Arpad, Thank you for your explanation. It was not that I have a need for better modelling of the clamp. In fact, it was a customer who raised the question why the clamp curves were crossing. They seem to be satisfied with the answer I gave them. But it made me re-think the way I make IBIS models. Kind Regards, René Voshol "Muranyi, Arpad" <Arpad_Muranyi@mentor.com> Sent by: owner-ibis-users@server.eda.org 2008-08-08 07:09 PM To <ibis-users@server.eda.org> cc Subject RE: [IBIS-Users] Conditions for clamp and pullup/down tables Rene, You are raising very good questions, and your observation is correct. We were aware of this problem, but didn't concern ourselves too much because this clamping diode behavior is only effecting the overshooting signals. In general, designs with overshoot should not rely on the clamping diodes of the devices for various reasons, and because of that we shouldn't expect hair splitting accuracy from the simulations based on clamping diode behavior. In addition to that, the parasitic clamps of MOSFET models in SPICE tools are quite often very inaccurate due to inadequate modeling techniques which adds another reason for not relying too much on the claming effects. Also, more modern signaling schemes use terminations which means that the signals will not get into the active region of the clamping diodes in which case they can be neglected in the models. In summary, even though your observations are correct, we didn't feel that this was an important deficiency in the IBIS specification and models. If you can show us an example which would justify to make corrections to this situation we could certainly address this issue. Thanks, Arpad ============================================================= From: owner-ibis-users@server.eda.org [mailto:owner-ibis-users@server.eda.org] On Behalf Of Rene Voshol Sent: Friday, August 08, 2008 7:25 AM To: ibis-users@server.eda.org Subject: [IBIS-Users] Conditions for clamp and pullup/down tables Hi, I have a question about the conditions used to extract the data for the clamp and the pullup/down curves. I understand that the temperature range given in the model should correspond to the min. and max. conditions of the device. Meaning that for the pullup/down curve of a CMOS device you will have maximum performance with low temperature, and vice versa. But the clamp of this same the device will most likely be a diode, of some sorts, which will have the reverse temperature behavior. Since both curves are substracted, you need to apply the same conditions when extracting the curves. In which case I use the pullup/down curve as reference. My questions are: 1. Is this correct thinking? (Or am I perhaps overthinking things here?) 2. What about an input model? This has only clamp curves. Is it OK to use a different temperature ranges for input and output of the same device? 3. Or a switch where the clamps and the series current are in separate models? Clarifications on this matter will be highly appreciated. Kind Regards, René Voshol -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -------------------------------------------------------------------- |For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org |with the appropriate command message(s) in the body: | | help | subscribe ibis <optional e-mail address, if different> | subscribe ibis-users <optional e-mail address, if different> | unsubscribe ibis <optional e-mail address, if different> | unsubscribe ibis-users <optional e-mail address, if different> | |or e-mail a request to ibis-request@eda-stds.org. | |IBIS reflector archives exist under: | | http://www.eda-stds.org/pub/ibis/email_archive/ Recent | http://www.eda-stds.org/pub/ibis/users_archive/ Recent | http://www.eda-stds.org/pub/ibis/email/ E-mail since 1993Received on Tue Aug 12 01:34:18 2008
This archive was generated by hypermail 2.1.8 : Tue Aug 12 2008 - 01:35:10 PDT