RE: Rising/Falling waveform and Ramp keywords

Hello Offer:

   The [Ramp] keyword was in the original (IBIS ver 1.1) specification and,
at the time, only resistive loads to VCC and GND were allowed. When the
waveform tables were introduced in ver 2.2 the L/C/R_dut and L/C/R_fixture
subparameters were added so that a model creator could document a test bench
setup. However -- and this is important -- the IBIS cookbook strongly
recommends that the [Rising Waveform] and [Falling Waveform] data be taken
with 1) no package and 2) into a purely resistive load. In other words, to
build the most accurate simulator model the waveform data should be taken
from a SPICE simulation of the buffer driving resistive loads only.

(Data from a waveform taken with the buffer driving package and fixture
L/R/Cs can be included in the IBIS data file as a "Golden Waveform". The
idea is that the user can set up a simulation with the buffer driving the
package and fixture L/R/Cs, then see how closely the simulated waveform
matches the expected waveform documented in the IBIS file.)

Note that for a standard CMOS buffer the cookbook recomends including four
waveforms: two rising, one with the buffer driving 50 ohms to VCC and
another with the buffer driving 50 ohms to gnd, and two falling, same
conditions. For more info, check out the IBIS cookbook at

 http://www.eigroup.org/ibis/tools.htm

  Regards,
  Stephen Peters
  Intel Corp.

-----Original Message-----
From: Offer Kaye [mailto:sofferk@techst02.technion.ac.il]
Sent: Saturday, December 09, 2000 5:17 AM
To: ibis-users@eda.org
Subject: Rising/Falling waveform and Ramp keywords

Hello IBIS gurus.
My question is this- in the IBIS spec, the Rising/Falling Waveform
keywords include the sub-parameters R/L/C/V_fixture and R/L/C_dut, while
the Ramp keyword does not.
I want to measure these parameters (from a packaged component), and I
don't understand why the difference in specs for keywords which it was my
understanding should be measured the same way (only different data used).

How should my test setup be different for the Waveforms vs. Ramp
measurements? Can someone explain in greater detail (than the spec) the
meaning of the R/L/C/V_fixture subparameters?

Thanks in advance,
Offer Kaye
sofferk@techst02.technion.ac.il

 
Received on Mon Dec 11 14:35:47 2000