how efficiency increased in LM317 from LM7805

Discussion in 'General Electronics Chat' started by Mr.EU, Mar 9, 2014.

  1. Mr.EU

    Thread Starter New Member

    Mar 9, 2014
    4
    0
    LM317 and LM7805 both are linear voltage regulator. But in 7805 efficiency is very low but in 317 it is good. how is that achieved? :confused:
     
  2. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    I really doubt that. Actually 7805 has lower dropout voltage than LM317, so if properly desinged the 7805 should be a tiny bit more efficient.
    Could you post where you got your info from?
     
  3. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,348
    Hello,

    The LM7805 and LM317 are both linear regulators.
    The efficiency will be about the same.

    Bertus
     
  4. Mr.EU

    Thread Starter New Member

    Mar 9, 2014
    4
    0
    well, i used it, to get 5v from 24v;
    7805 really heated up but not 317, after 1hr running

    is both use same technique to regulate voltage?
     
  5. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,394
    1,606
    These devices are THREE terminal regulators. One pin is the input, another the output. Any current that does not flow from input to output will reduce the efficiency beyond the mathematical minimum of (Vout-Vin)*Iout

    As each device comes in several varieties we need to pick an apple to compare to an apple, so I looked at the TO-220 package rated for 1A output current on each. The 7805 has a nominal third terminal current of 5 to 8 mA. (It is labeled "Quiescent Current.") The LM317 had a third terminal current of just 46-100 uA (labeled "Adjustable pin Current").

    So the LM317 takes a smidgen less current to operate, but I would hesitate to call a 7.9 mA difference out of 1,000 mA anything but a marginal improvement.
     
    Mr.EU likes this.
  6. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,348
    Hello,

    How many current did you demand from the regulators?

    Did you use the decoupling capacitors on the LM7805?

    [​IMG]

    Without them the 7805 can oscillate and get heated.

    Bertus
     
  7. Mr.EU

    Thread Starter New Member

    Mar 9, 2014
    4
    0
    @Bertus
    yeah i used proper circuit configuration
     
  8. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    You are misinterpreting the meaning of the word "efficiency".
    Just because one device gets hotter than another is not a measure of efficiency.

    As others have already pointed out, every linear voltage regulator has to drop the excess voltage and this is lost as wasted heat. If your load consumes 100mA @5V from a 24V source, the voltage drop is 24V - 5V = 19V.
    The regulator has to dissipate 19V x 100mA = 1.9W
    This is the same for both regulators.

    What you have not told us is the load on each regulator.

    What you are most likely experiencing as excess heat is the "quiescent" current of the regulator with no load attached.
     
  9. Mr.EU

    Thread Starter New Member

    Mar 9, 2014
    4
    0
    thank you very much to all of you, for quick reply
    Thank you very much.
     
  10. John P

    AAC Fanatic!

    Oct 14, 2008
    1,634
    224
    This can't possibly be true. "Efficiency" of a linear regulator is a meaningless concept; its function is to throw energy away. When it heats up, it's just doing its job, and if the component is cold, it must be because it's not loaded. If you could feel any heat in either regulator with no load on the output, something was wrong.

    Edited to ask if you're aware that the LM317 and the LM7805 have different pinouts. If you put one on a power supply with the same connections as the other, you may indeed find some heat being generated!
     
    Last edited: Mar 9, 2014
  11. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    Your assumption is completely wrong. Efficiency is the same for the same VIN, VOUT, and load current.
     
  12. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    I think ErnieM already solved it?

    If the OP's "5v load" is something like a microcontroller (very likely) that draws 10mA, the 7805 will draw 10 mA at 24-5v (19v) = 190mW
    PLUS 24v * 7.9mA (GND pin current) = 189mW
    total dissipation 379mW.

    The LM317 will only draw 10mA from the 24v supply, as it's ADJ pin current is so low as to be negligible. So it's dissipation is 190mW, almost half the dissipation of the 7805 in that application.

    The difference in 190mW and 379mW in a TO-200 package with no heatsink is that the first will be cool to the touch but the second one at 379mW will be quite "warm".
     
Loading...