AD9680 input impedance matching questions

good day

I have several questions regarding the input interface of the AD9680.

I would like to design a matching network using Snp file. Firstly I don't understand why S2p are provided whereas I expect only file having S11 parameter since the purpose is to provide matching impedance. In addition several files are provided, I don't know which one to use. Could you please clarify which context for each file?

In the datasheet page 36 at table 9 parameter values are provided for the components shown in figure 115, I was wondering if these values are still ok whatever differential input impedance value is set (see figure 116 where different resistors can be selected via registers). Is it possible to attach a drawing of an example of matching specifying the context ( frequency range...)

Thank you in advance for your reply

Best regards

Raphael

  • 0
    •  Analog Employees 
    on Oct 6, 2016 1:12 AM

    Hello Raphael

    Thank you for contacting engineer zone regarding the AD9680. since the AD9680 inputs are differential, we made 2-port measurements of the analog inputs with port 1 connected to VIN+ and port 2 connected to VIN-. 

    So for a 200ohm termination setup, the simulation setup will look like the one shown below

    Upon simulation, the results will look like the image shown below:

    If you want to create a matching network you can either use a fully differential approach, or use a single ended approach and then generate a differential network. 

    Hope this helps. Please let me know if you have any questions.

    Thanks

    Umesh

  • Hello UmeshThank you very much for your reply. I understand better now. However I have a question regarding the smith chart obtained. The M1 marker lies on the circle having unitary real part impedance. However the 

    marker indicates 399 ohms instead of 200ohms. Could you please tell me why?

    Thank you

    Best regardsRaphael

          De : UmeshJ <analog@analog-vm.hosted.jivesoftware.com>

    À : bendel raphael <raphael_bendel2003@yahoo.fr>

    Envoyé le : Mercredi 5 octobre 2016 23h13

    Objet : Re:  - Re: AD9680 input impedance matching questions

        

    #yiv5126335924 * #yiv5126335924 a #yiv5126335924 body {font-family:Helvetica, Arial, sans-serif;}#yiv5126335924 #yiv5126335924 h1, #yiv5126335924 h2, #yiv5126335924 h3, #yiv5126335924 h4, #yiv5126335924 h5, #yiv5126335924 h6, #yiv5126335924 p, #yiv5126335924 hr {}#yiv5126335924 .yiv5126335924button td {}

    |

      EngineerZone 

    |

    Re: AD9680 input impedance matching questions

    reply from UmeshJ in High-Speed ADCs - View the full discussionHello RaphaelThank you for contacting engineer zone regarding the AD9680. since the AD9680 inputs are differential, we made 2-port measurements of the analog inputs with port 1 connected to VIN+ and port 2 connected to VIN-.  So for a 200ohm termination setup, the simulation setup will look like the one shown below Upon simulation, the results will look like the image shown below: If you want to create a matching network you can either use a fully differential approach, or use a single ended approach and then generate a differential network.  Hope this helps. Please let me know if you have any questions.ThanksUmesh

    Reply to this message by replying to this email, or go to the message on EngineerZone

    Start a new discussion in High-Speed ADCs by email or at EngineerZone

    Following Re: AD9680 input impedance matching questions in these streams: Inbox

      This email was sent by EngineerZone because you are a registered user.

    You may unsubscribe instantly from EngineerZone, or adjust email frequency in your email preferences   |

      |

  • 0
    •  Analog Employees 
    on Oct 6, 2016 10:41 PM

    Hi Rafael,

    Thank you very much for pointing this out to me. In my simulation results I had wrongly used the denormalizing impedance value as  400Ohms instead of 200Ohms. See below for the corrected waveform.

    You can see that  impedance reported is more in line with the model. I should not have done the denormalization and plotted as below:

    Hope this helps answer your questions.

    Thanks

    Umesh

  • Hi Umeshthank you for the clarification.

    Can you please tell me if the values for the components provided in table 9 at page 36 permit to match  at 50, 200,400 ohms ?Could you please tell me which balun do you use in your simulation (a 1:1 or 1:sqrt(2) balun)? I have tried to reproduce the same setup as you but it doesn't work with the 400ohms S2pfile (see attached pictures)? Normally it should be a 1:sqrt(2) balun isn't it? I don't have the right results with this one (see S33 parameter value at 100MHz)

    Thank you very much

    Raphael

          De : UmeshJ <analog@analog-vm.hosted.jivesoftware.com>

    À : bendel raphael <raphael_bendel2003@yahoo.fr>

    Envoyé le : Jeudi 6 octobre 2016 20h43

    Objet : Re:  - Re: AD9680 input impedance matching questions

        

    #yiv4916292152 * #yiv4916292152 a #yiv4916292152 body {font-family:Helvetica, Arial, sans-serif;}#yiv4916292152 #yiv4916292152 h1, #yiv4916292152 h2, #yiv4916292152 h3, #yiv4916292152 h4, #yiv4916292152 h5, #yiv4916292152 h6, #yiv4916292152 p, #yiv4916292152 hr {}#yiv4916292152 .yiv4916292152button td {}

    |

      EngineerZone 

    |

    Re: AD9680 input impedance matching questions

    reply from UmeshJ in High-Speed ADCs - View the full discussionHi Rafael,Thank you very much for pointing this out to me. In my simulation results I had wrongly used the denormalizing impedance value as  400Ohms instead of 200Ohms. See below for the corrected waveform.  You can see that  impedance reported is more in line with the model. I should not have done the denormalization and plotted as below:  Hope this helps answer your questions. ThanksUmesh

    Reply to this message by replying to this email, or go to the message on EngineerZone

    Start a new discussion in High-Speed ADCs by email or at EngineerZone

    Following Re: AD9680 input impedance matching questions in these streams: Inbox

      This email was sent by EngineerZone because you are a registered user.

    You may unsubscribe instantly from EngineerZone, or adjust email frequency in your email preferences   |

      |

    attachments.zip
  • 0
    •  Analog Employees 
    on Oct 18, 2016 4:21 PM

    Hello Raphael

    I used the ideal balun component in ADS to simulate my circuit. I could not find an equivalent in QUCS. However, I can get the same response by running single ended s-parameter simulation and then using the following equation

    SDD11 = 0.5*(S[1,1]-S[1,2]-S[2,1]+S[2,2])

    See below for schematic

    See below for results

    See below for the same simulation using ADS.

    Thanks

    Umesh