[SI-LIST] Re: PCI Rise/Fall Time
- From: "Andrew Ingraham" <a.ingraham@xxxxxxxx>
- To: "si-list" <si-list@xxxxxxxxxxxxx>
- Date: Fri, 16 Feb 2007 13:00:17 -0500
> I have a design that has a 33MHz PCI bus. ...
> The setup and hold times of this PCI bus in my design are within the
> spec.; however, the rise and fall times (slew rates) are not. The rates
> are too slow (less than 1) compared to the spec (1 to 4).
There is no PCI bus spec for the slew rate on the bus.
The numbers you are looking at (1-4 V/ns) are in the section on "Component
Specifications," which are the specs for the ICs that are allowed to be used
on a PCI bus. It is the slew rate requirement of the output buffer inside
the IC ... not the slew rate once it gets out on a bus and interacts with
traces, connectors, other ICs, etc. That number could be anything, even
momentarily negative (plateaus, ringback).
So, I think you don't have a problem.
What matters is (1) that the clock IS monotonic and meets its slew rate spec
(which is a different slew rate spec) at each IC pin, and (2) that address/
data/control signals meet setup and hold times with respect to clock.
Between the hold time after one clock and the setup time before the next
clock, those signals might be anywhere (well, assuming we're not talking
about exceeding voltage limits or something). Also see section 4.2.4,
Indeterminate Inputs and Metastability.
> There's a serial termination of 33 Ohms at the beginning of the bus and
> an end termination of 3K at the end of the bus.
Are you talking about clock, or data signals here?
Usually discrete resistors are best avoided on the bussed data signals,
although once in a while they prove useful for devices mounted on the
motherboard. (Series resistors aren't OK on PCI plug-in cards.)
Generally, though, the source resistance should be taken care of by
controlling the drive strengths of the buffers inside the IC, making
external series resistors unnecessary. Though they can be good for clocks.
And your clocks are best done point-to-point, not daisy-chained. Perhaps
they already are.
Regards,
Andy
------------------------------------------------------------------
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
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:
