[SI-LIST] Re: PCB thin dielectrics
- From: steve weir <weirsi@xxxxxxxxxx>
- To: leeritchey@xxxxxxxxxxxxx
- Date: Mon, 21 Dec 2009 13:28:09 -0800
1 mil glass epoxy tends to be very twitchy stuff. I don't know of
anyone who has ever gotten good yields with it.
The closer you get to saturating the spreading inductance between the
bypass caps and the IC interconnect, the more valuable thinner
dielectric becomes. If one has a low performance requirement where a 40
mil cavity will work there is little to be gained performance wise by
using something that is 10X, 20X or 40X thinner. It also doesn't take
much in the way of bypass cap density to saturate such a thick cavity.
By a similar token if one has a 4mil cavity that works great with just a
few bypass caps, the gains from thin dielectric will be elusive.
However take any of the many cases where with 3-4mil cavities that
typically use enough caps to just equal the spreading inductance from
the IC attach to where the capacitors can be attached. If you go from 4
mil to 1 mil you can achieve the same electrical performance from 5MHz
to the PDN / power cavity cross-over frequency with 57% as many caps or
a 43% savings in component, and placement cost, not to mention freeing
up real-estate right where you want to break-out the the signal traces.
It also opens other options as to how to assign the power polygons and
do the stack-up. There are many cases where this allows fewer layers in
the design.
If you design using fewer caps and the thinner cavity then the cross
over frequency between the PDN and the cavity comes down substantially,
making it easier to manage the resonance at the transition. This is
becoming a more significant issue as memory busses now operate at bit
rates / 2 that can land squarely on the transition. Above the
transition, the characteristic impedance follows Hthin/Hthick * (
eRthick/eRthin )^0.5. For either the duPont HK04 polyimide or the
Oak-Mitsui epoxy based materials the eR is fairly close to that of glass
/ epoxy and so the characteristic impedance falls almost linearly with
the dielectric thickness. Even better, the thinner the dielectric, the
lower the cavity Q and so the lower ratio of 1/2 wave modal impedances
to the characteristic impedance. For example with the 24um HK04
material the modal peaks run about 1/6th ( -16dB ) the impedance of an
identically patterned 4 mil FR406 or similar cavity.
I've worked with a number of duPont customers who are very happy with
the improvements thin dielectrics have made to their designs be it for:
Cost, performance, and/or yield / reliability or a combination of all.
I hope your seminars in Sweden went well.
Best Regards,
Steve.
Lee Ritchey wrote:
> There is also 1 mil material available from Sanmina SCI if you are
> determined to use such thicknesses. My experience is that it is not
> necessary to do so.
>
> Lee Ritchey
>
>
>
>> [Original Message]
>> From: Andresakis, John <John.Andresakis@xxxxxxxxxxxxx>
>> To: Lee Ritchey <leeritchey@xxxxxxxxxxxxx>; Steve Weir
>>
> <weirsi@xxxxxxxxxx>; Bowden, Ivor <ibowden@xxxxxxxxxxxxxxxxx>
>
>> Cc: si-list@xxxxxxxxxxxxx <si-list@xxxxxxxxxxxxx>
>> Date: 12/18/2009 10:59:49 AM
>> Subject: RE: [SI-LIST] Re: PCB thin dielectrics
>>
>> Gentlemen,
>>
>> As Lee points out, it is an engineer's job to avoid single sourcing when
>>
> possible. But as Steve indicated, there are 2 sources of 1 mil material.
> The HK04 (DuPont) and BC24 (Oak-Mitsui) products are used by a number of
> companies as interchangeable. Ivor, for what it sounds like what you are
> trying to do, either material would be acceptable.
>
>> Let me know if you further information on our BC24 or other materials.
>>
>> Best Regards,
>>
>> John Andresakis
>> VP of Strategic Technology
>> Oak-Mitsui Technologies, LLC
>> 80 1st Street Hoosick Falls, NY 12090
>>
>> Office: (518) 686-8088
>> Cell: (518) 368-1556
>>
>>
>> -----Original Message-----
>> From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
>>
> On Behalf Of Lee Ritchey
>
>> Sent: Friday, December 18, 2009 1:17 PM
>> To: Steve Weir; Bowden, Ivor
>> Cc: si-list@xxxxxxxxxxxxx
>> Subject: [SI-LIST] Re: PCB thin dielectrics
>>
>> Problem with most of these materials is they are single source with all
>>
> that entails. I do just fine with standard laminates by using thin
> prepregs between the power planes. This does not result in single
> sourcing. I know that puts a crimp in those companies that offer special
> materials, but an engineer's job is to do what he can to avoid single
> sources when doing designs.
>
>> Sorry Steve! Didn't mean to step on you here.
>>
>> Lee Ritchey
>>
>>
>>
>>> [Original Message]
>>> From: steve weir <weirsi@xxxxxxxxxx>
>>> To: Bowden, Ivor <ibowden@xxxxxxxxxxxxxxxxx>
>>> Cc: <si-list@xxxxxxxxxxxxx>
>>> Date: 12/18/2009 9:24:18 AM
>>> Subject: [SI-LIST] Re: PCB thin dielectrics
>>>
>>> Ivor, there are about four materials to know about:
>>>
>>> Except for exotic applications the big thrill in thin dielectrics is
>>> reduction in inductance. This lets you get rid of a bunch of bypass
>>> caps, and/or yield lower impedance to the power attachment of your IC
>>> package, and in a number of cases do things like reduce layer counts
>>> depending on the specifics of your power distribution scheme. With
>>> proper design a thin dielectric lets you use far fewer bypass
>>> capacitors to cover the frequency ranges from about 5MHz up to around
>>> 100MHz - 300MHz where the bypass network impedance crosses over into
>>> the interconnect impedance. Above the crossover, the thinner
>>> dielectric gives you lower impedance period. For unloaded materials
>>> the impedance improvement is almost linear to the ratio of the
>>> thickness: IE 1mil material is about 1/3 the impedance of a 0.0035"
>>>
> glass epoxy cavity.
>
>>> The impedance peaks caused by modal resonances come down faster than
>>> the thickness ratio.
>>>
>>> In the thinner than 0.0035 camp there is:
>>>
>>> $ BC2000 50um
>>> $$ DuPont HK04 24um polyimide / Oak Mitsui BC24 24um epoxy $$$DuPont
>>> HK04 18um polyimide $$$$Dupont HK04 12um polyimide / Oak Mitsui BC12
>>> epoxy $$$$$$OakMitsui BC12T BT loaded epoxy / Oak Mitsui BC8 8um epoxy
>>> / 3M Cply BT loaded epoxy
>>>
>>> I provide application support to DuPont for HK04 materials. The 24um
>>> material is cheap enough now that in many applications it reduces
>>> total cost versus an equivalent performance bypass system using
>>> thicker dielectric.
>>>
>>> Best Regards,
>>>
>>>
>>> Steve.
>>>
>>> Bowden, Ivor wrote:
>>>
>>>> Hi SI people,
>>>>
>>>>
>>>> Could I get some comments on the use of "thin" (<0.0035")
>>>> dielectrics
>>>>
>> in PCBs, in terms of available materials, cost, reliability, dielectric
>>
> strength, trace width vs dielectric constant, power plane capacitance,
> stories, studies, useful web links, any other pertinent issues? Base
> question is, could this be a practical way to increase PCB layer count
> while maintaining overall PCB thickness?
>
>>>>
>>>> Any / all information / comments welcome.
>>>>
>>>>
>>>>
>>>> Thanks,
>>>>
>>>>
>>>>
>>>> Ivor Bowden
>>>>
>>>>
>>>>
>>>>
>>>> ____________________________________________________________________
>>>> ___ This e-mail and any files transmitted with it are proprietary
>>>> and
>>>>
>> intended solely for the use of the individual or entity to whom they are
>>
> addressed. If you have reason to believe that you have received this e-mail
> in error, please notify the sender and destroy this email and any attached
> files. Please note that any views or opinions presented in this e-mail are
> solely those of the author and do not necessarily represent those of the
> Curtiss-Wright Corporation or any of its subsidiaries. Documents attached
> hereto may contain technology subject to government export regulations.
>
>> Recipient is solely responsible for ensuring that any re-export, transfer
>>
> or disclosure of this information is in accordance with applicable
> government export regulations. The recipient should check this e-mail and
> any attachments for the presence of viruses. Curtiss-Wright Corporation and
> its subsidiaries accept no liability for any damage caused by any virus
> transmitted by this e-mail.
>
>>>> ------------------------------------------------------------------
>>>> 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
>>>>
>>>> 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
>>>
>>> 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
>>
>> 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
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
