I have been giving this situation some thought and would make the
following recommendations:

1.)     Set the backplane impedance at 50-75 ohms.  It would be
impractical to try to set the characteristic impedance to that of the
board since any of the eight *parallel* modules can be the driver. This
would require complex routing and impedance calculation models which
would take far too long to develop.
2.)     I would make it clear to the responsible engineer(s) that they
need to include termination for the incoming signals.  This would
probably be a pseudo-terminator of clamping diodes, as the amount of
actual load/discontinuity would be based on the position of the card
driving the signal in relation to the receiving card/cards.

Jeffrey McGlaughlin
Battelle Memorial Inst.
Columbus, Ohio

(The opinions expressed are strictly my own, and may not represent those
of any sane individual.;<D)

        -----Original Message-----
        From:   Doug McKean [SMTP:[log in to unmask]]
        Sent:   Saturday, February 14, 1998 6:03 PM
        To:     [log in to unmask]
        Subject:        Re: [DC] Controlled Impedance of a Backplane ...

        <hitting hand on forehead>
        Should have thought of that one...

        The shelf will carry either T1 or E1 telco in,
        but some of the processing is pretty quick.
        So how about 1 nsec rise time for the telco
        but other rates I'm guessing could be down
        in the 10's of psec or maybe even less.

        Oh, what the heck.  Let's say for the sake of
        argument slowest is 1 nsec for the telco,
        fastest will be 1 psec for processing and
        control signals.

        Regards,  Doug

        ----------
        > From: Doug Brooks <[log in to unmask]>
        > To: [log in to unmask]
        > Subject: Re: [DC] Controlled Impedance of a Backplane ...
        > Date: Saturday, February 14, 1998 4:27 PM
        >
        > GOOD QUESTION
        >
        > First I have two questions back
        > 1) what is the fastest rise time we are dealing with here?
        > 2) is there a driver at the edge of each card where the signal
        >    leaves the card? If not, how far does a driven signal
travel
        >    before it leaves the card and goes onto the backplane?
        >
        > Doug Brooks
        >
        > At 01:07 PM 2/14/98 -0500, you wrote:
        > >We have a product that is a shelf with a backplane into
        > >which plugs 2 power supplies and 8 modules. The 8 modules
        > >are all in parallel to one another and are capable of
        > >talking to one another through the backplane.  A single
        > >module contains 2 boards that are each 50 ohm boards.
        > >These two board are also in parallel with each other.
        > >
        > >16 connectors in the backplane for 16 PCBs each
        > >50 ohms. For any type of configuration, a single
        > >50 ohm board is capable of talking to as many as
        > >2*N + 1  other 50 ohm boards where N = # modules in shelf
        > >and  1 = the other board in its own engine.
        > >
        > >In other words,
        > ># modules in shelf      # of other 50 ohm boards one board
        > >                                  talks to through the
backplane
        > >                          dependent upon configuration of
        > >                          shelf.
        > >
        > >1 module                         1 other 50 ohm board
        > >2 modules                        3 other 50 ohm boards
        > >3 modules                        5 other 50 ohm boards
        > >  ...                   ...
        > >8 modules                       15 other 50 ohm boards
        > >
        > >2 connectors of the same module are about 3/4 inch apart.
        > >2 connectors of different modules are about 5/8 inch apart.
        > >
        > >        Module #1          Module #2
        > >      |<-------->|<------>|<-------->|
        > >          3/4"      5/8"
        > >
        > >      0          0        0          0
        > >      0          0        0          0
        > >      0          0        0          0
        > >      0          0        0          0
        > >      ...        ...
        > >
        > >What should the impedance of the backplane be?
        > >50 ohms?  100 ohms?  800 ohms?
        > >
        > >
        > >Regards, Doug
        > >
        > >Private emails welcome
        > >
        > >mailto:[log in to unmask]
        > >
        >
>################################################################
        > >DesignerCouncil E-Mail Forum provided as a free service by
IPC using
        > LISTSERV 1.8c
        >
>################################################################
        > >To subscribe/unsubscribe, send a message to [log in to unmask]
with
        following
        > text in the body:
        > >To subscribe:   SUBSCRIBE DesignerCouncil <your full name>
        > >To unsubscribe:   SIGNOFF DesignerCouncil
        >
>################################################################
        > >Please visit IPC web site (http://jefry.ipc.org/forum.htm)
for
        additional
        > information.
        > >For the technical support contact Dmitriy Sklyar at
[log in to unmask] or
        > 847-509-9700 ext.311
        >
>################################################################
        > >
        > >
        > >
        >
        >
################################################################
        > DesignerCouncil E-Mail Forum provided as a free service by IPC
using
        LISTSERV 1.8c
        >
################################################################
        > To subscribe/unsubscribe, send a message to [log in to unmask]
with
        following text in the body:
        > To subscribe:   SUBSCRIBE DesignerCouncil <your full name>
        > To unsubscribe:   SIGNOFF DesignerCouncil
        >
################################################################
        > Please visit IPC web site (http://jefry.ipc.org/forum.htm) for
additional
        information.
        > For the technical support contact Dmitriy Sklyar at
[log in to unmask] or
        847-509-9700 ext.311
        >
################################################################

        ################################################################
        DesignerCouncil E-Mail Forum provided as a free service by IPC
using LISTSERV 1.8c
        ################################################################
        To subscribe/unsubscribe, send a message to [log in to unmask]
with following text in the body:
        To subscribe:   SUBSCRIBE DesignerCouncil <your full name>
        To unsubscribe:   SIGNOFF DesignerCouncil
        ################################################################
        Please visit IPC web site (http://jefry.ipc.org/forum.htm) for
additional information.
        For the technical support contact Dmitriy Sklyar at
[log in to unmask] or 847-509-9700 ext.311
        ################################################################

################################################################
DesignerCouncil E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE DesignerCouncil <your full name>
To unsubscribe:   SIGNOFF DesignerCouncil 
################################################################
Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################