lead free bga pads

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
Hi all

I'm having trouble finding pcb layout recomendations on the altera and
xilinx web sites for lead free bga's. I'm wanting to use an altera F256
package and I can find all I need for the leaded version but nothing
for the rohs packages. Does anyone know where I should be looking or
what changes to the leaded footprints I should make?

all the best in 07

Colin


Re: lead free bga pads
Quoted text here. Click to load it

I'm facing that issue right now for a FT256 package, and I'll do what
I've done with all the other RoHS BGA parts I've had done leadfree (and
that's quite a few) - use non-SMD pads with the recommended sizes from
the leaded versions.

If my manufacturer thinks it needs to be tweaked - after all, they do a
*lot* of BGAs - they'll let me know.

So far I haven't had any problems provided I use a gold finish; of
course, I haven't had sufficient time to get decent long term
reliability data.

Cheers

PeteS


Re: lead free bga pads

Quoted text here. Click to load it

Thanks Pete, I will do the same.

Colin


Re: lead free bga pads
This link might be useful!
Quoted text here. Click to load it
Hi Vasile,
Good link! As you say, I think the OP meant something else, but thanks for
this! I'll add it to the RoHS BGA thread.
Cheers, Syms.



Re: lead free bga pads
Quoted text here. Click to load it

You're welcome.

As it happens, after re-evaluating the design, I am going to plonk down
a 456 ball RoHS compliant BGA (and 2 256 parts plus some - 6 -  at about
100 balls apiece), so I'll let you know how it all turns out.

Cheers

PeteS

Re: lead free bga pads
Quoted text here. Click to load it
Hi Pete,
I read somewhere* that engineers used solder mask defined (SMD) pads for
early BGAs because the parts were made of ceramic and could pull the pads
off the board during vibration testing. With a SMD pad I guess the pad area
can be larger to stop this. When lighter packages came along, the SMD pads
were no longer needed, but folks carried on with SMD pads maybe because**
"that's the way we've always done it". However, it turns out that SMD pads
have a proclivity for microvoids*** at the pad/solder boundary.
So, I wonder, now that Xilinx BGAs have metal heat spreaders built in,
whether their mass has increased enough to make the SMD pad necessary once
again. What do you think?

Cheers, Syms.

* http://www.smta.org/knowledge/proceedings_abstract.cfm?PROC_ID25%2
** http://www.ojar.com/view_2857.htm
*** http://www.microvoids.org/



Re: lead free bga pads
Quoted text here. Click to load it

I've been bitten *very* hard from SMD pads on BGAs, especially fine
pitch stuff. I've put down a lot of devices with heat spreaders on
non-SMD and never had a failure due to reflow, but I've had a *lot* of
failures (microvoids -> HiZ, and shorts pad to pad) using SMD whether
they had heat spreaders or not.

Unless there is a compelling reason for SMD pads (some specialised
manufacturers specify it and I *always* ask for the reason) I won't use
them.

Cheers

PeteS

Re: lead free bga pads

Quoted text here. Click to load it
Hi Pete,
Thanks for your comments. It more or less reflects my experiences, although
I've not vibration tested an FPGA with heat-spreader design.

On a related subject, what do you think about via-in-pad. I've had great
success with this, but recently people seem to be shying away from it.

http://www.pcblibraries.com/resources/GEN-docs.asp
Near the bottom, Solving the Metric Pitch BGA & Micro BGA Dilemma

On page 30, the guy says that the voids are no problem, which backs up my
experience. I wondered if you have a view on this?

Thanks, Syms.



Site Timeline