From: owner-ibis-users@eda.org (ibis-users) To: ibis-users-digest@eda.org Subject: ibis-users V1 #177 Reply-To: Sender: owner-ibis-users@eda.org Errors-To: owner-ibis-users@eda.org Precedence: bulk ibis-users Sunday, June 12 2011 Volume 01 : Number 177 ---------------------------------------------------------------------- Date: Fri, 10 Jun 2011 06:37:20 -0700 From: "Muranyi, Arpad" Subject: RE: [IBIS-Users] IV clamp range This is a multi-part message in MIME format. - ------_=_NextPart_001_01CC2773.8647A320 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Vinyak, =20 The crossing you show is perfectly normal. The knee voltage is determined by the P-N junction (parasitic diode), and the slope is determined by the resistance of the Silicon. These two have different dependencies on corner conditions. In other words, the slope doesn't get steeper when the knee voltage becomes smaller. =20 I hope this helps. =20 Arpad =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D =20 From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Ghatawade, Vinayak Sent: Friday, June 10, 2011 3:22 AM To: Lynne D. Green Cc: ibis-users Subject: RE: [IBIS-Users] IV clamp range =20 Lynne,=20 =20 Thanks for your comments. We are seeing gnd and powerclamp showing the crossovers.=20 Please see attached.=20 =20 We do not see crossover in the normal range 0 to VCC since the ESD diodes are not on.=20 We use same voltage range in the ibis model. But my concern is since the spice attached is showing the crossover. Is there anything that I need to do different to report in the model.=20 =20 In the spice we do below sweep ranges. While reporting we report back to - -1.8V to 3.6V since the total range (5.4V is fixed for all corners irrespective of the abs voltage sweep). =20 Thanks=20 vinayak =20 ________________________________ From: Lynne D. Green [mailto:lgreen22@mindspring.com]=20 Sent: Friday, June 10, 2011 1:19 PM To: Ghatawade, Vinayak Cc: ibis-users Subject: Re: [IBIS-Users] IV clamp range =20 Hello, Vinayak, You do not say which table exhibits a Crossing, or at what voltage. - - A clamp table can have a crossing for V<0 when the "clamping" is bipolar, while the drive is CMOS. Physically, bipolar and CMOS exhibit opposite dI/dTemp at fixed V. In this case, the total current may also have a crossing. - - A clamp table can exhibit a crossing due to numerical resolution when separating the clamp and drive currents into their respective tables. The total (summed) current usually does not show this type of crossing. - - It is rare to see I-V crossings in the range 0 to Vcc. It is strongly recommended for all I-V tables to use the same voltage range (-Vcc to 2Vcc). In your example, that would be -1.8V to +3.6V. As a model maker, you could submit a model to the IBIS Model Review Committee. http://www.eda.org/ibis/home/support/support.htm Cheers, Lynne Dr. Lynne Green Chair, IBIS Model Review Committee On 6/10/2011 12:00 AM, Ghatawade, Vinayak wrote:=20 Hello, =20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20 I have question reagarding the IV sweep ranges used for IBIS IV table generation.=20 Cookbook does not have detailed sweep range for all corners specified explicitly. The reason that I am asking this question is we have IV table which has cross-overs for nom, strong and weak corners. In spice I see this behavior which is expected as per diode behavior since the corners are temp variable as well.=20 Weak will turn on (125c) first and then followed by nom (25) and strong (-40c). This results in the crossover.=20 Some customers are not willing to have these crossover (not sure why?). But I feel since this is real behavior you should see the same=20 reflected in the ibis model. =20 Do you see a reason why IV crossovers are not allowed (recommended). I did not see any checklist or parser giving out any warning /error due this this. =20 Going back to the sweep voltage range when I see this crossover in spice (I sweep voltages as mentioned in cookbook) and report the strong, weak to fit in the=20 typical range voltage axis in the ibis model. =20 Ex 1.8V IO below is the numbers that I use for IV sweep range. =20 Sweep range that I use for typ=20 - -1.8 to 3.6=20 =20 Min/weak will be=20 - -1.962 to 3.428=20 =20 Max/strong will be=20 - -1.602 to 3.798 =20 =20 Thanks=20 vinayak =20 - --=20 This message has been scanned for viruses and=20 dangerous content by MailScanner , and is believed to be clean.=20 =20 - --=20 This message has been scanned for viruses and=20 dangerous content by MailScanner , and is believed to be clean.=20 - --=20 This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. - ------_=_NextPart_001_01CC2773.8647A320 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

= Vinyak,

 

The crossing you show is= perfectly normal.  The knee

voltage is dete= rmined by the P-N junction (parasitic

diode), and= the slope is determined by the resistance

of the= Silicon.  These two have different dependencies

=

on corner conditions.  In other words, the slope doesn’t

get steeper when the knee voltage becomes smaller.<= o:p>

 

= I hope this help= s.

 

Arpad

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

&= nbsp;

From:<= /span> owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.o= rg] On Behalf Of Ghatawade, Vinayak
Sent: Friday, June 10,= 2011 3:22 AM
To: Lynne D. Green
Cc: ibis-users
S= ubject: RE: [IBIS-Users] IV clamp range

 

Lynne, <= o:p>

 

Thanks for your comments.

We are seeing gnd and powerclamp showing the crossovers.

Please see attached. <= /span>

 

We do not see crossover in the normal range 0 to VCC since the ESD= diodes are not on.

We use sa= me voltage range in the ibis model. But my concern is since the spice attac= hed is showing the crossover.

I= s there anything that I need to do different to report in the model. <= /o:p>

 

In the spice we do below sweep ranges. While reporting we re= port back to -1.8V to 3.6V since the total range (5.4V is fixed for all cor= ners irrespective of the abs voltage sweep).

 

Thanks

vinayak

 

<= div class=3DMsoNormal align=3Dcenter style=3D'text-align:center'>

From:= Lynne D. Green [mailto:lgreen22@mindspring.com]
Sent: Friday, June 10, 2011 1:19 PM
To: Ghatawade, VinayakCc: ibis-users
Subject: Re: [IBIS-Users] IV clamp range<= /span>

 

Hello, Vinayak,

You do not say whi= ch table exhibits a Crossing, or at what voltage.
- A clamp table can ha= ve a crossing for V<0 when the "clamping" is bipolar, while th= e drive is CMOS.  Physically, bipolar and CMOS exhibit opposite dI/dTe= mp at fixed V.  In this case, the total current may also have a crossi= ng.
- A clamp table can exhibit a crossing due to numerical resolution w= hen separating the clamp and drive currents into their respective tables.&n= bsp; The total (summed) current usually does not show this type of crossing= .
- It is rare to see I-V crossings in the range 0 to Vcc.

It is = strongly recommended for all I-V tables to use the same voltage range (-Vcc= to 2Vcc).  In your example, that would be -1.8V to +3.6V.

As a= model maker, you could submit a model to the IBIS Model Review Committee.<= br>http://www.= eda.org/ibis/home/support/support.htm

Cheers,
Lynne

Dr= . Lynne Green
Chair, IBIS Model Review Committee



On 6/10/= 2011 12:00 AM, Ghatawade, Vinayak wrote:

Hello,

       = ;             <= /span>

I ha= ve question reagarding the IV sweep ranges used for IBIS IV table generatio= n.

= Cookbook does not have detailed sweep range for all corners specified expli= citly.

The reason that I am asking this question is we have IV table which has = cross-overs for nom, strong and weak corners.

In spice I see this behavior whic= h is expected as per diode behavior since the corners are temp variable as = well.

Weak will turn on (125c) first and then followed by nom (25) and strong = (-40c). This results in the crossover.

Some customers are not willing to have = these crossover (not sure why?). But I feel since this is real behavior you= should see the same

reflected in the ibis model.

 

Do you see a reason why IV cr= ossovers are not allowed (recommended). I did not see any checklist or pars= er giving out any warning /error due this this.

 

Going back to the sweep volt= age range when I see this crossover in spice (I sweep voltages as mentioned= in cookbook) and report the strong, weak to fit in the <= /p>

typical range voltage = axis in the ibis model.

 

Ex 1.8V IO below is the numbers that I use for IV s= weep range.

 

Sweep range that I use for typ

-1.8 to 3.6

 

Min/weak will be =

-1.962 to= 3.428

 

Max/strong will be

-1.602 to 3.798

 

 

Thanks

vinayak

=

 


--
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
believ= ed to be clean.


--=20
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean. - ------_=_NextPart_001_01CC2773.8647A320-- - -------------------------------------------------------------------- |For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org |with the appropriate command message(s) in the body: | | help | subscribe ibis | subscribe ibis-users | unsubscribe ibis | unsubscribe ibis-users | |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 1993 ------------------------------ Date: Tue, 10 May 2011 10:50:30 +0530 From: Muniswara Reddy Vorugu Subject: [IBIS-Users] IBIS Model Usage: - --_000_A7CAB96F02008C4AB1BED5606E2A7DFE04F8581EF8SAROVARAAsiap_ Content-Type: text/plain; charset=WINDOWS-1252 Content-Transfer-Encoding: quoted-printable Hi, For better delivery of IBIS model, we want to know better ways to get usab= lility and manage our deliverables. To make this balance, I want to understand the following: 1: What are the advantages of min/typ/max corners merged in single file, o= ver 3 files with one corner per file? What is the difficulty for running Min/Typ/Max corners analysis. Are there any Fundamental analysis limitations? Regards, Muniswara - -- IMPORTANT NOTICE: The contents of this email and any attachments are con= fidential and may also be privileged. If you are not the intended recipient= , please notify the sender immediately and do not disclose the contents to = any other person, use it for any purpose, or store or copy the information = in any medium. Thank you. - --=20 This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. - --_000_A7CAB96F02008C4AB1BED5606E2A7DFE04F8581EF8SAROVARAAsiap_ Content-Type: text/html; charset=WINDOWS-1252 Content-Transfer-Encoding: quoted-printable

Hi,

For better  delivery= of IBIS model, we want to know better ways to get usablility and manage ou= r deliverables.

To make this balance, I w= ant to understand the following:

 <= /p>

1:  What are the adv= antages of min/typ/max corners merged in single file, over 3 files with one= corner per file?

What is the difficulty for running Min/Typ/Max corners analysis.

Are there any Fundamental analysis limitations?

 <= /p>

 <= /p>

Regards,

Muniswara

 <= /p>

 <= /p>


-- IMPORTANT NOTICE: The co= ntents of this email and any attachments are confidential and may also be p= rivileged. If you are not the intended recipient, please notify the sender = immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the informat= ion in any medium. Thank you.
--=20
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean. - --_000_A7CAB96F02008C4AB1BED5606E2A7DFE04F8581EF8SAROVARAAsiap_-- - -------------------------------------------------------------------- |For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org |with the appropriate command message(s) in the body: | | help | subscribe ibis | subscribe ibis-users | unsubscribe ibis | unsubscribe ibis-users | |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 1993 ------------------------------ Date: Thu, 21 Apr 2011 00:11:44 +0800 From: ji xiao Subject: [IBIS-Users] pulse width of IBIS model response - --0016e648d1f88384c504a15be193 Content-Type: text/plain; charset=ISO-8859-1 Dear All, Thanks for your replies for my previous question! This time I got a new question about the pulse width of IBIS models. Since IBIS model does not include the propagation delay of the driver, it always has shorter time delay than SPICE model. And if we give a pulse input, at the rising edge, IBIS did rise faster than SPICE, but at the falling edge, it nearly falls *at the same time* with SPICE model, i.e., the total pulse width of IBIS response is wider than the SPICE response. It is easy to understand the faster response at the first switching edge (rise edge), but why it is not the same at the second switching edge (fall edge)? Is it also related to the threshold voltage defined by different simulation tool? Thanks and regards, Yuancheng - -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. - --0016e648d1f88384c504a15be193 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Dear All,
Thanks for your replies for my previous question! This time I = got a new question about the pulse width of IBIS models.
Since IBIS mode= l does not include the propagation delay of the driver, it always has short= er time delay than SPICE model. And if we give a pulse input, at the rising= edge, IBIS did rise faster than SPICE, but=A0 at the falling edge, it near= ly falls at the same time with SPICE model, i.e., the total pulse wi= dth of IBIS response is wider than the SPICE response.
It is easy to understand the faster response at the first switching edge (r= ise edge), but why it is not the same at the second switching edge (fall ed= ge)? Is it also related to the threshold voltage defined by different simul= ation tool?
Thanks and regards,

