LM317 does not vary output voltage

Discussion in 'General Electronics Chat' started by beta_male, Apr 2, 2015.

  1. beta_male

    Thread Starter New Member

    Apr 2, 2015
    6
    0
    I have connected LM317 voltage regulator to get variable voltage as per the circuit in the datasheet. Input voltage is 24V from a power supply. However, the output voltage is 24V which does not vary even if R2 is varied. I have checked all connections, added decoupling caps to input and output, tried replacing R1 and R2, replaced LM317, replaced the breadboard; but I still cannot get output voltage to vary !!
    I read online that LM317 needs a minimum load current to work properly, so I added a LED with 330 ohm resistor at the output. The LED turned on, but it received 24V and the resistor became hot very quickly.

    This should be an extremely simple thing but I am stuck as to why it does not work ! I have R1=250 ohm and R2 is 1K trimpot which I have set at 750 ohm, so I should get 5 V as output, but I end up getting 24V.
     
  2. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    try a little more load.330 ohm in series with an led isnt much current.
     
  3. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    The LM317 requires that the input voltage exceed the output voltage by a certain amount called the "dropout voltage". This is typically in the range of 1-2 Volts depending on the required current. You can find so-called LDO (Low Drop Out) regulators that reduce this to 0.2-0.3 volts. What the regulator is doing is "following the input" without trying to regulate anything.

    OOPS: I thought you were putting 24 in and getting 24 out on purpose. Standby one.

    https://www.fairchildsemi.com/datasheets/LM/LM317.pdf

    Do you have the TO-220 package?
    Have you matched the pinout in the datasheet, note the orientation of the part to determine the location of pin 1. The schematic symbol might tempt you to conclude otherwise.
    1. Adj
    2. Output
    3. Input
     
    Last edited: Apr 2, 2015
  4. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,965
    743
    I would make R1 120 ohms ,this would give a constant current of 10mA, but i bet your lm317 is connected wrong, can you show pics of the built device.
     
  5. beta_male

    Thread Starter New Member

    Apr 2, 2015
    6
    0
    Dropout voltage has nothing to do with my problem.
    All I am trying to do is this -

    That video does not even have load connected to LM317. I also checked the breadboard for accidental shorts but the breadboard is fine. I changed LM317 two times and now I am on my 3rd one. Maybe I fried the first two, I'll try tomorrow with a different power supply because that is the only thing left to change.
     
  6. beta_male

    Thread Starter New Member

    Apr 2, 2015
    6
    0
    @Dodgydave
    Yes, that should be it. Unfortunately I left the circuit in the lab, so I can share pics tomorrow.

    The main reason I posted here was to see if there are any non obvious gotchas while dealing with LM317.
     
  7. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,766
    1,101
    If the ground connection to the 250 Ohm resistor is missing you'll get ~24V out from the 317. Check the breadboard connections....again.
     
  8. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    Yes, I did figure that out after I reread your original post. Sorry for the confusion. I think your problem has to do with the particular package you have and the correct identification of the pins. If you assume that the package and the schematic symbol are arranged the same I believe you will get the behavior you described.
     
  9. wayneh

    Expert

    Sep 9, 2010
    12,087
    3,026
    One thing you might do is reduce the supply voltage below 9V until you get the LM317 working. It'll survive being hooked up wrong at lower voltage but I'm guessing 24V can toast it.
     
  10. beta_male

    Thread Starter New Member

    Apr 2, 2015
    6
    0
    @Papabravo
    Yes, I was pretty careful and it is connected properly.
    Having mainly used 7805, it would have been easy to slip up and connect ground to LM317's pin 2 !! :( Thankfully that is not the case.
     
  11. beta_male

    Thread Starter New Member

    Apr 2, 2015
    6
    0
    Thanks, I'll try that first thing tomorrow.

    @all
    Thank you for all the replies. I guess I'll try tomorrow by rechecking connections. There do not seem to be any non obvious complexities in 317. I guess it is the usual "bang your head and make a blood sacrifice to please the electronics goddess" thing going on.
     
  12. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,644
    759
    Could you show the circuit?
     
  13. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    do not ground the tab on the lm317 when putting it on a heatsink. it is internally connected.
     
    Papabravo likes this.
  14. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    If the picture represents what the TS/OP is doing it is hard to see how a tab sticking up in the air could be grounded. It is still an excellent point however.
    I guess I'm mystified as to what is going on.
     
  15. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    If it's not the circuit, perhaps it's the breadboard?

    Some breadboards have side slots connected differently.
    Some have side rails shorted together, some have side rails not connected.

    You say that you have 24 volts at both the input and output of the LM317.

    Then it is not doing anything at all.
    It is shorted.

    Did you measure these voltages at the pins of the IC?
    What is the ground/ variable pin voltage?
    Does this voltage vary when you rotate the pot?

    Finally you have a 24 volt power supply.
    Have you connected it the correct way round?
     
    Last edited: Apr 2, 2015
  16. blueroomelectronics

    AAC Fanatic!

    Jul 22, 2007
    1,758
    98
    R1 needs to be 120ohm on the 317
     
  17. WBahn

    Moderator

    Mar 31, 2012
    17,715
    4,788
    For starters, a circuit schematic would be nice. How are we supposed to know which circuit in which datasheet you are referring to?

    Are you sure that your R1 is 250Ω? They don't make a 250Ω resistor, so perhaps you have read the code on it incorrectly and it is a much larger resistor than you think it is.
     
  18. dl324

    Distinguished Member

    Mar 30, 2015
    3,211
    619
    There is no set value for the adjust current set resistor. I routinely use 274 ohms. In power sensitive applications, the National Semi datasheet shows an application where it's 1.2K; but you have to satisfy the minimum load requirement.

    BR
    Dennis
     
  19. WBahn

    Moderator

    Mar 31, 2012
    17,715
    4,788
    To satisfy the minimum load requirement (which has a max value of 12mA in the Fairchild part) you need to keep R1 to 100Ω or less, though 330Ω will be sufficient for the typical case. You also need to allow for up to 0.1mA of adjust pin current, but that's more of an issue at higher output voltages.
     
  20. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    2,778
    1,211
    As a side note please be advised of the 'safe area' limitation (i.e. Vin vs Vout vs Iout) lest the regulator enter 'current fold-back mode' in which case its output will resemble that of a relaxation oscillator!:eek: This, of course, has no bearing upon your present problem...

    FWIW I add my 'voice' to the other respondents' assertion that your difficulty owes to a faulty or erroneous connection -- have you verified that the breadboard is in good electrical contact with each pin?

    Best regards
    HP
     
Loading...