Hi, Tom, It all depends on where the pullup resistor is located. I assumed that the pullup was on-die with the buffer (esp with the high value, since some I/Os are designed that way). You are right that there are other cases I should have considered, particularly the case of the pullup being at the load. I think we both qualify as good engineers with an occasional unstated assumption. - Lynne _____ From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Tom Dagostino Sent: Friday, October 26, 2007 10:23 AM To: 'Lynne D. Green'; 'Ibis Model'; 'Server EDA' Subject: RE: [IBIS-Users] re: Warning message for IO_open_drain model Lynne I assumed (bad engineer, bad engineer) that the 1M resistor came from the datasheet and not a made up number. I too think it is a very high value but I had no basis to question it from the information in the question. I've seen a lot of open sink drivers in all the models we have done and have found many way too weak to drive a 50 Ohm pullup to any reasonable Vmeas. Open sink driver applications seem to be split between three different categories; 1. very low power status bits, 2. LED drivers and other applications where speed is not an issue and only require a mA or so and 3. High speed wired OR applications. A 1M pullup would fall into the first category but it still seems to be on the high side. But if your application is battery powered it may make sense. I'm not sure you have your transmission line theory correct. Let's take a single 1V, 1psec edge propagating down a 6 inch transmission line terminated into 50 Ohms to ground. Let's assume the output impedance of the driver is 50 Ohms. Let's also assume the Zo of the transmission line is 50 Ohms and the speed of propagation is 170 psec per inch. When the waveform is launched into the transmission line the only impedance seen by the driver is the Zo of the line. The voltage at the output of the driver is now 0.5V The signal has not been able to propagate down to the length of the line yet. As time goes by (6*170psec) the signal eventually reaches the end of the line and is totally absorbed by the termination, there is no reflection. So the driver sees a 50 Ohm load from the time it launches the edge down the line until the power is removed. The output voltage of the driver will remain at 0.5V If we were to replace the 50 Ohm termination with a short the driver again would see a 50 Ohm transmission line when it launches the edge. Again the output voltage of the driver will be 0.5V. After 6*170psec the edge will get to the short and a large reflection will occur. This reflection will take 6*170psec to get back to the driver. Up until the reflection returns to the driver the impedance seen by the driver is 50 Ohms. So for 12*170psec the driver sees 50 Ohms. After that time the reflection returns to the driver and the driver now "knows" there is a short at end of the line. The output voltage from the driver is now 0V. And you can see the logic with a open at the end of the line. The drive sees 50 Ohms for 12*170psec and then sees an open until the power is removed. After the reflection returns from the open the output voltage is 1V. Regards Tom Dagostino Teraspeed(R) Labs 13610 SW Harness Lane Beaverton, OR 97008 503-430-1065 tom@teraspeed.com www.teraspeed.com Teraspeed Consulting Group LLC 121 North River Drive Narragansett, RI 02882 401-284-1827 From: owner-ibis-users@server.eda.org [mailto:owner-ibis-users@server.eda.org] On Behalf Of Lynne D. Green Sent: Friday, October 26, 2007 9:46 AM To: tom@teraspeed.com; 'Ibis Model'; 'Server EDA' Subject: RE: [IBIS-Users] re: Warning message for IO_open_drain model Hi, Tom, Thanks for the catch. I must have confused Rref with a resistor-loaded table. Another thing what caught my eye was the value of 1e6. Wouldn't this be effectively be an "open", as it is much larger than any realistic Z0? (I seem to remember values like that preventing a Vmeas crossing.) For Rref: When the buffer is driving, one sees a Thevenin equivalent of the pullup (to Vcc) in parallel with the 50-ohm transmission line (to Vcc or GND) ["parallel" when both go to Vcc]. Many people take the rule of thumb to use Z0, which works as long at ZUP>>Z0. Best regards, Lynne _____ From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Tom Dagostino Sent: Thursday, October 25, 2007 10:43 AM To: 'Lynne D. Green'; 'Ibis Model'; 'Server EDA' Subject: RE: [IBIS-Users] re: Warning message for IO_open_drain model Lynne The problem is not the load resistance but the voltage it is terminated to, Vref = 0 in this case. Lowering the Rref is not going to help, Rref should be whatever resistance the datasheet says it should be. There will be no current flowing in this driver if both the load and the driver are tied to ground. From I=E/R, I will be 0 if E=0. I've seen this mistake way too many times and likely made it myself once or twice. I'm not sure what you mean by (Rref of 50 ohms for the transmission line load, in parallel with the pullup load.) Could you expand on it? What do you mean in parallel with the pullup load? Tom Dagostino Teraspeed(R) Labs 13610 SW Harness Lane Beaverton, OR 97008 503-430-1065 tom@teraspeed.com www.teraspeed.com Teraspeed Consulting Group LLC 121 North River Drive Narragansett, RI 02882 401-284-1827 From: owner-ibis-users@server.eda.org [mailto:owner-ibis-users@server.eda.org] On Behalf Of Lynne D. Green Sent: Thursday, October 25, 2007 10:02 AM To: 'Ibis Model'; 'Server EDA' Subject: RE: [IBIS-Users] re: Warning message for IO_open_drain model Hello, Timmy, The warning says if you hook the circuit up (in simulation), it will only switch a few millivolts, not enough for the output to cross Vmeas. You might try setting Rref and Vref to the normal load for this buffer. (Rref of 50 ohms for the transmission line load, in parallel with the pullup load.) One could also look at the V-t curves, to see where their endpoint voltages are. If you would like your model reviewed, the IBIS Model Review Committee welcomes models that pass the IBIS parser. Best regards, Lynne "IBIS training when you need it, where you need it." Dr. Lynne Green Green Streak Programs http://www.greenstreakprograms.com 425-788-0412 lgreen22@mindspring.com _____ From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Ibis Model Sent: Wednesday, October 24, 2007 11:04 PM To: Server EDA Subject: [IBIS-Users] re: Warning message for IO_open_drain model Hi everyone, I get a warning message "TYP VI curves cannot drive through Vmeas given load Rref=1e+006 Ohms to Vref=0V" for IO_open_drain model. What does this message mean ? Regards, Timmy __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- This message has been scanned for viruses and dangerous content by <http://www.mailscanner.info/> MailScanner, and is believed to be clean. -- This message has been scanned for viruses and dangerous content by MailScanner <http://www.mailscanner.info/> , and is believed to be clean. -- This message has been scanned for viruses and dangerous content by MailScanner <http://www.mailscanner.info/> , and is believed to be clean. -- This message has been scanned for viruses and dangerous content by <http://www.mailscanner.info/> MailScanner, and is believed to be clean. -- This message has been scanned for viruses and dangerous content by <http://www.mailscanner.info/> 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 Mon Oct 29 14:12:47 2007
This archive was generated by hypermail 2.1.8 : Mon Oct 29 2007 - 14:13:42 PDT