Yuancheng

--=20
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean. - --0016e648d1f88384c504a15be193-- - -------------------------------------------------------------------- |For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org |with the appropriate command message(s) in the body: | | help | subscribe ibis | subscribe ibis-users | unsubscribe ibis | unsubscribe ibis-users | |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 1993 ------------------------------ Date: Wed, 27 Apr 2011 08:29:14 -0400 From: "Timothy Coyle" Subject: RE: [IBIS-Users] Multi level IBIS model development (Ethernet) This is a multi-part message in MIME format. - ------=_NextPart_000_0016_01CC04B5.32399D10 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi Vishrant, Usually when developing an IBIS model that has anything more than standard output design you will have to look into some of the advanced IBIS keywords and/or modeling methods. Without fully reviewing the Ethernet 100 BT mode here are some of your options to model this in IBIS: . The [Driver Schedule] keyword can be used to model multi-stage drivers. You model different buffer stages separately and use timing parameters to dictate final output. Disadvantage here is that there are limitations on the frequencies you can run the model at depending on how you setup the timing parameters. . Everything else really falls into what I call Macro Modeling. You could try to use the external keyword support for Verilog/VHDL-AMS modeling but the end model will not be easily supported across all EDA tools. Main issue here is a lot of the PCB Signal Integrity tools don't have native support for *AMS models. There are probably other ways to try and model your design but those are the ones that I can think of off the top of my head. Best, Tim From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Vishrant GOYAL Sent: Wednesday, April 27, 2011 6:42 AM To: ibis-users@eda.org; ibis@eda.org Subject: [IBIS-Users] Multi level IBIS model development (Ethernet) Hello Expert, I am trying to develop ibis model for Ethernet Phy for 100 BT mode. In 100 BT mode there are 3 level possible at the output (TXP and TXN). Means for every 1's at the input data, output will switch between three levels and for every 0's output will remain on the same level. So for given below input data stream we got following output from the driver (Similar but opposite output on TXN) 0 1 1 0 1 1 1 1 0 0 0 1 0 1 0 1 1 0 (Input DATA) vcm - - How could I develop the ibis for Multi level signals? Can anyone give some light on this model development. Thanks & Regards, Vishrant Goyal - -- 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. - ------=_NextPart_000_0016_01CC04B5.32399D10 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi Vishrant,

Usually when developing an IBIS model that has anyt= hing more than standard output design you will have to look into some of th= e advanced IBIS keywords and/or modeling methods. Without fully reviewing t= he Ethernet 100 BT mode here are some of your options to model this in IBIS= :

<= o:p> 

·      &= nbsp;  Th= e [Driver Schedule] keyword can be used to model multi-stage drivers. You m= odel different buffer stages separately and use timing parameters to dictat= e final output. Disadvantage here is that there are limitations on the freq= uencies you can run the model at depending on how you setup the timing para= meters.

·     &nbs= p;   Everything else really falls into what I call Macro Modeling. You could tr= y to use the external keyword support for Verilog/VHDL-AMS modeling but the= end model will not be easily supported across all EDA tools. Main issue he= re is a lot of the PCB Signal Integrity tools don't have native support for= *AMS models.

 

There are probably other ways to try and model your design b= ut those are the ones that I can think of off the top of my head.

 <= /o:p>

Best,

=  

T= im

=  

 

From: owner-ibis-users@eda.org [mailto:owner-ibis-use= rs@eda.org] On Behalf Of Vishrant GOYAL
Sent: Wednesday, A= pril 27, 2011 6:42 AM
To: ibis-users@eda.org; ibis@eda.org
= Subject: [IBIS-Users] Multi level IBIS model development (Ethernet)

 

Hello Expert,

 

I am trying to develop ibis mode= l for Ethernet Phy for 100 BT mode.

<= o:p> 

In 100 BT mode there are 3 leve= l possible at the output (TXP and TXN).

Means for every 1's at the input data, output will switch between three = levels

and for every 0's output will = remain on the same level.

 =

So for given below input data stream we g= ot following output from the driver (Similar but opposite output on TXN)

0  1 1 0  1 1 1 1 0  0&n= bsp; 0  1 0  1  0  1 1 0 (Input DATA)

=

 

 

 

 =

 <= o:p>

 

 <= /span>

 

 

 

 

 

=

 

 

 

 <= o:p>

 

=

 vcm

 =

&= nbsp;

 

 

 =

 

 

=

&n= bsp;

 

 

 

 <= o:p>

 

=

 

 

 

 

 

 

- How could I= develop the ibis for Multi level signals?

 

Can anyone give some lig= ht on this model development.

&n= bsp;

Thanks & Regards,

<= p class=3DMsoPlainText>Vishrant Goyal


--=
This message has been scanned for viruses and
dangerous content by= MailScanner, and is believed to be clean.


--=20
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean. - ------=_NextPart_000_0016_01CC04B5.32399D10-- - -------------------------------------------------------------------- |For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org |with the appropriate command message(s) in the body: | | help | subscribe ibis | subscribe ibis-users | unsubscribe ibis | unsubscribe ibis-users | |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 1993 ------------------------------ Date: Thu, 2 Dec 2010 06:47:34 -0800 From: "Muranyi, Arpad" Subject: RE: [IBIS-Users] The [External Model] and [External Circuit] keywords were added in IBIS v4.1 and allow the usage of Verilog-A, Verilog-AMS, and VHDL-AMS languages with IBIS files. John is correct, the legacy IBIS specification does not include the capabilities to model feedback effects in the buffers. I don't know exactly what your application is but you might also consider the latest Algorithmic Modeling (AMI) features of IBIS 5.0. Arpad ===================================================================== - -----Original Message----- From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of John Phillippe Sent: Thursday, December 02, 2010 8:35 AM To: Ghatawade, Vinayak Cc: ibis-users Subject: Re: [IBIS-Users] Vinayak, The IBIS standard itself doesn't allow for any kind of feedback modeling of slew rate control. I forget the exact IBIS version, but either 4.0, 4.2 or 5.0 allows for the inclusion of verilog-A models which can be written to model this behavior. Ghatawade, Vinayak wrote: > All, > > > > This has been discussed previously in this forum on modeling Adaptive > feedback buffers. > > Is there anyway we can model adaptive feedback buffer which depends on > the PAD voltage and inturn switches on pmos/nmos depending on external > loading condition. > > > > I tried looking at few of the options like driver schedule but since > these are timing related I cannot use for my case since in my case the > buffer > > Adjust the slews based on PAD voltage and not with respect to the time > (delayed turn on). Also if I want to model the PAD dependency in time > domain this will be very > > Specific to a loading condition. Since IBIS modeling can be done using > R_load only we cannot make sure this approach can work across different > user conditions (tline, loading etc). > > > > Do you have any other suggestion to model a buffer with PAD feedback. > > > > Thanks > > vinayak > > > > > > - ------------------------------------------------------------------------ > > -- > This message has been scanned for viruses and > dangerous content by *MailScanner* , and is > believed to be clean. - -- John Phillippe MSG, I/O Pad Design Freescale - 512-895-1835 Austin, TX - john.phillippe@freescale.com - -- 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 | subscribe ibis-users | unsubscribe ibis | unsubscribe ibis-users | |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 1993 - -- 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 | subscribe ibis-users | unsubscribe ibis | unsubscribe ibis-users | |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 1993 ------------------------------ End of ibis-users V1 #177 *************************