Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 65 subscribers
  • Views 2579 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.

LM317 Maximum Output Current

Tim Sobering
Intellectual 860 points
Other Parts Discussed in Thread: LM317, TL783, PTH12040W

A student came to me with a spec from the LM317 datasheet (SLVS044U) and I cannot answer his question.  In the Electrical Characteristics table for Maximum Output Current, with Vi - Vo <= 40V, Pd < Pmax, and Tj  = 25C, the typical output current is 0.4A.  I'm confused as to this spec.  Does the maximum output current really drop when the differential voltage across the device exceeds 15V?  I have always assumed that as long as I provide sufficient heatsinking to keep the junction temperature below the maximum (preferably well below) and do not exceed the max input voltage, the regulator will provide up to the max output current.  The inclusion of Tj = 25C is confusing as well as it's unlikely the junction temperature would be at room temperature for all but the lightest of loads. 

The student pointed out the TL783 data sheet which has a similar reduction in peak output current, but the spec includes a time parameter as does the graph on page 4 of that TL783 data sheet.  Is this spec a pulsed current limit?

Could you shed some light on this for us.  The students in the Electronics Club are designing some high output current supplies for use in 3D printing and also some high voltage applications.

  • 0
    Mastermind 7395 points

    First of all, if you have big difference between vin and vout, these days you almost MUST use DC/DC, and not linear supplies.

    Even if you add heatsink, you improve the thermal Rca. You can never improve the thermal Rjc, so there will always be a limit to Pd. For higher vin-vout difference, you can expect lower output current.

    The Tj there that you mention... I don't have an answer for that.

  • 0 in reply to Avi Chami1
    Intellectual 860 points

    Thanks Albert.  But consider that LM317K package is 2 C/W for junction to case.  Assuming a 40V I/O differential the K package is limited to 0.4A typ., which is 16W of dissipation.  Yes, you would need a good heatsink, but a thermal resistance < 3 C/W for a heatsink is not difficult so that would yield a 80 C rise over ambient.  This is within the 150C max junction temperature assuming a 30C ambient (how reasonable that assumption is depends on the overall packaging). 

    I think the question still stands...is this simply a thermal limit involving Tjmax or is there something else in the device design that limits the current when the differential is 40V?

    Also, I'll respectfully quibble as to using a switcher.  I understand your point and a switcher is much more efficient, but there are situations where linear regulators are needed due to other requirements.  I was somewhat unfair as I didn't discuss all of the requirements for the particular supply they want to design.  They are trying to use one design topology to meet a number of different needs for the club.  The high differential voltage case is actually an outlier and the extreme.  The 317 might not be the right solution for the application.

  • 0
    Intellectual 530 points

    If I might suggest

    using an array of LM317 just connect each Adj pins and use 1 pot,,, also place a 2 Watt 10 OHM resister connecting each output. 

    I used 8 LM350 to power my 40 watt laser. Then I found Ti's PTH12040W modules. they rock! 

    It's just a thought  

  • 0 in reply to Jeff Lampe
    Mastermind 7395 points

    Hi Tim, 

    Tim, your question indeed still stands. I hope an SVA engineer will come to the rescue.

  • 0 in reply to Jeff Lampe
    Intellectual 860 points

    Jeff...that's not a bad suggestion and it meshes with the modularity we are trying to achieve in shoehorning these applications into a single solution.

    My guess is that when they see the PTH12040 there will be some excitement. It's not what we are trying to do but I'm certain a few of the students will want to play with that module.  They will find a use for it.  Thanks!

    Tim

  • 0 in reply to Avi Chami1
    Intellectual 860 points

    Albert...I'm sure they will.  TI is one of the better ones as to responding to the technical stuff.  I'm enjoying the responses so far and I'm guessing there will be a lot of things that the students can learn from.  Thanks again!

    Tim

  • 0 in reply to Tim Sobering
    Intellectual 530 points

    Thank you Tim.

    The PTH12040W works very well with a 500 Watt computer power supply.

    I just had a second thought

    I believe its possible to use the LM317 to set the Voltage and use the PTH to increase the Amperage,,,  Linking the two outputs,, to one output trace with high power Diodes so their is no back feeding.

    Even better use the LM317 to turn on and set a 2n3055 it will just need to be highly filtered.,,,  that should get you up to 15 amp up to 40 volt.

     

     

  • 0
    TI__Guru*** 149366 points

    Tim,

    The LM317 device protection includes a maximum current limit that is more than 1.5 Amp.
    It also has a maximum power protection that is more than 20 watts.

    Notice the two Zener diodes (about 7V each) that add the the current sense which lowers the maximum current.
    In general 25C specifications are production tested for each device and full range specifications are ensured based on design characterization.

     

     

    The TL783 figure 1 and figure 2 data are pulsed for 1mS or 30mS. Tj increase more for the 30mS test and there is some current limit reduction due to temperature.