TECHNET Archives

November 2001

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Earl Moon <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Thu, 8 Nov 2001 04:44:14 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (124 lines)
Ervin,

I?ve been laboring over your very important question and concerns as,
partly, they reflect some of mine. I first would note there are at least
three very highly regarded contributors to this forum that are acknowledged
(in 6012) members of the 6012 development committee - if that is the proper
terminology. They should be able to shed some light on your questions for
obvious reasons and I suspect they are readying a response. Considering this
as a valuable learning experience, my meager input
follows as Me:

Hi Technetters,
I wish to present you a dilemma we are facing while ordering complex boards.

Most of the boards we order are considered "High reliability" and we specify
them to comply with IPC-6012, Class3. The boards are all polyimide, rigid
multilayer boards.

Me: It is clear these requirements are specified  in Section 3. What else is
there anyway?

Recently, while ordering a complex board we faced a serious dilemma:
The board is still a rigid, polyimide multilayer, about 1mm thick and
contains blind via holes. Naturally, it has also fine lines on the outerlayers.

Me: At 1mm thickness, how thick are the three individual boards, should this
be the case, comprising the complete MLB structure? 1mm, as I recall, is
about .250? thick. Does this mean a 20 plus layer MLB? It must based on your
core thickness questions as 4 or 6 mil. It certainly could be a very high
layer count type.

As you may already suspect, the combination of fine lines + blind via holes
is an uneasy one since the blind holes are made through sequential
lamination (many plating steps----> thick plating on
outerlayer--->difficulty to produce fine lines). We don't order laser
drilled microvia holes from various reasons, some of them related to long
term reliability.

Me: Cannot correlate sequential lamination with blind via holes but for
filling vias with resin if we are talking about ?surface? blind or micro
vias. Just trying to get a picture here, so please set me straight.
Understand ?many? plating steps as sequential lamination is what it says:
The relamination of a double sided or MLB as a ?core? board sandwiched
between two other double sided or MLB?s, as but one example, and I have no
problem with this. One of the things happening here is that buried vias are
now filled with resin during the relamination process. Are these ?buried
vias? really blind vias in the ?core? board or are they interfacial? How
?thick? is the outer layer plating and are the fine lines as, say, 5 mils?
So how much hole plating is there on the outer boards in the sequentially
laminated structure? Surely meets 6012 requirements as you say and as is
indicated in table 3.6. Must be high aspect ratio holes in ?core? board ?
over 6:1? And we?re talking minimum plating thickness as specified in table
3.6 and in paragraph 3.2.6.2?

When looking carefully into the 6012, one sees that the requirements are
different for the various hole types -
Through hole must be at least 25 micron for Class 3.
Blind via must be at least 25 micron for Class 3.
Burried via must be at least 15 micron for Class 3.
Microvia must be at least 10 micron for Class 3 (O.K, I know it's not in the
6012 but I have also the 6016).
Now, my question is: Why ???

Me: These are plating thicknesses from table 3.6 ? correct?

What is the difference between burried via and blind via (we specify our
blind vias to be 100% filled with resin) that it is allowed to have only 15
microns of PTH?

Me: Again, are the buried vias blind or interfacial in the ?core? board?

What is the difference between a 0.004" (drill size) blind via over a 0.006"
or 0.004" core (produced through sequential lamination and is resin filled)
and a 0.004" microvia?

Me: Are we talking a 4 mil blind via on the outer board in the structure or
in the ?core? board? Is the 4 or 6 core in the outer boards or in the ?core?
board? How thick is the core in which the 4 mil via is drilled? I?m thinking
thick ?core? board with high aspect ratio holes and thin outer boards with
no aspect ratio concerns ? right?

Where would you draw the line? Does the aspect ratio of the hole play any
role in the decision?

Me: As above, the only place I see aspect ratio being a problem is in the
?core? board. It shouldn?t be an issue on the outer boards as they should be
relatively thin if double sided or even thin MLB?s. If this is true, even
the aspect ratio, as I see it (and a big question for the 6012 creators), in
the ?core board? may not be a problem provided plating thickness and
ductility requirements are met. Also, the fact these vias are resin filled
supports their mechanical integrity, through thermal stress and shock, even
though the polyimide resin system still expands about 50 ppm/C in the z axis.

You see, I need to specify Good, High reliability boards but I wonder maybe
some of these requirements add nothing to the reliability of the board but
makes it much more difficult to make and maybe even less reliable.

You, who specify, build and assemble printed circuit boards, have you had
this dilemma? What do you think should we specify?


Thank you,
Ervin Weisz
Process Engineer

Me: You, again, raise valid questions and concerns. Trying to think through
all this, for me, is difficult as those who created the requirements must
have done so based on evidence not available to us trying to specify and buy
such boards. I?m certainly looking forward to the answers but realize it
sure would be nice to see the design and to determine if it could be made
simpler thus avoiding such vexing questions.

Earl Moon

---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send the following message: SET Technet NOMAIL
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------

ATOM RSS1 RSS2