:Vout).

    The main reason for the cap on the adjust pin is that pot's are not perfect, they crackle and get high resistance when being turned. A cap on the ADJ pin means this HF noise from the pot won't be amplified by the LM317 at a high power level.

    In a worst case, if the pot wiper goes open circuit (even for a very short time) the voltage between Vout and Vadj drops to zero, and the LM317 turns on full and presents the full PSU (Vin) voltage at the output!
     
    Last edited: Jul 14, 2013
  13. Dr.killjoy

    Thread Starter Well-Known Member

    Apr 28, 2013
    1,190
    156
    Another Question
    I plan to use a 12v wallwart and want to use the LM317 to 1.5v or lower but I know the LM317 needs a heat sink but I am not sure what size to use ??

    Here my math but not sure if it's right
    12v-1v*.6ma = 6.6 watts of heat dissipation

    But would one of these work
    [​IMG]

    OR
    [​IMG]
     
  14. WBahn

    Moderator

    Mar 31, 2012
    17,757
    4,800
    Your math is not right, but at least by including the units this is catchable.

    First, I'm going to assume you meant:

    (12v-1v)*.6ma

    If so, the result is 6.6mW, not 6.6W.

    If, instead, you meant 0.6A and not 0.6mA, then your result is correct.

    What is the thermal conductance, case-to-ambient, for that heat sink?
     
  15. BobTPH

    Active Member

    Jun 5, 2013
    783
    114
    If you are tying to get 1V out of an LM317, good luck. The min is 1.2V.

    Bob
     
Loading...