[SI-LIST] Re: IBIS Model error
- From: "Chris McGrath" <chris.mcgrath@xxxxxxxx>
- To: "Kim Helliwell" <kimgh@xxxxxxxxxxx>
- Date: Mon, 17 Nov 2003 10:14:42 -0800
Kim,
> Are other curves apparently swizzled as well?
None of the other curves appear to be "swizzled".
> You probably should contact the supplier of the model
> to clarify this (I bet you already are doing that, though).
We did contact the supplier and they're pretty much telling us that the
model is what it is and that it was created from the SPICE output.
Basically, it is up to me to figure out if there are any valid
counterarguments that could get them to investigate the model.
> Another possibility is that the curves are so close together=20
> that they cross at some point. But your evidence (a clear=20
> difference in sim results) suggests that this is not the case here.
On the rising edge, this actually happens so that both the rising and
falling curves appear to yield somewhat dubious results. One thing that
we have noticed is that the difference between min (slow) and max (fast)
is extremely tight as compared to several other PCI-X models that we
regularly use. It may be correct, but anything "too good to be true"
deserves a flag to be raised to make sure that something isn't in error
especially in light of the VI curve and the dV/dt_f observations.
Thanks,
Chris
> -----Original Message-----
> From: Kim Helliwell [mailto:kimgh@xxxxxxxxxxx]=20
> Sent: Monday, November 17, 2003 11:57 AM
> To: Chris McGrath
> Cc: si-list@xxxxxxxxxxxxx
> Subject: Re: [SI-LIST] Re: IBIS Model error
>=20
>=20
> It seems probable that the typ, min, max order was
> swizzled by the model creator. I've seen cases where
> the model creator thought the order was "max, typ, min"
> or "min, typ, max"
>=20
> Are other curves apparently swizzled as well?
>=20
> You probably should contact the supplier of the model
> to clarify this (I bet you already are doing that, though).
>=20
> Another possibility is that the curves are so close together=20
> that they cross at some point. But your evidence (a clear=20
> difference in sim results) suggests that this is not the case here.
>=20
> Kim
>=20
>=20
> On Nov 17, 2003, at 8:29 AM, Chris McGrath wrote:
>=20
> > Speaking of IBIS model problems...
> >
> > I have a model that has a couple of odd parameters that are
> > inconsistent
> > with all other models we have used.
> >
> > First, in the VI curve, the falling edge "maximum" is between the=20
> > "typical" and "minimum" curves.
> >
> > Second, dV/dt_r and dV/dt_f do not increase from minimimum=20
> to typical
> > to
> > maximum.
> >
> > In both cases when firing these drivers into test loads=20
> (PCI-X), they
> > do
> > not follow the expected trend where the "maximum" settings show the
> > fastest rise/fall time and the "minimum" settings show the slowest
> > rise/fall times.
> >
> > This operations seems very odd, but given my limited=20
> experience with=20
> > dissecting IBIS models I'm open to suggestions on why this=20
> component=20
> > behaves the way it does.
> >
> > Thanks,
> > Chris
> >
> >
> >
> >> -----Original Message-----
> >> From: Ingraham, Andrew [mailto:a.ingraham@xxxxxxxx]=3D20
> >> Sent: Monday, November 17, 2003 9:38 AM
> >> To: si-list@xxxxxxxxxxxxx
> >> Subject: [SI-LIST] Re: IBIS Model error
> >> =3D20
> >> =3D20
> >> Suresh,
> >> =3D20
> >> All these errors point to an incorrect IBIS model. Contact=3D20 =
IDT=20
> >> and ask them to fix it. =3D20
> >> The IBIS checker in the simulator tries to verify that the=3D20
> >> waveform data is consistent with the I-V tables. Ohm's law=3D20
> >> and the I-V tables should tell you the steady-state (initial=3D20
> >> and final) voltages, which ought to closely match those in=3D20
> >> the waveform tables. In this model, they do not come close.
> >> =3D20
> >> The I-V data doesn't seem to be pulling the output down. =3D20
> >> Even when driving a 50 ohm load resistor to 1.5V, the LOW=3D20
> >> state is 2.13V. Something is seriously wrong here.
> >> =3D20
> >> Regards,
> >> Andy
> >> =3D20
> >> =3D20
> >> ...
> >>> ERROR- Model IOPAD2546Z: The [Rising Waveform]
> >>> with [R_fixture]=3D3D50 Ohms and [V_fixture]=3D3D1.5V
> >>> has TYP column DC endpoints of 0.50V and 2.21v, but
> >>> an equivalent load applied to the model's I-V tables yields=3D20
> >> different=3D20
> >>> voltages ( 2.13V and 2.69V), a difference of 288.80% and=20
> 84.70%,=3D20=20
> >>> respectively.
> >> ...
> >> =3D20
> >> =3D20
> >> ------------------------------------------------------------------
> >> To unsubscribe from si-list:
> >> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the=20
> Subject field=20
> >> =3D20 or to administer your membership from a web page, go to:=3D20
> >> //www.freelists.org/webpage/si-list
> >> =3D20
> >> For help:
> >> =3D20
> >> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
> >> =3D20
> >> List technical documents are available at:
> >> http:/www.si-list.org
> >> =3D20
> >> List archives are viewable at: =3D20
> >> //www.freelists.org/archives/si-list
> >> or at our remote archives:
> >> http://groups.yahoo.com/group/si-list/messages=3D20
> >> Old (prior to June 6, 2001) list archives are viewable at:
> >> http://www.qsl.net/wb6tpu
> >> =3D20
> >> =3D20
> > ------------------------------------------------------------------
> > To unsubscribe from si-list:
> > si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the=20
> Subject field
> >
> > or to administer your membership from a web page, go to:=20
> > //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.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
> >
> >
> >
> Kim Helliwell
> kimgh@xxxxxxx
> Homepage: http://homepage.mac.com/kimgh
>=20
------------------------------------------------------------------
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.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
Other related posts:
