• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 852 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.

LM2585 CURRENT OUTPUT

John Greenhalgh
Prodigy 50 points
Other Parts Discussed in Thread: LM2587, LM2585, LM3478, LM3481, LM5118

I have implemented a boost circuit like the one in Fig 45 of the LM2585 spec sheet.  I am using a Li-on battery pack that supplies my circuit with 7.2VDC.  I am boosting it up to +12VDC to run my electronics.  I have two parts to my electronics package.  The first is a special controller we designed (which) has the 2585 on it and the second is a single board computer (SBC) that we buy.  When I run just our board it draws a little over 1A and everything is fine.  When we add the SBC it draws about 1A and the regulator drops regulation.  The output drops about 1/3 and we can see what we assume is the dreaded "Subharmonic Oscillation". We have changed the value of the inductor from 150uH to 100uH to 68uH and have seen no change.  I have read some discussion about the voltage on the Vcomp pin effecting the output current, but do not see any info on what and how that may effect it.  I am looking at the LM2587 which is a 5A part as a possible solution.  Can anyone help me with this problem.  I need to have a solid _12V @ 3A for this project.  My appreciation for any help on resolving this problem.

THANKS

JG

  • 0
    TI__Guru 64515 points

    Remember that a boost regulator is rated for its switch current and not the load current, like a buck.  The 3A rating is the maximum switch current and not the load current.  Assuming 100% effeciency and ignoring the ripple current, we have: 12V*3A=36W; 36W/7.2V = 5A.  That means you need a switch current limit greater than 5A; so I even the '87 might not do it.  You may want to look at a controller like the LM3481 or LM3478.

    FD

  • 0 in reply to Frank De Stasi
    Prodigy 50 points

    Frank,

    Thanks for the info and the lesson.  I always believe that school is never out.  I will look at the 3481 and 3478.

    JG

  • 0 in reply to John Greenhalgh
    Prodigy 50 points

    Frank,

    I have been looking at the LM3481 and need help some more info.  \

    First, with my need for 3A of output current, it appears that the external MOSFET is the part that

    will provide that level of current switching.  Is that correct?

    Next, adjusting the Rsen resistor will provide the feedback to the chip to do what it does to allow the switch controller to turn that MOSFET on/off to pump enough current through the inductor for this current level requirement.  Is that correct?  I have used the IRFP064N in other high current switching applications.  Will that work here?

    The last is the inductor selection.  In my use and calculations it appears that the actual value of the inductor does not effect the output current very much.  I have done the calculations and applied it to a circuit and it was comparatively small.  I would like to keep the switching frequency down to avoid EMI problems.  Do you have a suggestion for a 6+ amp device? 

    Thanks,

    JG

  • 0 in reply to John Greenhalgh
    TI__Guru 64515 points

    You are correct in your understanding.

    The IRFP064 might be a little large for this application.

    The best way to design this would be to go to Webench.  It will give you a full list of components.

    There are so many good inductors that it is had to select the best one; Wurth is a good vendor that you can get from DigiKey.

    The most important thing about the regulator is the PCB layout.  Try to follow the sugestions in the data sheet and in the app note for the EVM.

    FD

  • 0 in reply to Frank De Stasi
    Prodigy 50 points

    Off to Webench......................................

    JG

  • 0 in reply to John Greenhalgh
    Prodigy 50 points

    So, Frank...

    I went to Webench and ran my numbers.  There was no solution in the list of devices.  The note on the part you suggested said that it could not identify a compatible pair of FET and DIODE.  So I looked farther in the list and found the LM5118 "could" do what I needed.  So I went through the process with the LM5118 and Webench produced a suggested circuit diagram.  I studied it and decided that I would try it out.  I had a small PCB made with this circuit on and when I applied an input of 6VDC I did get a 11.86V output.  However, when I added a load requiring 2A, the voltage dropped off to less than half.  I then tried 1A and the result was the same.  So, I purchased an evaluation board for the LM5118 to see what I was missing or did not do.  Following the same test and loading procedure I goy almost the same results.

    I am a little dismayed and disappointed.  The eval board spec was for about Vin=5 to 75V and Vout=+12VDC at 3A out.  Never got there.  Decided to experiment a little and added a 5K POT in parallel with the larger of the two resistors in the FB circuit and eventually got near 9VDC out at about 2A. 

    What am I missing that makes my test with your board not work????????

    JG