Yes, we do need a lot of room to play. But after thinking about it for a few minutes, I guess my answer is “it depends”. I know you were waiting for that.
But it really depends on the process being discussed.
For example, the change that Ed described, the slight change in conveyor speed for a standard CCA would probably be a minor change, it probably would not affect form, fit, function, or reliability for almost any standard CCA, whether the solder paste used was Sn63 or SAC305.
But for example, a CCA that is being reflowed with PowerPC CBGA parts, and the solder paste used to solder the PowerPC Pb90 balls to the PWB pads is a SAC305, and the profile is set up to precisely heat the SAC305 paste to form an intermetallic (not “inner metallic” like the don’t-solder brothers say)  with the surface of the PWB pad and the surface of the Pb90 ball, but not allow the Pb90 ball to go into full liquidus, that little change in belt speed would have such a huge impact on that process that it would affect both fit and reliability. That would be a candidate for a “quick and easy requal” and a minor update to the qual plan document in the form of an addendum.

Likewise, for a CCA where flipchips are being reflowed in a reflow oven, and the dwell time above solidus must be strictly controlled from 10 to 20 seconds to prevent too much of the flipchip solder ball from flowing down the unmasked traces, again the minor change in belt speed would cause a huge impact on the product, and again a re-qual would certainly be a good idea.

I could go on and on all day about different examples, such as the rate of cure of epoxy, coating, and cooldown after curing, and the impacts on the properties of the epoxy or adhesive as a result. For most assemblies a +-10% change in ramp-up in a convection oven will not change the cured material properties enough to make a difference, but for a CCA with known issues of warping during curing and cooldown, with the possibility of separation of the bonding materials from the surfaces of the parts or from the soldermask as a result, the change could have a major impact.
Another example is cleaning. One assembly will see no real impact in cleanliness from a 4% minimum saponifier level versus a 7% saponifier level, but for another, the 7% level could wash the ink right off the parts.

So there is no clear cut tolerance or general answer for that question, because there are too many different processes, applications, CCA types, part types, requirements, and so on and so forth.

One general rule that does exist is that the tolerances should be known for every process and fully documented whenever possible, but sometimes that takes time and money.

Take some Advil.

dean

From: Douglas Pauls [mailto:[log in to unmask]]
Sent: Thursday, May 26, 2016 10:15 AM
To: Stadem, Richard D.
Cc: TechNet E-Mail Forum
Subject: Re: [TN] How much change is change?

Chicken.

I agree with what Dave said, at a high level.  But where it gets to the headache part is: what is the burden of proof or how much data is needed to demonstrate that you do not affect form fit or function.

Everything I have seen so far leaves it pretty much at the discretion of the Subject Matter Experts (SMEs), which leaves a lot of room to play.  Maybe we need a lot of room to play.


Doug Pauls
Principal Materials and Process Engineer
Rockwell Collins

On Thu, May 26, 2016 at 10:11 AM, Stadem, Richard D. <[log in to unmask]<mailto:[log in to unmask]>> wrote:
It is pretty subjective, and it gives me a headache to think about it, so I am going to take the easy way out and agree with Hillman.

-----Original Message-----
From: TechNet [mailto:[log in to unmask]<mailto:[log in to unmask]>] On Behalf Of Douglas Pauls
Sent: Thursday, May 26, 2016 10:00 AM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: [TN] How much change is change?

Good morning all,

One of the IPC groups that I am leading is presently wrestling with the issue of minor vs. major change.  Generally along the lines that if you have a baselined or qualified manufacturing process, how much can that process change before it needs to be re-baselined or re-qualified?
Sometimes this is referred to as Level 1 vs. Level 2 change.

So far, every quality documentation system that I have looked at, like AS9100, ISO9000, etc., gets really fuzzy and uses vague terms when you approach this issue.  Most of these documentation systems have change better defined for products, but get extremely fuzzy and extremely vague about manufacturing processes.

This forum has a lot of very smart, very experienced people. How would you differentiate a minor change, which would not impact form fit or function, from a major change, which "could/would" impact form fit or function?

And I want all you lurkers to come out of the woodwork on this one.

Doug Pauls
Principal Materials and Process Engineer Rockwell Collins


______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]<mailto:[log in to unmask]> ______________________________________________________________________


______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________