LM338 power supply not going below 7volts

Discussion in 'The Projects Forum' started by Anubhav Sharma, Apr 28, 2016.

  1. Anubhav Sharma

    Thread Starter Member

    Mar 21, 2016
    57
    2
    Hello Everyone,
    This is for the second time am posting here, after getting great help from experts in here.

    First of all i must tell you that am hobbyist and dont know much.

    I have built a LM338 power supply using 12-0-12 1Amp transformer and diode (IN5399) bridge rectifier.
    Also 2200uf 50v cap + 0.1uf ceramic cap, LM338 with nice big heat sink, 220ohms 1watt resistor (between adj pin and out pin), 10K pot (also tried 5k 1watt pot), 1K load resistor with LED and 0.1 ceramic cap.

    Everything is working fine except 1 thing. I am getting 30v max as output BUT cannot go below 7v.
    I turned pot to its very starting position but not showing below 7v.
    7 volts is the minimum output voltage showing.

    As per data sheet LM338 should produce 1.2v as minimum volts.

    Please see the attached diagram of same circuit.
    Power Supply1.jpg

    Please guide me where am wrong.
    Help me please.

    Thanks in advance.
     
  2. wayneh

    Expert

    Sep 9, 2010
    12,123
    3,048
    Try jumping around (shorting) the variable resistor to see what happens. The output voltage should go to ~1.25. If so, your variable resistor appears to be unable to go to ~0Ω.
     
  3. hp1729

    Well-Known Member

    Nov 23, 2015
    1,952
    219
    Only 7 V out? That amounts to a difference of 1,000 ohms. I agree, check and see if the pot goes all the way to zero ohms.
    With that wide a range in output voltage a set value of a resistor for the LED may not be the best idea, especially if you are going to spend a lot of time with a higher voltage out. If you want the LED to be an indicator of how high the voltage is your design is good. If you want a more consistent brightness a constant current circuit might be better than the fixed resistor. Not a bad design as it is.
     
    Last edited: Apr 28, 2016
  4. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,539
    1,251
    Check the pinout.

    ak
     
  5. hp1729

    Well-Known Member

    Nov 23, 2015
    1,952
    219
    Good possibility on the pinout. It is correct if he has the TO-220, going left to right facing the front, but if he has the TO-3 case pin 1 is different.
     
  6. Anubhav Sharma

    Thread Starter Member

    Mar 21, 2016
    57
    2
    I got your point and i understand this is a bad design, I'll change it once all working well.
    I checked pot... its going to 0 ohms and max to 4.99k-5 as its a 5k and same with 10K pot
     
  7. Anubhav Sharma

    Thread Starter Member

    Mar 21, 2016
    57
    2
    Shorting ? Its already showing 0 ohms at minimum and 5k at max. I checked with multi meter.
    As i mentioned am new to this field, can you please tell me What pins of pot do i have to short. ( I have already joined adj pin and right pin )
     
  8. Anubhav Sharma

    Thread Starter Member

    Mar 21, 2016
    57
    2
    am using TO-220 and checked pin out.. middle one is out, right one is IN and left one is ADJ ( facing front )
     
  9. Anubhav Sharma

    Thread Starter Member

    Mar 21, 2016
    57
    2
    I am using TO-220, Pins are connected fine ( facing front )
     
  10. wayneh

    Expert

    Sep 9, 2010
    12,123
    3,048
    So with the variable resistor completely bypassed with a wire ("shorted", as I said, as if it is not there), you still get 7V at the output? I would begin to wonder about the IC.

    But let's stick to simple things first. How certain are you of the 220Ω resistor? A 2200Ω resistor might cause your problem.

    Post a picture of your build. Another set of eyes is often useful.
     
    Anubhav Sharma likes this.
  11. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,052
    3,817
    If wayneh can't see the problem, he'll put these guys to work - laser eyes solve everything...
    upload_2016-4-28_14-1-34.png
     
    wayneh likes this.
  12. wayneh

    Expert

    Sep 9, 2010
    12,123
    3,048
    They dim down a little after feeding time. :p
     
  13. Anubhav Sharma

    Thread Starter Member

    Mar 21, 2016
    57
    2
    I think its IC.... I have updated circuit, replaced LM338 with LM317 and BINGO.
    its wiorking fine. 1.4 to 31v

    Is there any way to to get 1.4v from that bad LM338 i mean if its actually bad. may be its working, beacuse i ordered it from Texas Instruments directly.
     
  14. wayneh

    Expert

    Sep 9, 2010
    12,123
    3,048
    Two things to try:
    1. A larger load on the output, or
    2. A larger capacitor on the output. You might have ringing.​

    If neither of those work, I'd say you have a bad part. Ask TI to replace it.
     
  15. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,052
    3,817
    Check that your pot actually goes to zero ohms when turned fully.

    Replace the 220 Ohm resistor with a 120 Ohm resistor. The datasheet says voltage will go higher than expected if a minium load is not maintained. The table on the datasheet says a minimum load of 3.5mA is TYPICALLY needed. Unfortunately, you don't always get a Typical regulator, you get what you get (a quote from AudioGuru). So, the datasheet also says that the maximum value for the minimum load (confusing?) will be 10 mA. Therefore, you need a 1.25V/0.01 amp = 120 ohm load. Note that that load starts to change as you turn your potentiometer. Do the calculations to make sure you always have at least 10 mA load, no matter where you turn the pot (or place the load between out and ground directly).
     
  16. hp1729

    Well-Known Member

    Nov 23, 2015
    1,952
    219
    Then I am at a loss to an idea why the output won't go down all the way. On paper the design looks good. The LED, as it is, is a quick indicator of output voltage. A brighter LED indicates a higher output voltage. 30 mA out may make a bright LED, but will work such as it is.
     
Loading...