OK, probably against my better judgement to weigh in on this but here it
goes.
Barrel cracks? Precise location is important here. If at or near the knee of
the hole we are likely looking at thermal stress failure of the copper
deposit as this is where columnar grain structure of the copper will exhibit
itself 1st. If down in the barrel then we look elswhere and drill
characteristics can have an effect. Typically drill parameters are
controlled and usually based in some way on drill manufacturers
recommedations. Feeds and speeds are adjusted to generate a specific "chip
load", a theorhetical particle size to be evacuated up the flute of the
drill. Some adjustments are generally made for material and board type
(multilayer, double sided, polyimide etc.). Better shops will make these
adjustments based on experimentation and empirical evaluation of hole wall
quality (lotsa microsections).
Maximum hit counts are a function of drill wear (he said, stating the
obvious). Again, drill manufacturer's recommendations are a baseline.
Adjustments generally made after close monitoring of drill point geometry
with respect to wear. Hit counts are tracked automatially, and tools offered
for change by most modern drilling equipment.
What is commonly referred to as barrel cracking is often the result of a
very narrow circumferential (or sometimes not circumferential) void in the
hole coming out of the metallization process, be it eletroless copper or
some of the direct metallization methods. The plating as boards dwell in an
electroplating bath will tend to close around the void a bit leaving the
appearance of a crack but in general these are often referred to as plating
folds. They will tend to fail after thermal excursions. Non-circumferential
voids can be the result of drill parameters as debris impacted in the hole
wall comes loose after metallization and before plating. Direct
metallization methods are more sensitive as the deposit is far thinner than
electrolless. True circumferential problems generally have other contibuting
factors rooted in the through hole processing (desmear, metallization, etc.)
A severe enough case of pink ring can even contribute. Micosections should
be reviewed carefully to point toward root cause. Have you considered using
a failure analysis lab? Trace or Robisan might be able to help.
Hope I muddied the waters sufficiently. <G>
Bob Dube
-----Original Message-----
From: Paul Klasek <[log in to unmask]>
To: [log in to unmask] <[log in to unmask]>
Date: Friday, January 07, 2000 5:01 PM
Subject: Re: [TN] Standard No. of via drill cycles
>Thanks folks , been bit vague, sorry,
>
>standard (whatever that stands for)laminate FR4 4layer 1/16 ,
>mainland china via HK,
>
>Yes the cracks are on the barrel rings (not a pad/track separation)
>That is why i wonder if anybody would have a set routine,
>say 2000 and change (for a particular board configuration )?
>Or how is the "due" determined ?
>
>From the fab side , am i on the right wavelength assuming blunt
drill>coarse
>wall>uneven plate would be the most common cause of cracked barrels ?
>
>No high resin content (can see some 7-8 layers of glass from routed edge);
>overcured laminate ? = harder ? ; aggressive desmear ? (now 'm definitely
in
>deep end)
>
>thanks again paul
>
>-----Original Message-----
>From: [log in to unmask] [mailto:[log in to unmask]]
>Sent: Friday, 7 January 2000 21:36
>To: [log in to unmask]; [log in to unmask]
>Subject: Re: [TN] Standard No. of via drill cycles
>
>
>Paul,
>
>Depending on the number of layers and aspect ratio of the product you are
>drilling, most folks are getting 3 to 7 repoints and 1000 to 3000 hits
>per use with small drills (.0210" - .0120"). All of this depends on what
>level of positional accuracy, hole quality, and breakage the quality
>departments of PCB manufactures can tolerate. As for utilization habits,
>the US tends to be more aggressive in tool utilization in the drill room
>than most other countries, however, when competing with countries that
>have much lower wage and salary structures you can understand why.
>
>A question that I raise is what is the substrate material that is giving
>you cracking? There are several resin systems with higher Tg's that when
>drilled aggressively tend to crack easily. Also, you mention uneven
>barrel plating (rough holes). This can be a sign of over-cured material,
>an aggressive desmear process, or a board design with a high resin
>content.
>
>Regards,
>
>Mark Ford
>
>
>On Fri, 7 Jan 2000 16:00:53 +1100 Paul Klasek <[log in to unmask]>
>writes:
>>Dear fab folks ;
>>could you give me indication of "standard" No. of cycles on small via
>>holes
>>drill sizes ,
>>on carbide and steel bits please ,
>>and would there be a difference of utilization habits between say US
>>and
>>SEA/CHINA/SOB regions ?
>>
>>Facing patches of cracked vias , suspecting blunt drills > uneven
>>barrel
>>plate .
>>
>>And class 3 , is there via fill (solder) mandatory ?
>>
>>Thanks a lot
>>
>>Paul Klasek
>>ResMed
>>
>>##############################################################
>>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 Gayatri Sardeshpande at
>>[log in to unmask] or
>>847-509-9700 ext.5365
>>##############################################################
>
>##############################################################
>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 Gayatri Sardeshpande at [log in to unmask] or
>847-509-9700 ext.5365
>##############################################################
>
##############################################################
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 Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################
|
