Yuan,
        Unless I am mistaken, the increase in reliability is mostly due to the reduction of the tin/lead grain size in the solder joint.  I have personally done some research which confirms that the grain size can be reduced be a faster cooling rate.  That's the up side.
        The down size is that once you have this small grain size, you can't keep it.  As soon as the assembly leaves the reflow process, this grain size starts to grow.  This means that after several months, you are no better off than you would have been with a slower cooling rate.  Also, be having an extreme cooling rate, you may be introducing large amounts of thermally induced stresses and damaging some of your components, which may show up as increased field failures.  Definitely not a good thing.
        The bottom line is this.  If it makes you feel better to have a faster cooling rate, by all means do it, just be careful not to damage you components in the process.
	
John Guy
ACI / EMPF
714 N. Senate Ave.
Indianapolis, IN 46202-3112
voice:  (317) 655-EMPF ext.130
fax:    (317) 655-3699
e-mail:         [log in to unmask]

-----Original Message-----
From:   [log in to unmask] [SMTP:[log in to unmask]]
Sent:   Tuesday, August 05, 1997 4:13 PM
To:     [log in to unmask]
Subject:        [TECHNET] Rework cooling rate

     Hello,

     Many articles recommend rapid cooling rate during rework for reliable
     solder joints. What cooling rate can be readily achieved with most
     rework stations?  One article recommended 100 C in 5 seconds, is it
     reasonable? I thought it would be hard to achieve such a rapid
     cooling.

     Thanks,

     Yuan