Reports

SMART  GROUP  VISIT  TO  ELECTROVERT SMART  REFLOW 

DAY A large group of  SMART Group  members visited  the Henley  premises of Electrovert on 9 May. After  the welcome and introduction  where Mike Judd  emphasised  Electrovert's commitment  to SMT, a number of  papers were  presented  by  the  Electrovert  team  during  the  morning  session. Alan Weston covered the Vapour Phase Reflow process. There is a keen interest in this process again, with the equipment now in third  generation. Thermal shock is eliminated using pre-heating as part of the machine. Fluid prices are reducing.  Systems that  eliminate  the secondary  blanket  ensure that  vapour  phase  uses  no  CFCs. So the pendulum  now appears to be swinging back in favour  of  vapour phase. Quality, reliability and consistency ensure that this process  should be  compared  against  other  reflow  methods. Phil Atkinson presented the advantages of infra-red  reflow soldering of PCBs. He covered recent advances in machines as well as the process. Panel heater  design,  'edge  heater'  technology,  controlled  convection, exhaust/cooling,  computer  control,  and  surface  mounted  closed  loop thermocouple sensors were all described. The importance of evaluating  IR reflow  systems, with guidelines for  selection, were covered in Phil's  paper. Geoff  Pickles spoke on rework, referring  specifically  to a  sophisticated system  that  monitors  and  controls  the  rework  process,  automatically develops  time/temperature  profiles  for  every  component,  location  and board.  Geoff  stressed  the  importance  of  getting  the  reflow  process  right first  time,  but  where  rework  was  necessary  the  equipment  should  allow similar  temperature  profiles  with  repeatability. Tony Welford  took the opportunity to cover briefly the recent advances in SMT cleaning—batch  and in-line, alternative solvents to CFCs and  the ozone issue, terpenes and the trend  towards water cleaning. Tony  offered some  suggestions  on  future  equipment. Two guest  speakers  balanced  the  programme.  Gordon  Prior  of  Alpha Metals discussed solder paste and screens. Russ Wood of 3M gave his usual high  quality  presentation  on  vapour  phase  fluids.  Time  was allowed  for questions  to  all  speakers.

 #p#分页标题#e#

ELECTRONICS  BY THE  SEASIDE

The  small  English  seaside town  of  Weymouth  played  host  on  27  June to a gathering  of  members  of  the  SMART  Group.  The  one-day  seminar concerned  itself with the current issue of  Fine Pitch Technology within the surface  mount  industry.  Presentations  were given  by Alan  Hobby,  DEK Printing Machines, on 'Fine Pitch and the Solder Paste Printer',  Malcolm Huntley-Thomas,  DEK  Precision  Screen  Division,  on  'SMT  Screens  for Fine Pitch Applications', and Bob Hudson of Alpha Metals on 'Fine Line Solder  Paste'. All  presentations  were  very  informative,  and  'well  received'  was  the verdict  given  by  all  20  attendees.

COMPONENT SOLDERABILITY  IN ELECTRONICS  MANUFACTURE —A  DISCUSSION  MEETING

National  Physical  Laboratory,  4 July  1990 The  above  meeting  was  organised  jointly  by  the  National  Physical Laboratory,  the  Surface  Mount  Club  and  the  SMART  Group. The effectiveness  of automated  soldering processes depends critically upon the  solderability  of  the  two  surfaces  being joined.  In  recent  months  there has been much speculation and debate in the technical press about the fitness-for-purpose  of the solderable surfaces  of components, in particular  surface mount components. Especially at issue are the solderability at goods inwards, the  deterioration  of  solderability  during  storage,  the  date  coding  of components,  and  the  interval  between  solderability  re-tests. This  one-day discussion  meeting  was  aimed  at  providing  a  forum  for  component  users to give their experience and component suppliers to voice their  manufacturing and  distribution  constraints,  in  terms  of  both  materials  properties  and economics,  that  may  limit  component  quality. The proceedings of  the day began  with a review of  the present  scientific understanding of the materials issues that impact upon electronic component solderability. This was given by Colin Lea (NPL). Component  solderability comprises both the intrinsic wettability of the terminations with an acceptable electronics-grade  flux and  the component's ability to withstand  the heat  of the soldering process. Examples of inadequate heat resistance are the thermal cracking of chip capacitors and the popcorn effect  from  which some plastic surface  mount  components  can  suffer.  This meeting,  however,  considered primarily  the  phenomenon  of  wetting  and  wettability. The majority  of component terminations have fusible coatings, normally of  tin or  tin-lead. The  factors  that  control  the wettability of  a  coating  are (i) the oxide thickness, (ii) the morphology  and  thickness of  the  interfacial intermetallic phase formed by diffusion  between the substrate and its coating, and (iii) the thickness of the remaining  fusible  coating.  Each of  these three parameters can, by suitable selective heat treatments, be studied independently. In  general  terms,  the  thicker  the coating,  the better  will  be the  wettability and  the  longer  the  solderability  will  be  retained.  All  parts  of  the  coating initially should be at least 5 μm thick. Under normal storage conditions  the intermetallic  compound  will grow,  and  reduce the  thickness of  the  fusible coating,  very  slowly.  If,  during  manufacture,  component  leads  were  hot dipped  or  a  plated  coating  was  fused,  the  molten  solder  at  the  corners  of leads of rectangular cioss-section can be very thin even if the average thickness is well in excess of 5 μm. Thus, either leads should be circular or the plated coating  should  be left  unfused.  The problems associated  with  using  bright plated  finishes  rather  than  matt  were also  discussed. Some components have no fusible coating on their terminations. Generally these  are  more  difficult  to  wet  with  solder  during  assembly,  but  any deterioration  in wettability  during  ageing  is more easily  controlled  since it is dependent on the cleanliness and the temperature of the storage atmosphere. Species  such  as  hydrogen  sulphide,  sulphur  dioxide,  hydrogen  chloride, nitrogen dioxide and organic acids at very low levels can have a marked  effect on  component  wettability. The  rôle  of  component  lead  geometry  (in  addition  to  intrinsic  surface wettability)  in  determining  the  ease  of  solder  fillet  formation  was  also discussed.  The  design  engineer  can  modify  the circuit  board  pad  size  and shape but is stuck with a component  lead design that  is often  incompatible with  good  soldering  practice.  The  problems  of  goods  inward  testing  of solderability and the usefulness  of artificial  ageing procedures such as high temperature  dry  heat  and  steam  were  discussed. Two case studies  were used  by Martin  Wickham  (Surface  Mount  Club) to  illustrate  the  crucial  aspect  of  component  solderability  in  electronics manufacture.  The  following  points  have  great  relevance  to  the  debate: (i)  Electronics assemblers usually cannot disrupt  a production  schedule to complain or return goods to their component  suppliers if a  batch is delivered  with  poor  wettability. They  tend  to grit their  teeth  and rework,  or  change  to  a  more  active  flux. (ii)  There  is  a  trend  from  RMA  towards  RA  fluxes,  to  overcome experienced  problems  of  wettability,  especially  in  surface  mount. (iii)  Reflow  soldering  of  solder  paste  requires  a  higher  quality  of component  wettability than does wave soldering because the  flux  is coming  at  the  metallisation  'from  a  standing  start'. (iv)  In a recent Surface Mount Club survey, 55% of electronics assembly companies experienced in in-house manufacture  perceived component solderability  as  being  a  major  problem.  Only  12%  of  component manufacturers  acknowledged  that  there  was any  problem  at  all. Ron  Neale  (RA  Neale  Associates)  talked  about  quality  control  and assessment, through goods inward acceptance testing. He has been party to the  writing  of  many  of  the  BS and  international  standards  that  relate  to solderability  of  wires, components  and  circuit  boards.  For  any  debate  on solderability  to be meaningful,  tests and  measurement  methods have to be standardised  and,  if  possible, quantified.  The dip and  inspect  method  can be agreed procedurally but the visual assessment is always open to subjectivity. The wetting balance, using a solder bath and more recently a solder globule, gives  the  opportunity  for  quantitative  measurement. Ron  also explained  how  the choices  for  accelerated  ageing  procedures were made. Often  these procedures are used in regimes, or for  component types, for which they were never intended and are inappropriate. An example of this is a short-term (one hour) steam test during which the solderability can  actually  improve  as  the  surface  cleans. The  morning  session  concluded  with  an  amusing  but  very  salient  talk about the experiences of one particular surface mount company. This was given by the Chairman of the SMART Technical Committee, Glyn Jackson (Multitone). Through a concerted  programme to improve component  and process control, solder joint defects have been reduced from  3000 to 50 ppm on a turnover  of  half  a million components a week. Much  of  this  success is due to improved solderability acceptance testing and component storage. Glyn has questioned all his component  suppliers over a number of issues. One  issue  is  the  sealing  of  component  reels  in  airtight  bags:  those  who offered  this  claimed  (quite rightly) a  significantly  longer  shelf  life;  those who  did  not  offer  airtight  bags  claimed  there  would  be  no  advantage. Another  issue  raised  was that  of  date  coding:  only  6 of  his  10 suppliers datecode YY.WW (Year.Week) so as to be understandable to the user. For some manufacturers  the code refers to date of manufacture while for others it is the date  of  taping.  Sometimes  these dates are a long way apart.  The implementation of a YY.WW date of manufacture coding as a requirement for  component  supply  and  acceptance  would  represent  a  very  tangible benefit.#p#分页标题#e#

Reports SMART  GROUP  VISIT  TO  ELECTROVERT SMART  REFLOW  DAY

A large group of  SMART Group  members visited  the Henley  premises of Electrovert on 9 May. After  the welcome and introduction  where Mike Judd  emphasised  Electrovert's commitment  to SMT, a number of  papers were  presented  by  the  Electrovert  team  during  the  morning  session. Alan Weston covered the Vapour Phase Reflow process. There is a keen interest in this process again, with the equipment now in third  generation. Thermal shock is eliminated using pre-heating as part of the machine. Fluid prices are reducing.  Systems that  eliminate  the secondary  blanket  ensure that  vapour  phase  uses  no  CFCs. So the pendulum  now appears to be swinging back in favour  of  vapour phase. Quality, reliability and consistency ensure that this process  should be  compared  against  other  reflow  methods. Phil Atkinson presented the advantages of infra-red  reflow soldering of PCBs. He covered recent advances in machines as well as the process. Panel heater  design,  'edge  heater'  technology,  controlled  convection, exhaust/cooling,  computer  control,  and  surface  mounted  closed  loop thermocouple sensors were all described. The importance of evaluating  IR reflow  systems, with guidelines for  selection, were covered in Phil's  paper. Geoff  Pickles spoke on rework, referring  specifically  to a  sophisticated system  that  monitors  and  controls  the  rework  process,  automatically develops  time/temperature  profiles  for  every  component,  location  and board.  Geoff  stressed  the  importance  of  getting  the  reflow  process  right first  time,  but  where  rework  was  necessary  the  equipment  should  allow similar  temperature  profiles  with  repeatability. Tony Welford  took the opportunity to cover briefly the recent advances in SMT cleaning—batch  and in-line, alternative solvents to CFCs and  the ozone issue, terpenes and the trend  towards water cleaning. Tony  offered some  suggestions  on  future  equipment. Two guest  speakers  balanced  the  programme.  Gordon  Prior  of  Alpha Metals discussed solder paste and screens. Russ Wood of 3M gave his usual high  quality  presentation  on  vapour  phase  fluids.  Time  was allowed  for questions  to  all  speakers. ELECTRONICS  BY THE  SEASIDE The  small  English  seaside town  of  Weymouth  played  host  on  27  June to a gathering  of  members  of  the  SMART  Group.  The  one-day  seminar concerned  itself with the current issue of  Fine Pitch Technology within the surface  mount  industry.  Presentations  were given  by Alan  Hobby,  DEK Printing Machines, on 'Fine Pitch and the Solder Paste Printer',  Malcolm Huntley-Thomas,  DEK  Precision  Screen  Division,  on  'SMT  Screens  for Fine Pitch Applications', and Bob Hudson of Alpha Metals on 'Fine Line Solder  Paste'. All  presentations  were  very  informative,  and  'well  received'  was  the verdict  given  by  all  20  attendees.#p#分页标题#e#

COMPONENT SOLDERABILITY  IN ELECTRONICS  MANUFACTURE —A  DISCUSSION  MEETING

National  Physical  Laboratory,  4 July  1990 The  above  meeting  was  organised  jointly  by  the  National  Physical Laboratory,  the  Surface  Mount  Club  and  the  SMART  Group. The effectiveness  of automated  soldering processes depends critically upon the  solderability  of  the  two  surfaces  being joined.  In  recent  months  there has been much speculation and debate in the technical press about the fitness-for-purpose  of the solderable surfaces  of components, in particular  surface mount components. Especially at issue are the solderability at goods inwards, the  deterioration  of  solderability  during  storage,  the  date  coding  of components,  and  the  interval  between  solderability  re-tests. This  one-day discussion  meeting  was  aimed  at  providing  a  forum  for  component  users to give their experience and component suppliers to voice their  manufacturing and  distribution  constraints,  in  terms  of  both  materials  properties  and economics,  that  may  limit  component  quality. The proceedings of  the day began  with a review of  the present  scientific understanding of the materials issues that impact upon electronic component solderability. This was given by Colin Lea (NPL). Component  solderability comprises both the intrinsic wettability of the terminations with an acceptable electronics-grade  flux and  the component's ability to withstand  the heat  of the soldering process. Examples of inadequate heat resistance are the thermal cracking of chip capacitors and the popcorn effect  from  which some plastic surface  mount  components  can  suffer.  This meeting,  however,  considered primarily  the  phenomenon  of  wetting  and  wettability. The majority  of component terminations have fusible coatings, normally of  tin or  tin-lead. The  factors  that  control  the wettability of  a  coating  are (i) the oxide thickness, (ii) the morphology  and  thickness of  the  interfacial intermetallic phase formed by diffusion  between the substrate and its coating, and (iii) the thickness of the remaining  fusible  coating.  Each of  these three parameters can, by suitable selective heat treatments, be studied independently. In  general  terms,  the  thicker  the coating,  the better  will  be the  wettability and  the  longer  the  solderability  will  be  retained.  All  parts  of  the  coating initially should be at least 5 μm thick. Under normal storage conditions  the intermetallic  compound  will grow,  and  reduce the  thickness of  the  fusible coating,  very  slowly.  If,  during  manufacture,  component  leads  were  hot dipped  or  a  plated  coating  was  fused,  the  molten  solder  at  the  corners  of leads of rectangular cioss-section can be very thin even if the average thickness is well in excess of 5 μm. Thus, either leads should be circular or the plated coating  should  be left  unfused.  The problems associated  with  using  bright plated  finishes  rather  than  matt  were also  discussed. Some components have no fusible coating on their terminations. Generally these  are  #p#分页标题#e#http://www.ukthesis.org/ygsslwdx/ more  difficult  to  wet  with  solder  during  assembly,  but  any deterioration  in wettability  during  ageing  is more easily  controlled  since it is dependent on the cleanliness and the temperature of the storage atmosphere. Species  such  as  hydrogen  sulphide,  sulphur  dioxide,  hydrogen  chloride, nitrogen dioxide and organic acids at very low levels can have a marked  effect on  component  wettability. The  rôle  of  component  lead  geometry  (in  addition  to  intrinsic  surface wettability)  in  determining  the  ease  of  solder  fillet  formation  was  also discussed.  The  design  engineer  can  modify  the circuit  board  pad  size  and shape but is stuck with a component  lead design that  is often  incompatible with  good  soldering  practice.  The  problems  of  goods  inward  testing  of solderability and the usefulness  of artificial  ageing procedures such as high temperature  dry  heat  and  steam  were  discussed. Two case studies  were used  by Martin  Wickham  (Surface  Mount  Club) to  illustrate  the  crucial  aspect  of  component  solderability  in  electronics manufacture.  The  following  points  have  great  relevance  to  the  debate: (i)  Electronics assemblers usually cannot disrupt  a production  schedule to complain or return goods to their component  suppliers if a  batch is delivered  with  poor  wettability. They  tend  to grit their  teeth  and rework,  or  change  to  a  more  active  flux. (ii)  There  is  a  trend  from  RMA  towards  RA  fluxes,  to  overcome experienced  problems  of  wettability,  especially  in  surface  mount. (iii)  Reflow  soldering  of  solder  paste  requires  a  higher  quality  of component  wettability than does wave soldering because the  flux  is coming  at  the  metallisation  'from  a  standing  start'. (iv)  In a recent Surface Mount Club survey, 55% of electronics assembly companies experienced in in-house manufacture  perceived component solderability  as  being  a  major  problem.  Only  12%  of  component manufacturers  acknowledged  that  there  was any  problem  at  all. Ron  Neale  (RA  Neale  Associates)  talked  about  quality  control  and assessment, through goods inward acceptance testing. He has been party to the  writing  of  many  of  the  BS and  international  standards  that  relate  to solderability  of  wires, components  and  circuit  boards.  For  any  debate  on solderability  to be meaningful,  tests and  measurement  methods have to be standardised  and,  if  possible, quantified.  The dip and  inspect  method  can be agreed procedurally but the visual assessment is always open to subjectivity. The wetting balance, using a solder bath and more recently a solder globule, gives  the  opportunity  for  quantitative  measurement. Ron  also explained  how  the choices  for  accelerated  ageing  procedures were made. Often  these procedures are used in regimes, or for  component types, for which they were never intended and are inappropriate. An example of this is a short-term (one hour) steam test during which the solderability can  actually  improve  as  the  surface  cleans. The  morning  session  concluded  with  an  amusing  but  very  salient  talk about the experiences of one particular surface mount company. This was given by the Chairman of the SMART Technical Committee, Glyn Jackson (Multitone). Through a concerted  programme to improve component  and process control, solder joint defects have been reduced from  3000 to 50 ppm on a turnover  of  half  a million components a week. Much  of  this  success is due to improved solderability acceptance testing and component storage. Glyn has questioned all his component  suppliers over a number of issues. One  issue  is  the  sealing  of  component  reels  in  airtight  bags:  those  who offered  this  claimed  (quite rightly) a  significantly  longer  shelf  life;  those who  did  not  offer  airtight  bags  claimed  there  would  be  no  advantage. Another  issue  raised  was that  of  date  coding:  only  6 of  his  10 suppliers datecode YY.WW (Year.Week) so as to be understandable to the user. For some manufacturers  the code refers to date of manufacture while for others it is the date  of  taping.  Sometimes  these dates are a long way apart.  The implementation of a YY.WW date of manufacture coding as a requirement for  component  supply  and  acceptance  would  represent  a  very  tangible benefit. +++morning starts. On the first day, a basic history and overview of Statistical Process Control was presented. The need for SPC was introduced and the definition  of quality discussed. Through a simple bead box experiment, the idea of controllable and uncontrollable variations was illustrated by Zachery Brice. Later in the day, Paddy Regan discussed the organisation involved in co-ordinating an SPC programme, and the importance of both management  commitment  and  employee  involvement  for  successful implementation. The second day began with Paddy Regan outlining the different  types of control charts that can be used, depending on the type of data that can be measured during the process. Various examples of charts were shown and  discussed.  Dr  Eamon  Murphy  followed  with  a  presentation  on experimental design, which in theory should minimise the 'tweaking' on a  process and identify  the optimal process settings. He also discussed a number  of  the  statistics  packages  available.  Daithi  Fallon  of  Digital Equipment in Galway presented a case-study of a designed experiment which involved just one day's work to optimise a rework system. The attendees were then divided into three groups, and each group was presented with a problem. Each group was then asked to define its problem, the process steps involved and then design an experiment to identify the main possible causes of the problem through the different  process stages. This proved a most interesting and informative exercise for most of the attendees who had no experience in this area, particularly when it came to presenting the results. The workshop adjourned early on the first day to a tour of the facilities at the National Microelectronics  Research Centre, where attendees were able  to  see  state-of-the-art  microelectronics  facilities  including  silicon fabrication,  IC  test,  failure  analysis,  computer-aided  design,  II1-IV semiconductor  processing, and  interconnection  and  packaging  facilities consisting of high pincount IC packaging, SMT and thickfilm/hybrid circuit processing, environmental test equipment and IC package characterisation. The tour was followed  by a reception given by the NMRC during which attendees were addressed by Cian O Mathuna of SMART Group-Ireland and Bob Willis, chairman of the SMART Group.#p#分页标题#e#