• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 4 answers
  • Subscribers 62 subscribers
  • Views 321 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM117HV: Regulator Overheating when no load with little change in voltage

Steven Pavesio
Prodigy 110 points
Part Number: LM117HV

I am designing a circuit where the voltage is regulated from 15V to 9.15V. However, it overheats when there is no load for the regulator. And, even when I set the input close to the desired output input set at 13V. The regulator is overheating. R1 is set at around 100 ohms while R2 is set at 15 ohms. I don't have much experience with this and would appreciate it if someone can help explain what is happening. This is for LM117H

  • 0
    TI__Genius 17035 points

    Hi Steven,

    Please note that there must be at least a 5 mA load or the part will not be stable, as shown in the datasheet.  Also check your resistor values and the equation in the datasheet.  It sounds like you have R1 and R2 mixed up.

    If these items don't correct your issue, please provide more details.  Include a schematic of our application if you can.   What makes you think the part is overheating?  What are the symptoms?  What are the output and input doing?  Can you measure the currents along with the voltages?

    Kirby

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 110 points

    Thank you for mentioning the minimum load. This solves part of a bigger problem that I am trying to work with. I am trying to bring down 15V to around 3.3V.  R1 is 61 ohms and R2 is 100 ohms. The measured current in is 28.6 mA. The measured output voltage is 3.27 V. The load I have is running at 8.3 mA from a 382 ohm load. The input current is 28.6 mA. The current going from R2 to ground is 28.6 mA. The current going into the adjust pin is 36.6 uA.  I can tell the regulator is overheating since I can’t touch it without burning myself. Supposedly I could use a heat sink. But the design my company is using has no room for heatsinks. I’m looking around for answers not sure what to do

  • 0 in reply to Steven Pavesio
    TI__Genius 17035 points

    With this load current and voltage drop, you are dissipating around 340 mW.  Under worse case conditions with no heatsinking, you might see a temperature rise of greater than 60C.

    The thermal shutdown point of the part is around 170C.   When you the part goes into thermal shutdown, the output voltage and current will start to decrease.  

    Continuous operation of the part above 150C could impact the life of the part.

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 110 points

    Ok I see. Would a switching regulator make it easier to not have this overheating issue?

  • 0 in reply to Steven Pavesio
    Prodigy 110 points

    A switching regulator or a configuration that can drop the power required for the regulator to run and be enough to run the load without overheating the regulator or the resistors?

  • 0 in reply to Steven Pavesio
    TI__Genius 17035 points

    If your load is only 28 mA, then a switching regulator probably will not be more efficient.  

    The issue you have is heat sinking the package.  If you are running the part at a room temperature ambient, then the die temperature increase under the conditions you indicate should not be a problem in regards to the reliability of the part.   If you increase the load current or the ambient temperature then you can get into a condition where the junction temperature will be too high.

    You might consider a different package.  Do you need a hermetic package or will a plastic package work?  What grade of product do you need?   The LM117K STEEL is the very similar but comes in power package for heat sinking.

    If you would like to explore power product options, please visit our WeBench page that can assist you in picking the best product for your application.

     www.ti.com/.../webench-power-designer.html

  • 0 in reply to Kirby Kruckmeyer96848
    Prodigy 110 points

    Ultimately I am planning on using a military rated CFP since it is the only one available in stock. I am testing with commercial parts for the sake of testing costs and working with breadboard compatible components. The LM117GW/883 part is the part that would've been incorporated had it not been for the heating issue that is taking place now.

  • 0 in reply to Steven Pavesio
    TI__Genius 17035 points

    The LM117GW/883 comes in a 16 lead package so that all the no connect pins can be used for thermal dissipation.   If those pins are connected to the heat sink of our board, you should not have much die temperature rise.

    The LM117K/883 comes in a TO-3 package, which has even better thermal dissipation.