Hot!Issue with PIC24 QFN-ML44 soldering for Deep Sleep app

Author
Mush
New Member
  • Total Posts : 11
  • Reward points : 0
  • Joined: 2011/06/20 05:26:08
  • Location: 0
  • Status: offline
2018/01/03 22:09:40 (permalink)
0

Issue with PIC24 QFN-ML44 soldering for Deep Sleep app

Design uses PIC24FJ128GA204 in Deep Sleep mode. Constantly having problem with current leakage between pins and/or Thermal pad. Around 100-600 uA. 
I use standard footprint design with manual and automatic soldering. All cases have leakage.
Only I could remove leakage if do manual soldering on pins only with no flux use (almost dry joints).
 
Have anyone succeed using QFN-44 package PIC24 chips in Deep Sleep mode (0.4 uA current use) for battery powered applications? What measures should be taken to stop leakage?
#1

12 Replies Related Threads

    qɥb
    Monolothic Member
    • Total Posts : 2966
    • Reward points : 0
    • Joined: 2017/09/09 05:07:30
    • Location: Jupiter
    • Status: online
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/01/04 03:27:10 (permalink)
    0
    Are the boards being washed after soldering, to remove flux?
     

    This forum is mis-configured so it only works correctly if you access it via https protocol.
    The Microchip website links to it using http protocol. Will they ever catch on?
    PicForum "it just works"
    #2
    Mush
    New Member
    • Total Posts : 11
    • Reward points : 0
    • Joined: 2011/06/20 05:26:08
    • Location: 0
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/01/04 04:59:00 (permalink)
    0
    Yes I tried to wash in ultrasonic with alcohol as well as with toothbrush. Issue is these washing cannot reach underneath QFN case.
    #3
    crosland
    Super Member
    • Total Posts : 1198
    • Reward points : 0
    • Joined: 2005/05/10 10:55:05
    • Location: Bucks, UK
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/01/04 05:05:52 (permalink)
    0
    What's the leakage through decoupling caps? You might find that it dominates.
    #4
    Mush
    New Member
    • Total Posts : 11
    • Reward points : 0
    • Joined: 2011/06/20 05:26:08
    • Location: 0
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/01/04 05:14:36 (permalink)
    0
    I use only ceramic caps (3V battery design) and they are low leakage. 
    As I mention above problem goes away if I do "dry joint" processor soldering.
    #5
    RISC
    Super Member
    • Total Posts : 5221
    • Reward points : 0
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/01/04 07:08:26 (permalink)
    0
    Hi,
    Have you checked carefully all you pins configuration during deep sleep ?
    100 to 600uA is too far away from spec.
    which technology of capacitor do you use for Vcap/Vddcore (low ESR cap is needed)
    I would personally do a trial with all I/O pins not connected (unpopulated PCB except caps) and programmed as outputs to make sure the SW sequence is OK
    Regards
     
    #6
    mbrowning
    Just a Member
    • Total Posts : 988
    • Reward points : 0
    • Joined: 2005/03/16 14:32:56
    • Location: Melbourne, FL
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/01/04 10:15:03 (permalink)
    5 (1)
    Active solder flux residue is a problem in many ways, and QFNs and LGAs are notoriously hard to clean. Have you tried reflow with no-clean flux? The flux is deactivated by heat. Not recommended for hand soldering as the heat is too uneven to guarantee all the flux is deactivated.
     
    Perhaps too much solder paste is being deposited. Maybe smaller openings in the stencil would help.

    Can't remember. I've slept since then - Mark
    #7
    Mush
    New Member
    • Total Posts : 11
    • Reward points : 0
    • Joined: 2011/06/20 05:26:08
    • Location: 0
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/02/13 14:47:40 (permalink)
    0
    Thanks mbrowning. 
    On new production sample prototypes (SMT soldering) was the same issue. Only was achieved target low leakage by re-soldering with completely removing paste from thermal pad under QFN (not soldering thermal pad). 
    #8
    mbrowning
    Just a Member
    • Total Posts : 988
    • Reward points : 0
    • Joined: 2005/03/16 14:32:56
    • Location: Melbourne, FL
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/02/13 18:55:24 (permalink)
    0
    Good to know. It would be a rare pic design that actually needed the added thermal conductivity of that exposed pad

    Can't remember. I've slept since then - Mark
    #9
    NKurzman
    A Guy on the Net
    • Total Posts : 16029
    • Reward points : 0
    • Joined: 2008/01/16 19:33:48
    • Location: 0
    • Status: online
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/02/13 21:57:20 (permalink)
    0
    It may not need the thermal properties, for the pad also provides mechanical support.
    #10
    Mush
    New Member
    • Total Posts : 11
    • Reward points : 0
    • Joined: 2011/06/20 05:26:08
    • Location: 0
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/02/14 03:25:58 (permalink)
    0
    Yes initially thermal pad was needed for mechanical strength.
    Now after removal I may need to glue chip on all edges.
    #11
    lichau
    New Member
    • Total Posts : 30
    • Reward points : 0
    • Joined: 2010/02/05 12:14:25
    • Location: 0
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/03/12 08:08:12 (permalink)
    0
    Can't help much with your leakage problem--only to say if you have that much board leakage current, you need to solve it regardless.
     
    But, I am using the same chip ( PIC24FJ128GA204 ) and cannot get it INTO Deep Sleep.  It goes into normal Sleep, but I CANNOT get it into deep sleep.
    I have tried various variations of the datasheet, etc. instruction sequence.  Right now, I am using:
     
    __asm__ volatile ("disi #3");
    __asm__ volatile ("bset DSCON, #15");
    __asm__ volatile ("bset DSCON, #15");
    __asm__ volatile ("pwrsav #0");
     
    But, when it wakes up, it executes the next instruction, not the reset vector.
     
    Appreciate a hint.
    Just, ONCE, I would like to implement the Microchip cookbook solution and have it actually work.
     
    Hints would be appreciated.
    #12
    JuliaDee
    Super Member
    • Total Posts : 172
    • Reward points : 0
    • Joined: 2009/05/04 13:51:40
    • Location: Connecticut
    • Status: offline
    Re: Issue with PIC24 QFN-ML44 soldering for Deep Sleep app 2018/03/21 17:13:22 (permalink)
    0
    Your Deep Sleep entry does not conform to either of the two examples in the data sheet. I would start by trying those, exactly as published.

    "Given one hour in which to chop down a tree, I'd spend the first 30 minutes sharpening my axe" - Abraham Lincoln
    #13
    Jump to:
    © 2018 APG vNext Commercial Version 4.5