Point to point simulation results

Hi,I am trying to simulate a topology.it consists of driver --> 0 ohm resistor --> MS line (length varied from1000 mils to 10000 mils) ---> DDR2 dq 533 io(receiver).I am just simulating to findout how much can be the maximum length, beforesignal degrades badly.As i am increasing the length, i am seeing more degradation at the Driverthan at the receiver. there is degradation at the receiver , but it is OK.I have uploaded 3 files 1000mils,4800mils and 10000mils@ 10000mils there is delay between input and output but still outputquality and input quality are good compared to 4800 mils case.why is this? 10000mils case should have more degradationhttp://s3.postimg.org/q6wi1e283/ddr2_io_0000_dq_533_MS3_1000.pnghttp://s2.postimg.org/4qw22kazt/ddr2_io_0000_dq_533_MS3_4800.pnghttp://s27.postimg.org/o988g61zn/ddr2_io_0000_dq_533_MS3_10000.png
malli.1729 6 years 1 month 11 days

3 answers


The best answer


You can select the best answer for current question!
Answered bysherman.chen 6 years 22 days
Hi, The 4800mil waveform shows both source and load ends are mismatch. The reason why 1000mil and 10kmil give a less distorted signal is bcs at these lengths the impd seen from the source is close to Zload - you will see it clearly if drawing a Smith chart. -----
Answered bymalli.1729 6 years 1 month 10 days
The receiver IO model i selected is Without ODT.I will try with single Pulse.....On Wed, Oct 22, 2014 at 5:39 PM, preetesh rathod  wrote:> Hello>          You haven't mentioned anything about end termination (ODT) if> you using it. In your case it seems impedance mismatch at both ends. So> your single pulse reflections may not die out in single cycle which can> create such type of results at different trace lengths. I suggest you to> simulate with single edge which will help you to understand what is going> on in more detail in terms of reflections.>> Regards> Preetesh Rathod>>>   On Tuesday, 21 October 2014 7:30 AM, Bert Simonovich <> bertsimonovich@xxxxxxxxxx> wrote:>>> This looks like a case of driver impedance mismatch to the transmission> line. The driver likely has a lower impedance than the transmission line.> Try increasing the 0 ohm resistor to a value equal t-line impedance minus> the driver impedance. Eg. if the driver impedance is 15 ohms, and the> t-line is 50 ohms, then the resistor should be 35 ohms.>> Sent from my iPad>> > On Oct 20, 2014, at 8:41 PM, Mallikarjun K  wrote:> >> > Hi,> > I am trying to simulate a topology.> > it consists of driver --> 0 ohm resistor --> MS line (length varied from> > 1000 mils to 10000 mils) ---> DDR2 dq 533 io(receiver).> >> > I am just simulating to findout how much can be the maximum length,> before> > signal degrades badly.> >> > As i am increasing the length, i am seeing more degradation at the Driver> > than at the receiver. there is degradation at the receiver , but it is> OK.> >> > I have uploaded 3 files 1000mils,4800mils and 10000mils> > @ 10000mils there is delay between input and output but still output> > quality and input quality are good compared to 4800 mils case.> >> > why is this? 10000mils case should have more degradation> >> > http://s3.postimg.org/q6wi1e283/ddr2_io_0000_dq_533_MS3_1000.png> >> > http://s2.postimg.org/4qw22kazt/ddr2_io_0000_dq_533_MS3_4800.png> >> > http://s27.postimg.org/o988g61zn/ddr2_io_0000_dq_533_MS3_10000.png> >> >> > 
Answered bybertsimonovich 6 years 1 month 12 days
This looks like a case of driver impedance mismatch to the transmission line. The driver likely has a lower impedance than the transmission line. Try increasing the 0 ohm resistor to a value equal t-line impedance minus the driver impedance. Eg. if the driver impedance is 15 ohms, and the t-line is 50 ohms, then the resistor should be 35 ohms. Sent from my iPad> On Oct 20, 2014, at 8:41 PM, Mallikarjun K  wrote:> > Hi,> I am trying to simulate a topology.> it consists of driver --> 0 ohm resistor --> MS line (length varied from> 1000 mils to 10000 mils) ---> DDR2 dq 533 io(receiver).> > I am just simulating to findout how much can be the maximum length, before> signal degrades badly.> > As i am increasing the length, i am seeing more degradation at the Driver> than at the receiver. there is degradation at the receiver , but it is OK.> > I have uploaded 3 files 1000mils,4800mils and 10000mils> @ 10000mils there is delay between input and output but still output> quality and input quality are good compared to 4800 mils case.> > why is this? 10000mils case should have more degradation> > http://s3.postimg.org/q6wi1e283/ddr2_io_0000_dq_533_MS3_1000.png> > http://s2.postimg.org/4qw22kazt/ddr2_io_0000_dq_533_MS3_4800.png> > http://s27.postimg.org/o988g61zn/ddr2_io_0000_dq_533_MS3_10000.png> > >