Welcome to the NEW iPad Rehab Forum! Everyone is welcome to read posts and learn for free. If you'd like to post your own question on the forum, please click "register" above or go here: to enroll as a paid subscriber for $29/mo for unlimited access to our forum support and private signature failure section.
See more
See less

Strategy for reflowing/replacing ICs when CPU/NAND/RAM on opposite side of PCB

  • Filter
  • Time
  • Show
Clear All
new posts

  • Strategy for reflowing/replacing ICs when CPU/NAND/RAM on opposite side of PCB

    This is a real bug bear for me. I need a strategy for replacing PMIC when sensitive chips ie CPU, NAND, RAM are on the opposite side.
    Samsung G7 G930F
    This was a fresh board with no prior work tested fully working.

    I tried the following
    Quick hot air station PMIC 400 degrees 110L air large nozzle size 42s to reflow
    I held off the full blast of hot air for about 6 -9 secs working in the flux at a distance before coming in close and hard and giving PMIC a tiny nudge to confirm reflow at 42seconds and straight off with the hot air.

    Clean off the flux allow to cool and back into known good housing for a retest.
    Now it does not power up - draws 230mA via USB (no charging)
    DCPS 0mA before PTB climbs up to about 430mA before dropping back to about 300mA where it stays.

    I killed it.

    Prior to that I did the following on a couple of Samsung S7 edge boards

    PMIC reflow killed at 385 degrees 110 air 52s
    PMIC ok at 385 degrees 110 air 30s ok (no reflow)
    PMIC ok at 385 degrees 110 air 40s ok (no reflow)
    PMIC 390 degrees 43s 110 air ok (no reflow)
    PMIC 400 degrees 43 secs ok (NB it IS possible to desolder PMIC in this time)

    Any suggestions?

    Attached Files

  • #2
    You want to be in the 15s range to remove chips. “Reflow” is not a strategy the we ever use in iPhone repair, unless it is to confirm a newly placed chip is in fact adhered if we aren’t sure. In that case, if you are working on a chip opposite cpu where the board is also a giant heat sink, then you’d use leaded solder and again be working around 10-15s.


    • #3
      Thanks Jessa. 15s desoldering time - that's quite a feat though reflow was probably not the correct word for me to use in this instance.
      I was really referring to the point at which the unleaded solder becomes molten and the IC IS able to be removed. 10 -15 seconds is the objective so as a strategy does warming up the board on a preheater first help to achieve such a low desoldering time.


      • #4
        If your hot air station can’t get a chip off in 15-20 seconds then it is underpowered for this work, OR you have poor angle control. Practice on working boards.


        • #5
          OK thanks Jessa. I would say that the old Hakko FR-801s(with the manual knobs) are definitely under powered but the standard Quick 861DW hot air station with 500 degrees C and 120L air, which became so popular in recents years and I have, is not. Do you agree? So, with the Quick 861, if I practice till I get the angle of the hot air nozzle right I should be able to get the IC off in 15- 20s seconds around the 380 - 400 degrees celcius temperature operating range that I quoted or maybe even a little lower? I say this because you didn't say "Hey, that temperature is way to low!" ie Crank up the hot air to get it off is less time.

          "Practice on working boards" - I assume you meant non working boards?


          • #6
            I don't need the recent Quick DW881high power (1300W) hot air station?
            Attached Files


            • #7
              Quick 861DW in working order is fine. How do we know if yours is in working order or not? By whether or not you can get chips off in 15 seconds. Assume that the digital temperature reading on those is completely inaccurate (maybe it is maybe it isn’t, but they aren’t precision medical equipment, so let’s assume it is completely wrong). Then you go by feel.

              If you need to set yours at 500 degrees and 30% air to get a chip off in 15 seconds, then do it. Or 400 and 40%. Or 360 and 80%. It doesn’t matter—the end goal is always the same. Get the chips off in the range of 15-20 seconds.

              work on WORKING boards. If not, then how will you know if you took a chip off and put it back on and actually successfully completed the job?

              you need to work on technical proficiency before
              you try to fix anything else. Most of your posts are “I tried to fix something and whatever I did messed up other stuff because of technical proficiency problems”

              or come to Practical Board Repair School and get up to speed in 5 days!


              • #8

                I take it then that achieving the 15-20s IC removal time is in itself no guarantee of success - I just thought it would be reasonable to first achieve that on a non working board and then verify on a working board.
                Then, if it doesn't work, back to the drawing board to rework my technique?


                • #9
                  Okay sure--makes sense.


                  • #10
                    I remember a time earlier on in your You tube videos when you'd use the preheater as a strategy to avoid motherboard problems in "sensitive" situations eg Audio IC jumper repair on the iphone 7 causing the loss of baseband CPU function(requiring a reball) on the opposite side. Would it be correct to say that with experience, as your skills with hot air evolved, that the use of the preheater in many situations became no longer necessary.. even when sensitive ICs like the mainboard CPU are involved? For myself, reballing Baseband CPU is one thing but I wouldn't like to wager my chances when it's motherboard CPU.