[SI-LIST] Re: SMA Connectors

Hello Padma,

Scott is correct (as usual!), and I will provide some additional
information. We actually use quite a few of those Molex connectors in
our Final Inch test board kits and in other test applications, so we are
very familiar with it.  It is a unique connector that fills a useful
niche.

The Molex connector is optimized for use at higher frequencies than the
Samtec part you mentioned.  As such, it is most likely manufactured to
tighter tolerances (although I am not privy to Molex's manufacturing
methods and tolerance limits).  That will certainly drive up the cost.
It also uses some more costly materials.   And as Scott mentioned, the
Molex part is a semi-custom design, which drives up cost as well.  I
expect the volumes Molex sells are fairly low too.  So there are quite a
few reasons it would cost more.

As background, the original Molex design consisted of a longer pin and
socket arrangement for connection to the PCB.  A socket (or pin, I
forget which) was soldered into the PCB.  The center conductor of the
connector would mate into that terminal.  So in effect, the top assembly
with the SMA interface was replaceable, with the soldered pin remaining
permanently in the PCB.    The semi-custom design basically trims the
connector center conductor at the top of the PCB, and eliminates the
soldered socket.  This turns the center conductor / PCB connection into
a compression type interface.

This approach offers some significant SI advantages, and allows the
total connector/launch bandwidth (based on return loss) to be pushed
higher in frequency.  Thru hole center conductor connectors generally
perform less well than surface mount types, and this is almost entirely
due to effects which occur in the PCB.  Most of us are familiar with the
negative effects of pin and via stubs.  These issues can be minimized
with stopped or back drilled vias, which are much easier to implement
with surface mount parts.  In addition, the surface mount or compression
type interface also gives us the ability to use a smaller diameter via
down to lower signal layers which allows more leeway in tuning the
impedance of the launch.

The Samtec connector you mentioned is very similar to the original Molex
design, in that it requires a pin that is permanently soldered into the
pcb.   With out back drilling, it is possible to push the performance of
that connector/launch combo to around 7-8 GHz or so.  With the Molex
parts, we can get launch/connector designs that function well to 20 GHz.
Again, these numbers assume return loss or VSWR is the primary issue, as
is often the case in SI measurements.  If your primary design concern is
another parameter such as insertion loss or isolation, then both
connectors can be used to higher frequencies.  

One downside to the "quick and dirty" compression pin design is that it
is less reliable than a soldered joint.  It is more susceptible to
intermittent connections and corrosion over time.  This is usually not
an issue in an SI lab environment, where an engineer is sitting in front
of an instrument and can observe the quality of the connection in real
time, and tweak it as necessary.   

The connector / PCB interface also tends to "wear out" with repeated
applications and removal from test boards.  The center conductor can
push up into the connector body, which can lead to intermittent
connections.  And we have learned that you should not subject these
connectors to severe temperature cycling when mounted on a PCB.   But
again, in a typical SI test lab environment, we find these downsides to
be perfectly acceptable trade offs for the increased bandwidth.

We have experimented with some compression type designs of our own. We
were able to easily improve the performance in the launch area.  But
without tightening up the tolerances and changing some materials in
other areas of the connector, we didn't get as much bang for the buck
when the entire launch/connector area was considered.  We might continue
development work on an improved version that will hopefully address the
reliability issues, while still offering the bandwidth improvements.
But such a design will no doubt end up costing more, and we're still a
bit uncertain about the overall demand for such a product.

So in the mean time, if your application requires excellent return loss
performance to 20 GHz or so, and you can live with potential
intermittencies, then the Molex connector is a good choice.  If you
require better long term reliability, and/or lower cost, and if return
los performance above 8-10 GHz is not a concern, then the Samtec
connector is certainly worth considering.


I hope you find this information useful.

Julian Ferry
High Speed Engineering Manager 
Samtec, Inc.

-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On Behalf Of Scott McMorrow
Sent: Wednesday, May 27, 2009 6:08 PM
To: padma gundala
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: SMA Connectors

Padma

The Molex connector was one that I had George Hubbard design for me back

in 2004, when he worked at Molex.  It takes a precision flange mount 
design and incorporates a built in beryllium copper pin that forms a 
signal contact with pad on a PCB.  It was designed specifically to be a 
low return loss connector for precision measurements.

The Samtec flange mount connector is designed to be a low cost 
connector, similar to other standard SMA flange mount connectors in the 
industry.  The Samtec boys in Harrisburg can give you any information 
about it that you desire.  Depending on your application, it may meet 
your needs.

regards,

Scott


padma gundala wrote:
> Hi,
>  
> Anyone know the technical reason why one costs $40 and the other $3.75
>  
> Molex SMA Jack/connector :P/N 73251-1850       $40
> and 
> Samtec SMA Jack :P/N SMA-J-P-H-ST-PN4       $4
> Both are 50ohms, Samtec rates 12.5 GHZ.
>  
> Any inputs from people who have experience using any of the above
> is appreciated.
>  
> Thanks a lot.
>  
> Padma
>
>       
> ------------------------------------------------------------------
> 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:
> http://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:     
>               http://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
>   
>
>   

-- 
Scott McMorrow
Teraspeed Consulting Group LLC
121 North River Drive
Narragansett, RI 02882
(401) 284-1827 Business
(401) 284-1840 Fax

http://www.teraspeed.com

Teraspeed(r) is the registered service mark of
Teraspeed Consulting Group LLC

------------------------------------------------------------------
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:
http://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:     
                http://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:
http://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:     
                http://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: