[SI-LIST] Re: Backdrill under BGA
- From: "Lee Ritchey" <leeritchey@xxxxxxxxxxxxx>
- To: "Aubrey Sparkman" <asparkman@xxxxxxxxxxxxx>, "Bill Stube" <Bill.Stube@xxxxxxxxxx>, cygnul@xxxxxxxxx, si-list@xxxxxxxxxxxxx
- Date: Wed, 22 Jul 2009 10:09:12 -0700
I have tests that show that the usual 12 mil drill diameter vias in PCBs at
least as thick as 120 mils are of no concern up to 6.4 Gb/S and of no
concern to 10 Gb/S as long as the signal travels from the BGA pin down to
one of the bottom layers of the PCB. (These vias tend to measure about 0.5
pF.)
Backdrilling comes into play when the holes are the size used for press fit
connectors (24 mil drill) when the backplane exceeds 100 mils thick. I've
got several test PCBs that show this. (These vias can measure up to 2 pF.)
I think the worry over the effects of vias is based on supposition rather
than actual measurements in many cases. We have a habit of worrying about
things rather than measuring them to see if they are significant as opposed
to just visible. To avoid this, I've built almost a dozen test PCBs to
measure what the effects are of various structures. Much of this data has
been presented in public forums.
Lee Ritchey
Speeding Edge
> [Original Message]
> From: Aubrey Sparkman <asparkman@xxxxxxxxxxxxx>
> To: Bill Stube <Bill.Stube@xxxxxxxxxx>; <cygnul@xxxxxxxxx>;
<si-list@xxxxxxxxxxxxx>
> Date: 7/22/2009 9:41:16 AM
> Subject: [SI-LIST] Re: Backdrill under BGA
>
> We looked at backdrilling an 82 mil thick pcb under a 1mm (40 mils) pitch
> BGA once. Almost all of the low cost PCB suppliers said no.
>
>
> Aubrey Sparkman
> Aubrey.K.Sparkman@xxxxxxxx
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On
> Behalf Of Bill Stube
> Sent: Wednesday, July 22, 2009 11:36 AM
> To: cygnul@xxxxxxxxx; si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Backdrill under BGA
>
> Sam,
>
> I do not have experience with backdrilling under BGAs specifically.
>
> My suggestion would be to avoid putting vias on the signal at BGA in the
> first place, or if you must have vias to transition from layer 1 to layer
N
> and avoid the stub. I realize this doesn't exactly address your question,
> but if you avoid having a via/stub altogether, you will avoid the added
> complexity and cost of having to backdrill and account for the items David
> mentions below.
>
> Just thought I would throw it out there.
>
> Bill Stube
> Hardware Engineer
>
> Plexus - http://www.plexus.com
> The Product Realization Company
>
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On
> Behalf Of David Mullenex
> Sent: Wednesday, July 22, 2009 4:13 AM
> To: cygnul@xxxxxxxxx; si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Backdrill under BGA
>
>
> Sam,
>
> The critical factors missing here are your BGA pitch and location of the
> proposed back-drills relative to the BGA grid and each other. This as
> Back-drilled vias require larger anti-pads to accomodate the larger
diameter
> of the back-drill head that assures removal of the matal stub, as well as
> the minor mis-alignment tollerace for the back-drill. Note, that you will
> also lose routing channels on the back drilled layers due to the larger
> antipad and back-drill diameter.
>
> Looking at the drill size, anti-pad, routing channels, board thickness,
and
> ground/power plane coverage under the BGA, may indicate that back-drilling
> is not the preffered solution for fine pitch BGAs. Also, selection of the
> right routing layers and pad stack may allow you to get by with negligable
> stubs... assuming you do not have a boat load of multi-gig signals on a
> poorly pinned-out BGA.
>
> Porvide more details here or work with your fab house to arrive at
> the manufacturable solution.
>
> cheers,
> - David.
>
>
> ----- Original Message ----
> From: "cygnul@xxxxxxxxx" <cygnul@xxxxxxxxx>
> To: si-list@xxxxxxxxxxxxx
> Sent: Tuesday, July 21, 2009 5:46:21 PM
> Subject: [SI-LIST] Backdrill under BGA
>
> SI Gurus,
> Has anybody done back drilling on BGA vias for multi gig traces?
> is that achievable? If so, what was the via structure?
>
> -Sam
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List technical documents are available at:
> http://www.si-list.net
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
> List technical documents are available at:
> http://www.si-list.net
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List technical documents are available at:
http://www.si-list.net
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
