[SI-LIST] Re: 1000BaseT Common-mode output voltage test
- From: Douglas Smith <doug@xxxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Sat, 06 Aug 2011 09:00:37 -0700
Hi Matthew,
Is the common mode voltage due to the Ethernet signal or other noise
source? Have you confirmed the source, unbalanced Ethernet or common
mode other source? Poking around with spectrum analyzer settings can
confirm this.
Different brands of magnetics have different performance so you would
have to be using the exact same model of magnetics to expect the same
performance.
You say that you connected the grounds together. I find most attempts to
do this are ineffective. A 5 mm connection can be way too long in many
cases. I would need to see exactly how you did this.
Other that than, your description is too general. There are subtle
layout effects that can cause problems. To see an example of this look
at my latest Technical Tidbit article at:
http://emcesd.com/tt2011/tt080111.htm
Also, possibly the measurement setup has a problem. Pictures would be great.
By playing around with the terminations of the transformer center taps
(if possible) you can get a lot of information on the source of the
common mode voltage (through transformer capacitance or noisy ground for
instance).
Doug
On 8/6/11 8:34 AM, Matthew Fornero wrote:
> Hello All,
> I'm performing system evaluation on an embedded SBC using a couple of Intel
> ethernet controllers/PHYs 82577 and 82574). I am able to pass parts of the
> 1000BaseT compliance test, with the exception of common-mode output voltage.
> Here, the spec requires< 50 mV pk-pk, while I am measuring ~70-90 mV pk-pk.
>
> I've run the same tests on reference platforms for the two controllers, and
> they are able to pass without difficulty (< 25 mV pk-pk). Comparing my
> schematics and layout to the reference platforms, I can find no obvious
> differences other than the following:
>
> 1) Different grounding of the ethernet connector shield: The reference
> platforms tie the shield to digital ground (commercial boards), while mine
> is tied to frame ground/earth. Shorting the connector ground to digital
> ground does not seem to make any difference
>
> 2) Different connectors/magnetics: one reference platform uses an ethernet
> connector (Pulse I think) with built-in magnetics and USB ports, the other
> platform uses discrete magnetics. I am using a Tyco connector with built-in
> magnetics. All are rated for gigabit ethernet.
>
> I've looked over my circuit in general, and can't find anything obvious--
> the power delivery is good, and well within spec for the controllers, the
> MDI traces follow the Intel recommendations for isolation from aggressor
> signals, and are all length matched to within 10 mils.
>
> Any ideas what would be causing this level of common-mode output voltage?
>
>
> For reference, the common-mode output voltage test is described here:
> ftp://ftp.iol.unh.edu/pub/ethernet/test_suites/CL40_PMA/PMA_Test_Suite_v2.6.pdf
>
> Test 40.1.8 (IEEE Std 802.3-2008, clause 40.8.3.3)
>
> Best Regards,
>
> Matthew Fornero
>
>
> ------------------------------------------------------------------
> 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
>
>
>
--
--------------------------------------------------------------
___ _ Doug Smith
\ / ) P.O. Box 1457
========= Los Gatos, CA 95031-1457
_ / \ / \ _ TEL/FAX: 408-356-4186/358-3799
/ /\ \ ] / /\ \ Mobile: 408-858-4528
| q-----( ) | o | Email: doug@xxxxxxxxxx
\ _ / ] \ _ / Web: http://www.dsmith.org
--------------------------------------------------------------
------------------------------------------------------------------
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:
