Fast-20 Active Negation Drivers

Tak Asami asami at dt.wdc.com
Mon Jan 30 16:54:57 PST 1995


Thank you for all the help from Bill Ham, John Lohmeyer, Paul Aloisi 
and Kevin Gingerich.

I now understand how it works and how it is different from Fast SCSI.
We have run several simulations to verify the theory, and agree with you.
It seems we need twice as much pullup current for heavily loaded Fast-20.
I agree with John Lohmeyer's proposal (shown below) to cut the shoulder 
off the "minimum side" rectangle (I am impressed with John's talent to
draw a smooth curve using teletype graphics).

There are two concerns:
- Do we want to keep 2.0V @ Ioh=7mA requirement?
  If a driver is powerful enough to supply sufficient current in transition,
  then the requirement is almost automatically satisfied, but is this DC
  spec still required?  Perhaps not, since the value of active negation is
  really in the rise transit.  So John's picture (if it is drawn 
  with finer resolution graphics) is totally agreeable.
- The load model on (current) Figure 1 is totally inadequate to be used
  for slew rate and AC transition characteristics design / validation.
  I prefer to use (unfortunately) a transmission line model like shown
  below as Figure 2.


> [ASCII rendition of figure 1:]
>                                                      3.0V
>       |                                              |xxxxxxxxxxxxxxxxxx
>       |                                              |xxxxxxxxxxxxxxxxxx
> 30 mA |                                              |xxxxxxxxxxxxxxxxxx
>       |                                              |xxxxxxxxxxxxxxxxxx
>       |                                              |xxxxxxxxxxxxxxxxxx
>       |                                              |xxxxxxxxxxxxxxxxxx
> 22 mA +                                              |xxxxxxxxxxxxxxxxxx
> 20 mA |\___                                          +---+xxxxxxxxxxxxxx
>       |xxxx\___                                          |xxxxxxxxxxxxxx
>       |xxxxxxxx\__                                       |xxxxxxxxxxxxxx
>       |xxxxxxxxxxx\___                                   |xxxxxxxxxxxxxx
>       |xxxxxxxxxxxxxxx\__                                |xxxxxxxxxxxxxx
> 10 mA |xxxxxxxxxxxxxxxxxx\__                             |xxxxxxxxxxxxxx
>       |xxxxxxxxxxxxxxxxxxxxx\___                         |xxxxxxxxxxxxxx
>  7 mA |xxxxxxxxxxxxxxxxxxxxxxxxx\__                      +--------+xxxxx
>       |xxxxxxxxxxxxxxxxxxxxxxxxxxxx\__                            |xxxxx
>       |xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx\                           |xxxxx
>  0 mA +---------------+---------------+---------------+---+-------+-----
>       0V              1V              2V              3V  3.24V   3.7V
> 
>               Figure 1 -- Active Negation Current vs. Voltage


 0.3m x 10
  Zo=90ohm    25pFx8           Driver under test
 +-====-+-====-+-====-+-====-+-====--+--====-+-====-+-====-+-====-+-====-+
_|_     |      |      |      |       |       |      |      |      |     _|_
|T|    ---    ---    ---    ---   -|>+      ---    ---    ---    ---    |T|
|_|    ---    ---    ---    ---             ---    ---    ---    ---    |_|
 |      |      |      |      |               |      |      |      |      |
 +-====-+-====-+-====-+-====-+-====--+--====-+-====-+-====-+-====-+-====-+
                                     |                            2.5V 110ohm
                                     V GND                        Terminatorx2

                 Figure 2 -- Transition Load model
                   (How's my teletype graphics?)


So any one of us can generate the proposal.
John, since you have taken a shot at it, would you like to do that?
I will be happy to do that, if that is more convenient for you.

Tak Asami =================================================================
Western Digital Corp
I/O Product Engineering
asami at dt.wdc.com
(714) 932-7621 : Voice
(714) 932-6496 : Fax




More information about the T10 mailing list