In a message dated 08/24/2000 8:42:24 AM US Eastern Standard Time,
[log in to unmask] writes:
> Hi all. I'm hoping to pull on the experience on this list...
Good choice. You have gotten some excellent advice so far, reading the
existing contributions.
>
> Performing a flux evaluation, we performed SIR tests with 8 different
> conditions (4 fluxes, 2 temperatures). The IPC coupons were processed
> through our wave soldering/spray fluxing process, doing our best to
simulate
> regular production. The fluxes we are evaluating are all VOC free/no
> clean/water soluble/low solids/low residue fluxes.
Out of curiosity, what patterns or test boards are you using? Bellcore is
based on either the Bellcore pattern or the B-25 B/e or B-25A Pattern D, but
you can use other patterns with a suitable calculation of pass-fail based on
square count.
>
> The test was performed to Bellcore standards.
Which one? The older NWT-TR-78 or the newer GR-78-CORE? The SIR test is the
same, but the requirements are different.
>
> The SIR results we received were lower than expected - ranging from 2.74E+9
> to 4.27E+10. The pass value assigned by the laboratory is 3.2E+10. Only
3
> of 32 samples actually passed this standard. 6 of the 32 samples passed
the
> Bellcore minimum value (2.0E+10).
Were these the values at the first reading, i.e. after 24 hours unbiased at
35/90? Or were they the second values, obtained after 5 days with bias
applied after 24 hours?
Another two factors to consider relate to chamber dynamics. If your chamber
is heavily loaded with test samples, and you don't have good air flow over
your sensor elements, then one portion of your chamber may be at 35/95 and
another at 35/60. I have seen a decade difference before just based on
humidity variations. I did a paper for the 1998 Georgia Tech Flux Conference
regarding some SIR research I did on BGA attachment. We found that dendrites
grew most often in the main airflow of the humidity stream, and very seldom
on the coupons that were off in the corners of the chambers.
> What sort of implications can we expect
> with selecting a flux with slightly lower SIR results than the minimum
> value?
Well, there's an open ended question if I ever heard one. You need to select
a flux or paste based on the best blend of properties. Most flux vendors I
know could make you a flux that would have SIR readings that were 2-3 orders
of magnitude above the current J-STD-004 SIR minimums. Just yank out all the
activators and you get great SIR. Guess what happens when you try to solder
with it though. Just passing the SIR test is not a magic bullet. Having a
material that meets Bellcore, or IPC, or ISO, or JIS, etc., consistently is
no guarantee that your product will be free from failures.
When I do SIR evaluations in screening experiments, I look for three things:
- what do the SIR levels do in the first 24 hours? If they go real low and
come back quick, there is a hydroscopic agent that volatilizes easily with
heat. Not a desirable thing.
- is there any metal migration. Automatic disqualification unless it came
because of water droplets in the chambers.
- is there any corrosion? Depends on what you call corrosion, and where it
occurs.
You say that you are evaluating a VOC free/no clean/water soluble/low
solids/low residue flux. I was with you until the water soluble part. What
are the activity levels of the fluxes you are evaluating. I would expect
lower SIR from an M level flux than I would from and L level flux.
> Each of the flux vendors provided their own passing SIR test results.
Wow, there's a shocker!. You don't really expect them to publish results
showing they fail the standards do you? Keep in mind that when vendors
process samples for Bellcore or IPC conditions, you can bet your bottom
dollar they are processed under optimal optimal conditions. And, there are
ways in which you can improve the SIR responses in qualification............
>
> One other thing that was noted was that the boards with the fewest defects
> came back with the poorest SIR results, and vice versa.
What do you consider a defect? What is your criteria for judgement. The
Bellcore spec talks about not having visible corrosion, but then ducks the
issue of defining what visible corrosion means.
> Is this just a coincidence, or is there a solid explanation for this?
There is one, but its a secret. I'd have to kill you...........
>
> Thanks in advance,
You're welcome, in advance.
Doug Pauls
Technical Director and Diet Mt. Dew Addict
##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe: SUBSCRIBE TECHNET <your full name>
To unsubscribe: SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################
|
