Is it this your first experience of C-CGA RWK ?
In order to approach a RWK of this kind of pakage it requires a capable process and clear operating procedure.
In the past, the company who I use to work for, we rwkd plenty of this kind of CCGA. Earliest version with WIRE Column Type, CAST Column Type and last tech CLASP column type. The earliest column pitch was 1,27 mm, the one like yours it should be 1,00 mm column pitch and i suppose CLASP column attach Type.
The last C-CGA size we use to place and rwk was 52,5 x 52,5 mm square, more than 2650 column, 1,00 mm pitch Clasp type.
You need a very robust process: - to know the technology of this pakage - to know the technology of multilayers PCB (your's should be at least 24 layers), because, for istance, Internal plane Separtions or delaminations it is a matter to keep strongly under consideration, when setting : pre-bake time , pre-heat and reflow profiles of rework machine.
- RWK machine used had bottom (under board) heaters (hot air obtained by 2 powerfull plates 1500W plus 1500W. - Heating media: Hot Gas N2 (Top Nozzle) - Solder paste: Water Soluble (Type 3) - PCB finisch OSP - PCB thickness from 2,4 mm up to 3,4 mm
Your stencil specs seams to be OK, but what you need to do is make sure for the right voulme of paste released on pads. We run several SPC by measuring paste volume and registration. But at the end we found reliable enough to visual inspect by microscope (40X) and only randomly measure also the volume.
Dedicated and Very skilled RWK Operator is the keey point for a succesfull and reliable CCGA rwk. I have seen some of them, after long time operating in rwork, be able to place CCGA on paste by hands and obtain reliable results same as placing CCGA by machine.
Pad cleaning and dressing operation after bad CCGA removal: Make sure do not stay for long time with hot solder tip (hot gas + vac) on pads in order to avoid copper dissolution problem. Same when redressing pads.( IMC grow very fast and you could get copper removed from pads very quickly)
Procedure ( aproximatelly): - preheat board (remove moisture,also shortening the rwk cycle time - romove bad CCGA - pad cleaning/redressing - By mini stencil deposit new paste - remove gently the ministencil - chec/inspect for correct paste on pads (at 20X to 40 X) - place a new CCGA and verify for a correct positioning ( using also smal mirrors, some time also X-ray this was at the beginning, before to get familiar with this process) - put the bord inside oven for preheating time/temp - after ....time remove board from oven and place on RWK machine, that must be sat already for right parameters - start the automatic reflow cycle on RWK machine - once completed the reflow cycle including cooling, remove board and do visual inspection, plus X-ray inspection ( bridges, XY registration, solder joint assessment,etc. - wash board not later the 30 minutes after RWK reflow - final inspection - additional bake ( 30 min at 90�C can be enough)of the whole PBA (Board) in order to remove all the residual of water and moisture under CCGA but also many other SMDs with low gap. Water (even if DI water) if left with high concentrationof Pb (like Column or Balls made by 90/10)can generate some problem (can see white residual like......lead )
So my suggestion is make sure you have all the knoledges enough to approach this kind of RWK, in my opinion not so difficult, but if you fail the risk is to make very high $ damages. Normally boards are very expensive and kill them by delamination or IP Separation, it is very easy. Best results obtained by AirVac (Zevac in Europe)DRS-24
Any way you schould follow also instructions suggested by CCGA Supplier. The famous CCGA maker (IBM) released plenty of papers dedicated to the SMT Attachment and RWK Process for CBGA and CCGA.
Have a good luck
Regards............GS
This message was posted the
Electronics Forum @
reply »