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.

Soldering an F28M35 by hand

I am hand soldering a 144 pin Concerto part to a PCB; any advice on what I can do with the PowerPad?

I need to replace the headache known as REV 0 Concerto part with a REV B. I want to do this for all of my controlCards and for a board or two.  I do not have a reflow machine but I have a tech here with a super steady hand.

Besides, "don't do it", what is my best path forward? This document SLMA002 does not recommend running the processor without soldering the pad - What if I connected a heat sink instead?

Any advice is appreciated.

  • Apply Solder paste to the power pad with a syringe. Solder the pins as usual using a fine pitch iron . Reflow the device utilizing a hot air source such as a Hakko fr801

  • Do you think I can do this without messing with the other parts of the board?  Our tech almost cooked the surrounding components getting the MCU off in the first place.

    What about fastening some heatsinks on the surface of the packaging with thermal epoxy?  I know it is ghetto, but this is only for a test board that will be used in the lab.

  • William,

    I don't think the main issue will be from heat if you don't solder down the PowerPAD.  The main problem is that this MCU relies completely on the PowerPAD for grounding and therefore won't work properly if it's not soldered down relatively well. 

    Note that I believe that getting the MCU on should be easier than getting it off.  Using leaded solder paste vs non-leaded solder paste on the PowerPAD's connection may make things easier (if this is an option).  If you do this make sure the pad is cleaned thoroughly.  The board will obviously also no longer be RoHS compliant.  :)

    Tom has actually done exactly what he has described on the F28M35x cCARD a few times in our lab.


    Thank you,
    Brett

  • If you're not sure you can accomplish this without damaging the board you can protect the surrounding area with polyamide tape.

  • Great folks, this is all good information!

    I will pass this on to our tech.

    ___________

    When will you be putting newer Si revisions on cCards?  I can not explain to you what an epic PITA the Rev 0 has been to my design.  Ok, well maybe I can...

    *inhale*

    Can you guess which pins the 2 MOST important GPIO in the entire freaking project were connected to?  

    Also, our internal development tool required the M3 to be able to read the state of the pins assigned to the C28.

    Also, GPIOTOGGLE in port C (I think Brett actually helped me with this)

    And finally, I spent so long on that corrupted DWARF issue with CCS 5.3 that I thought I was on the set of Game of Thrones.

    *gasp*

    Not gonna lie, I sent quite a bit of curse words in the direction of Dallas.

  • BTW, we love lead solder here, can you tell?

    :D

  • William,

    Depending on your location, Dallas was probably in roughly the right direction, but most of the C2000 team is actually based in Houston(in case you want to more finely directly your curses in the future) :)

  • William,

    My best guess based on MCU availability dates and existing inventory of kits is (please don't hold us too firmly to this - I'm giving you an actual estimate, not a padded estimate):

    DOCK kits (w/ RevA/B F28M35x MCUs): ~end of July
    CNCD kits (w/ RevA/B F28M35x MCUs): ~beginning of September

    In particular, you should be looking for TMDS_ kits as opposed to TMDX_ kits.


    Thank you,
    Brett

  • William, for what it is worth - we are also looking forward to the new Rev of silicon being widely available.