[SI-LIST] Re: Test Loading Issues on Quad XTK

Dear Lawrence and others,

In my opinion, when considering the effects of TIME_TO_VM in XTK
simulations we need to focus on two operations:

I.  IBIS2XTK
II. XNS

How TIME_TO_VM comes into existence and influences results are
explained below.

I. When translating an IBIS model into QUAD format,  IBIS2XTK normally
computes TIME_TO_VM (by calling the "mtest" program) if the timing  test =
load=20
parameters (e.g.  Vmeas, Vref, Rref, Cref ) are present in the IBIS =
file.=20

However, to ensure reliable TIME_TO_VM values in XTK models, it is a =
good=20
practice to manually determine/check TIME_TO_VM values. This usually =
involves=20
obtaining the test load parameters ( Vmeas, Vref, Rref, Cref ) either =
from the=20
IBIS model (or preferably from the device data sheet), carrying out XNS =
simulation
and then measuring the rise and fall delays (by definition TIME_TO_VM is =
the time
it takes the driver output to cross the Vmeas voltage when the driver is =
connected to a=20
desired test load).=20


II. With the V(m) correction Enabled, XNS will subtract the driver =
TIME_TO_VM and
report the  corrected delays in .idd,  .rep and .gen files.  Such =
TIME_TO_VM
correction (to prevent double counting of buffer intrinsic delay) can =
prove critical when
flight time values are to be applied for setup and hold margin =
computations.
Furthermore, the delays measured by the XNS (or TLC) cursors differ from =
the delays
reported when TIME_TO_VM correction is enabled.  However, as noted by =
Jon Powell
TIME_TO_VM does not have any effects on signal quality such a
overshoot/undershoot, ringback, etc.

In summary, it is important that during the model =
translation/verification phase,=20
accurate TIME_TO_VM parameters be incorporated into the model's =
DRIVERSPEC.  This requires utilization of correct timing parameters ( =
Vmeas, Vref, Rref, Cref )
and measurement approach.   During the XNS simulations, with the V(m) =
correction
enabled the reported rise and fall delays (in .rep , .idd, and .gen =
files) differ from those
which can be manually measured by the XNS (or TLC) cursors.
 =20

Best Regards,

Abe Riazi
ServerWorks

-----Original Message-----
From:   lting168 [SMTP:Lawrence.Ting@xxxxxxxxxxxx]
Sent:   Monday, April 28, 2003 3:30 AM
To:     si-list@xxxxxxxxxxxxx
Subject:        [SI-LIST] Test Loading Issues on Quad XTK


Dear all,

 I am using Quad XTK to do simulation. When I get the rising delay or=20
falling delay data, those values should be substrated by the=20
Time_TO_VM values from the IBIS model of driver. However, if the IBIS=20
model doesn't include Time_TO_VM data, XTK automatically generates=20
the Time_to_VM values. Does anyone know how these values come from?

Thanks in advance

Lawrence



------------------------------------------------------------------
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                http://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: