The consensus of the member participation on the IPC-RB-276 committee feels that 1.5% for all boards is adequate. They feel that tighter requirements could cause unnecessary scrapping of perfectly usable boards. Tighter requirements can be invoked by cust
omer in situations where needed.

The IPC-A-600 looks to the IPC-RB-276 for its requirements, and this must stay with the 1.5%. The A-600E, however, does have a note in it that states "The stated requirements may not meet surface mount board requirements". This is to inform you that you c
an tighten the requirements if needed. 

The consensus of the J-STD-001 and IPC-A-610 committees is that tighter requirements are needed for some surface mount assemblies, due to solder joint damage concerns from assembly mounting in the final housing/card cage. Both of these committees have add
ed the 0.75% requirement in their standards for surface mount assemblies.

We sent the above statement to Phil Hinton, Hinton "PWB" Engineering, who is the chairman of the Rigid Board Performance  Task Group and received the following response:

The Rigid Board Performance Task Group, formerly the IPC-RB-276
Task Group, at the meeting in San Diego changed the Bow and Twist
requirements to 1% for Surface Mount and 1 1/2% for all other
boards ( See IPC-6012 para 3.4.4, 2nd working draft dtd. June
1995).  There seemed to be little controversy among the group which
was composed almost entirely of printed board fabricators.

It is obvious that an assembler cannot take a 2mm thick printed
board that is 250mm long and has a bow/twist of 1 1/2% or even 1%
(3.75mm or 2.5mm of flat) and screen paste onto the surface mount
lands using a stencil and a metal squeegee. The numbers that the
assembly group adopted of 0.75% is probably a more realistic
number. There will undoubtedly be more discussion of the subject at
the interim meeting in San Jose.

I would hesitate to give the impression that the rigid board task
group favors the 1 1/2% percent limit for surface mount since in
the RB-276 spec. we side-stepped the assemblers needs and made it
mandatory for the user to otherwise specify, if he wants a tighter
requirement. Today, with many boards going from the fabricator to
a contract assembler it is certainly desirable that he gets what he
can use the first time without a later engineering change needed to
add a tighter requirement to the procurement documents.

For lead-in-hole boards the 1 1/2% is a valid limit since the
assembler cant and most often does, restrain the board as it passes
over the wave; he can allow the looser requirements.

****************************************************
Jon Holmen
Technical Project Manager
IPC
7380 N. Lincoln Ave
Lincolnwood IL  60646
Phone (708) 677-2850
Fax   (708) 677-9570
e-mail  [log in to unmask]
*****************************************************


On Thu, 6 Jul 1995, Peter Swanson wrote:

> One of our customers has a query:
> 
> "I am adopting IPC standards in our company, especially in regard to pcb 
> flatness.
> 
> IPC 610 (para 8.3) gives 0.75% for populated surface mount boards.
> 
> IPC 600 (para 10.1) requires on average 1.5% with only 6 instances (T4 sized 
> boards) where flatness is better than 0.75% and this is for _non-populated_ 
> boards.
> 
> I find this confusing - can you advise."
> 
> We note that IPC 600 says that flatness requirements stated may not meet 
> surface mount requirements. What will the new revision of 600 say?
> 
> o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o
>                Peter Swanson, Oxfordshire, England
>                      Dynamix Technology Ltd
>                    [log in to unmask]
>                     CompuServe: 100120,3641
> 
>  If you tied buttered toast to the back of a cat and dropped it from a height,
>                         what would happen?
> o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o
> 
>