Mikhail, We performed signal integrity simulations for various CompactPCI switch fabric architectures using the 2MM connectors: PICMG 2.16 (10/100/1000 Base-T Ethernet) PICMG 2.17 (StarFabric SERDES at 622 Mbps) PICMG 2.18 (Serial RapidIO SERDES at 1.25 Gbps, and 2.5 GBPS also studied) We have also performed signal integrity studies for a number of custom backplane interconnects in these same frequency ranges. One major issue with the 2MM connector is that it was not originally designed to operate at these high rates, and there is a very wide variation in electrical characteristics from one vendor to another, in terms of both crosstalk and impedance. With the worst vendors that we have seen, 1.25 Gpbs is marginal, so 2.5 Gbps would be problematic. If you are willing to surround the active pairs with grounds on all sides, restrict the SERDES rise times, and restrict the connector vendors to only the best ones, then 2.5 Gbps might be possible, depending on your specific SERDES characteristics. Note that fast SERDES rise times will exascerbate the connector problems. If you plan to go down this path, signal integrity simulations should be performed. The major connector vendors all have SPICE models that you can get under NDA, but the last time I checked, only a few vendors had models that were characterized for fast rise times. Regards, Bob Sullivan Vice President and Corporate Director of Technology Hybricon Corporation 12 Willow Road Ayer, MA 01432 Tel: 978-772-5422 x250 Fax: 978-772-7570 Cell: 978-502-0948 -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Mikhail Matusov Sent: Friday, March 18, 2005 5:45 PM To: si-list@xxxxxxxxxxxxx Subject: [SI-LIST] 2.24 Gb/s over standard 2mm HM connectors? Dear experts, I am designing a system where two Xilinx FPGAs on 2 separate PCBs will be connected with a differential serial 2.24 Gb/s RocketIO link. The cards are so called front and rear Compact PCI cards and are expected to go into a standard PICMG 2.16 chassis. Thus, they will be connected with a 2mm Hard Metric "pass-through" connector. In other words there will be a backplane or rather a midplane between the cards, but no tracks on this midplane, the signals will pass directly through the vertical male connector pins sticking out on both sides of the plane, while each of the cards will have a right angle female connector. My questions are as follows: 1. Has anyone done this successfully? 2. Which connector pins should I use for signaling and which for GND? 3. Can I get away without simulating this? If not, could someone please point me to where I can get appropriate SPICE(?) models for the connectors? Thanks, ======================= Mikhail Matusov Hardware Design Engineer Square Peg Communications Tel.: (613) 271-0044 ext.231 Fax: (613)271-3007 http://www.squarepeg.ca ------------------------------------------------------------------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org 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 ------------------------------------------------------------------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org 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