[SI-LIST] Re: [Fwd: coax to PCB transition - return loss]

  • From: "Loyer, Jeff" <jeff.loyer@xxxxxxxxx>
  • To: "Perry Qu" <perry.qu@xxxxxxxxxxx>, <scott@xxxxxxxxxxxxx>
  • Date: Tue, 22 Apr 2003 08:33:58 -0700

Ray Anderson put the information at:
http://groups.yahoo.com/group/si-list/files/Technical%20Documents/
under "SMA Launch Structure Info"

I always went to microstrip.  I would think your S11 resonance merely =
indicates a noisy launch.  I wouldn't be surprised to find that the =
microstrip launch would be noisier than that for coplanar - the =
transition (from SMA connector environment to PCB) is more complex and =
easier to induce discontinuities.

Jeff Loyer



-----Original Message-----
From: Perry Qu [mailto:perry.qu@xxxxxxxxxxx]
Sent: Tuesday, April 22, 2003 7:29 AM
To: scott@xxxxxxxxxxxxx; Loyer, Jeff
Cc: si-list
Subject: Re: [SI-LIST] Re: [Fwd: coax to PCB transition - return loss]


Scott, Jeff:

Thanks for your info. It's not clear to me what type of transmission =
line you
used on PCB, microstrip or coplannar waveguide ? Jeff, can you send me =
the
slides you are talking about ?

I re-examined some of our measured data for various coax-to-PCB =
interface and it
appears that whenever microstrip is used, I see some resonance in S11 =
frequency
domain plot (10 - 240 MHz). This resonance basically screw up my return =
loss out
of -15 dB. However, for coplannar waveguide, it appears to be a =
linearly-rising
curve with no resonance and it meets -15dB faily well.

BTW, I need to clarify a bit about the interface: The interface I'm =
talking
about here is the analogue link between SMZ connector and Tranceiver =
analogue
side. There are a bunch of passive elements in between (transformer, =
protection
diode, etc.). The "ground" in microstrip or coplannar waveguide is =
completely
isolated from PCB digital ground.

Regards

Perry

Scott McMorrow wrote:

> Bill,
>
> I'm not quite sure what you are asking.  An SMA by nature is a
> single-ended creature.  Measurements were made single-ended with one
> SMA, and differentially with two SMAs.  But in both cases single-ended
> cables and TDR heads were used.  The TDR profiles matched simulation
> with extremely good accuracy, including all high frequency resonances.
>
> In about a month we'll see how close the optimized SMA launch pattern
> matches simulation.  It will no doubt be better than our first try and
> will allow us to further refine our modeling and simulation =
techniques.
>
> scott
>
> bpanos wrote:
>
> >Scott-
> >Were your measurements made with a single ended line or a =
differential?
> >
> >-Bill
> >
> >
> >
> >Scott McMorrow wrote:
> >
> >
> >
> >> Ron,
> >>
> >>I have a vertical (top mount) SMA design that I've simulated and
> >>measured with the following characteristics:
> >>
> >>Insertion loss -1 dB @ 16 GHz
> >>Return loss -20 dB @ 12 GHz
> >>
> >>With this top mount sma we see a worst case TDR profile off a Tek
> >>TDS8000 17 ps head of 49 to 54 ohms.
> >>
> >>This design has been correlated to 3D full-wave simulation by
> >>measurement in both the time and frequency domain. As a result, I =
have
> >>redesigned the launch to have the following characteristics.
> >>
> >>Insertion loss -1 dB @ 27 GHz
> >>Return loss -30 dB @ 12 GHz
> >>-20 dB @ 22 GHz
> >>
> >>with a simulated TDR profile of 50 ohms +/- 1 ohm.
> >>
> >>Depending upon where the signal is that needs to be tested, the =
launch
> >>can either be done from top or the back side of the board, which =
reduces
> >>the via stub problem.
> >>
> >>regards,
> >>
> >>scott
> >>
> >>--
> >>Scott McMorrow
> >>Teraspeed Consulting Group LLC
> >>2926 SE Yamhill St.
> >>Portland, OR 97214
> >>(503) 239-5536
> >>http://www.teraspeed.com
> >>
> >>ron miller wrote:
> >>
> >>
> >>
> >>>Hi
> >>>
> >>>It looks like whether coplanar or coax the vertical mount is not to =
be
> >>>recommended.
> >>>
> >>>The big problem is the via,and the via capacitance, and the via =
stub.
> >>>This can go as low
> >>>as 30 ohms. Your worst case stub is .093 mils or about 10 ps one =
way or
> >>>20 ps round trip and
> >>>about two degrees 1/200 of error at 240 Mhz so it should probably =
be ok
> >>>unless you must have a very
> >>>accurate timing edge in your signal.
> >>>
> >>>However, be sure to remove internal pads, and keep the spacing to =
the
> >>>antipads as large as possible.
> >>>
> >>>For test purposes it is better to connect a piece of semirigid .047 =
or
> >>>.093 OD with a female sma
> >>>attached to two surface pads.
> >>>
> >>>Ron Miller
> >>>
> >>>dmitrik@xxxxxxxxxxx wrote:
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>>-------- Original Message --------
> >>>>Subject: [SI-LIST] coax to PCB transition - return loss
> >>>>Date: Wed, 16 Apr 2003 10:50:17 -0400
> >>>>From: "Perry Qu" <perry.qu@xxxxxxxxxxx>
> >>>>Reply-To: perry.qu@xxxxxxxxxxx
> >>>>Organization: Alcatel Canada
> >>>>To: si-list <si-list@xxxxxxxxxxxxx>
> >>>>
> >>>>
> >>>>Hi,
> >>>>
> >>>>I have been having problems in designing STM1 electrical interface =
to
> >>>>pass return loss requirement (< -15 dB up to 240MHz). We don't =
have a 3D
> >>>>full wave tool in house so I can't really simulate the interface =
and
> >>>>provide proper design guidelines. My question, what will be the =
best
> >>>>coax to PCB transition (75 ohm vertically-mounted SMZ connector):
> >>>>
> >>>>1. a microstrip, where center conductor of SMZ is connected to
> >>>>microstrip lines on PCB and the 4 ground pins are connected to the
> >>>>reference copper on PCB;
> >>>>
> >>>>2. A coplannar waveguide structure, ground - signal -ground, where
> >>>>signal is connected to center conductor and ground is connected to
> >>>>ground pins on SMZ.
> >>>>
> >>>>As long as we maintain 75 impedance control, I don't see much =
difference
> >>>>in terms of return loss performance for the above two.
> >>>>
> >>>>Also, how significant will through hole via along the signal path =
affect
> >>>>return loss at 240 MHz (assume 0.093" thickness) ?
> >>>>
> >>>>Thanks for your input.
> >>>>
> >>>>Regards
> >>>>
> >>>>Perry
> >>>>
> >>>>--
> >>>>Perry Qu
> >>>>
> >>>>SDN D & Q            |      600 March Road
> >>>>Alcatel Canada       |      Ottawa, ON K2K 2E6, Canada
> >>>>
> >>>>DID: (613) 7846720   |      FAX: (613) 5993642
> >>>>
> >>>>
> >>>>------------------------------------------------------------------
> >>>>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 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
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>--
> >>Scott McMorrow
> >>Teraspeed Consulting Group LLC
> >>2926 SE Yamhill St.
> >>Portland, OR 97214
> >>(503) 239-5536
> >>http://www.teraspeed.com
> >>
> >>------------------------------------------------------------------
> >>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 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
> >>
> >>
> >>
> >
> >
> >-- Binary/unsupported file stripped by Ecartis --
> >-- Type: text/x-vcard
> >-- File: bill.panos.vcf
> >-- Desc: Card for bpanos
> >
> >
> >------------------------------------------------------------------
> >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 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
> >
> >
> >
> >
>
> --
> Scott McMorrow
> Teraspeed Consulting Group LLC
> 2926 SE Yamhill St.
> Portland, OR 97214
> (503) 239-5536
> http://www.teraspeed.com
>
> ------------------------------------------------------------------
> 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 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
>

--
Perry Qu

SDN D & Q            |      600 March Road
Alcatel Canada       |      Ottawa, ON K2K 2E6, Canada

DID: (613) 7846720   |      FAX: (613) 5993642

------------------------------------------------------------------
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 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: