This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TRF372017: Baseband drive circuit

Part Number: TRF372017
Other Parts Discussed in Thread: DAC3283, THS4509, ADS5424, OPA355, OPA814, OPA817, THS4551, THS4541, THS4503

Hi

we design a PCB with TRF372017 , a waveform generator IC ( AD9833) will be used to generate BB waveform, a 5th order Chepyshev LPF shall be used after AD9833 as a reconstruction filter to eliminate harmonics, then Unipolar to bipolar converters, and finally using high speed differential amplifier to convert single-Ended to Diff, and provide a DC shift to 1.7V as a Vcom to be suitable for TRF372017 with( Vp-p)  almost 1V as shown in next Fig.

so after all, we have a differential signal with Bias 1.7 V and almost 0.3 Vrms as in datasheet and the reconstruction filter already applied at the beginning.

i can't find an example for circuit to connect op-amp differential output with input baseband pins of TRF372017. 

According to TRF372017 datasheet figure 86 "DAC to TRF372017 Interface With External VCM Generation".  I have these questions:

   1-could I use this configuration to connect op-amp differential outputs to TRF372017 BB directly or need modifications ?

   2- is this configuration suitable for Wideband applications, can explain this configuration

   3- how can make output impedance of differential op amp match with input impedance of IQ modulator (5K) ohm?

   4- can i use following circuit to drive BB of IQ modulator? (like this technical article https://www.analog.com/en/technical-articles/baseband-design-example-for-ltc5589-ltc5599-low-power-iq-modulator.html

I tried hard, i have read. TRF372017, DAC3283, SLUA647A but all discuss another case

Appreciate your support.

regards

Eng. maged

  

  • Hi Maged,

    Thanks for your question, we have received and will get back to you. Due to the holiday season, please allow for some delay in our responses.

    Thanks,

    Aman

  • Hello,

    a PCB with TRF372017 , a waveform generator IC ( AD9833) will be used to generate BB waveform, a 5th order Chepyshev LPF shall be used after AD9833 as a reconstruction filter to eliminate harmonics, then Unipolar to bipolar converters, and finally using high speed differential amplifier to convert single-Ended to Diff, and provide a DC shift to 1.7V as a Vcom to be suitable for TRF372017 with( Vp-p)  almost 1V as shown in next Fig.

    Please help provide a block diagram.

    Please also provide the schematic for the high speed differential amplifier and the TRF372017 connection.

     1-could I use this configuration to connect op-amp differential outputs to TRF372017 BB directly or need modifications ?

    The diagram below does not apply to the high speed differential amplifier. The resistor ladder network is based on the concept of the DAC3283 having current sink structure as oppose to the typical high impedance output of the high speed amplifier. You can study the concept of the resistor ladder network design in the following app note to understand the differences

    https://www.ti.com/lit/an/slaa399/slaa399.pdf

    2- is this configuration suitable for Wideband applications, can explain this configuration

    Please advise your connection currently (schematic) of the high speed amp and TRF372017. 

     3- how can make output impedance of differential op amp match with input impedance of IQ modulator (5K) ohm?

    Please advise the differential amplifier that you are using. TI has differential amplifier group who may help out on this. Typically, the output impedance of the differential amplifier should be as low as possible by design. For example, the output impedance of the THS4509 is the following:

    An example circuit would be the following. you can replace the ADS5424 input with the I or Q input of the TRF372017.You can replace 2.7pF with 100ohm 

    the 2x 50ohm can create source impedance of 100ohm differential, with the other 100ohm as load impedance for matching.

    The THS4509 should have sufficient drive strength to drive 200ohm load total:

       4- can i use following circuit to drive BB of IQ modulator? (like this technical article https://www.analog.com/en/technical-articles/baseband-design-example-for-ltc5589-ltc5599-low-power-iq-modulator.html

    The output impedance of the 2.5k is severely high. The app note did not talk about this. You will have difficulty designing the low pass filter between the amplifier and the TRF372017. 

    I would recommend that you first select the amplifier with the right bandwidth and gain for the application that you are building. We can then refer you to the right amplifier support team to support you. On the TRF372017 side, again, we can add 100ohm differential resistor as load termination for your filter load.

  • Hello, Kang Hsia 

    Thanks for your reply, a lot of important suggestions, let me check it, I'll post the block diagram and schematic. 

    regards     

    Eng. Maged

  • Hello, Kang Hsia

    Sorry for being late, I 've been revised your reply and study some technical data sheets.

    First, this is my suggestion block diagram to interface waveform generator (AD9833) with IQ modulator (TRF372017) 

    Second, the schematic shall be something like this , first AD9833 then a unipolar to bipolar converter, then LPF

    then ,  Single-Ended to Diff. finally IQ modulator

    So

    regarding to  unipolar to bipolar converter,

    I imitate both ti technical documents "SBAA603 and SLAA869   Analog Engineer's Circuit

    this part will interface between AD9833 with internal resistor 200 Ohm and LPF with both input and output impedance are 50 ohms

    and simulation results for both input and output 

    is something wrong with this circuit? or do you have some notes about it?

    regarding to the second part LPF then Single-Ended to Diff. I used THS4509, like your previous suggestion and the circuit such that 

    AND SIMULATION WILL BE 

    Also , is something wrong with circuit? do you have notes about it ?

    Finally the 100 ohm resistor between BB inputs for TRF372017 as shown in Fig up is this represent RL ?

    I'm really sorry for this long post, 

    thanks for your interest and help

    regards

    Eng maged

    Also, 

  • Hello,

    Please advise your highest frequency and bandwidth out of the amplifier circuit to the TRF372017. I will need to refer you to our amplifier experts to comment, and need the information to reference you to the associated forum group.

    -Kang

  • Hello

    The frequency for TRF372017 will not exceed than 4.2GHz and bandwidth less than 3MHz   

    Eng Maged

  • Please advise the maximum frequency out of the amplifier circuit (i.e. baseband input of the TRF372017, not output). 

  • Hello

    I got it, the maximum baseband is 3MHz

    Regards

  • Hi Maged, 

    Could you clarify which part of the circuit you are inquiring about? The differential amplifier circuit or the buffer amplifier circuit? 

    Regards,
    Ashley

  • Hello Ashely

    actually both, but if I have to choose. The differential amplifier circuit and interfacing it with TRF372017 is more important for me since for buffer amplifier I've referenced with "SBAA603 and SLAA869" .  

    Regards,

    Maged

  • Hi Maged, 

    In that case, I will move this thread over to this High Speed Amplifier team to further support the differential amplifier circuit you have referenced above as it is using the THS4509. 

    They can help to support this further with this device application. 

    regarding to the second part LPF then Single-Ended to Diff. I used THS4509, like your previous suggestion and the circuit such that 

    AND SIMULATION WILL BE 

    Also , is something wrong with circuit? do you have notes about it ?

    Finally the 100 ohm resistor between BB inputs for TRF372017 as shown in Fig up is this represent RL ?

    Thank you!
    Best Regards,
    Ashley

  • Hello Maged,

    Our HSAMPs team will look over the thread and provide any additional insight we can offer.

    We may have more follow-up questions, thank you for your patience.

    Best,

    Alec

  • Hello Alec

    waiting to hear from your team.

    Regards

    Maged  

  • Hello Maged,

      Thank you for your patience. We reviewed your design and requirements, and have a few comments and suggestions. 

    1. Looked into the waveform generator, and it has a voltage output between 38mVpp to 0.65Vpp with a max frequency of 12.5MHz with either a sine, triangular, or square wave output. I am unsure what the DC offset is for the device, but you have your design set to 1.7V. I will assume you will AC couple from output of the generator into the buffer/amplifier prior to your 5th order LPF. 

    2. Your chosen OPA355 is a good choice, but I also recommend looking at these other amplifiers depending on your other specifications: link comparing automotive/catalog qualified amplifiers. In your OPA355 circuit, you have VS+ set to 3.3V and VS- set to 3.2V. The total of 6.5V is below the abs max voltage (7.5V) of the part, but we do recommend operating the device between 2.7V and 5.5V. If you need higher operating power, I suggest looking at the link between OPA814 and OPA817. 

    3. Will you be using the full range of the waveform generator from 38mVpp to 0.65Vpp? If it is the entire range, I suggest this system layout instead: ADS98333 Waveform Generator --> OPA355 at a gain of 2V/V --> 5th order LPF --> Variable gain Fully Differential Amplifier --> TRF372017. The variable gain fully differential amplifier is a catalog of devices we provide which are fully differential amplifiers with gains that can be either altered digitally or via analog voltage. Since you would need single-ended to differential conversion, I would recommend either the LMH6518 or the LMH6881. If you have a specific smaller voltage range, I would then recommend choosing our regular FDAs, as you have chosen the THS4509. I am unfamiliar with TRF372017, but THS4509 could be overkill for a bandwidth of 3.5MHz. I would suggest looking at the THS4551 and THS4541 if you want to use 3.3V power supplies. Otherwise, we also have many options for 5V supply range which can be found at this link.

    4. Simulation of OPA355:

      I can recommend a different circuit if your voltage supplies are flexible. Can you work with these supplies: +/-2.5V and you would have to consider your waveform generator's required single 2.3 V to 5.5 V power supply. Or, we can work with single +5V supply for your entire design including your waveform generator. 

    5. THS4509 voltage supply is from 3V to 5V recommended, or 6V max. I believe you are violating the supplies which is the reason for the simulation result. Depending on the answers above, we can redesign with THS4509 or we can help with designing with the variable fully differential amplifiers 

    Thank you,

    Sima

  • Hello Sima,

    first sorry for being late,

       Thank you for your helpful notes, i'll give my comment in red one by one 

    1- "Looked into the waveform generator, and it has a voltage output between 38mVpp to 0.65Vpp with a max frequency of 12.5MHz with either a sine, triangular, or square wave output………."

          Exactly I’m planning to use AC couple between waveform stage and the next

    2- "Your chosen OPA355 is a good choice, but I also recommend looking at these other amplifiers depending on your other specifications: link comparing automotive/catalog qualified amplifiers. ……."

     

             Actually I figured out this point, so I’ planning to use  3.3V as a VS+ and – 1.8V as a VS-   for  OPA 355 since it will be suitable for the required output             voltage swing (1.7+0.6 V to 1.7-0.6V)

     

    3- "If you have a specific smaller voltage range, I would then recommend choosing our regular FDAs, as you have chosen the THS4509. I am unfamiliar with TRF372017,.........." 

          Actually according to datasheet, TRF372017 accept BB input around 0.3V rms with CMV equal to 1.7V, so the output voltage swing (2.3~ 1.1) V is                suitable for IQ baseband  signal

     

    4- “ but THS4509 could be overkill for a bandwidth of 3.5MHz I would suggest looking at the THS4551 and THS4541 if you want to.......... 

              I couldn’t get this point as per THS4509 datasheet,  BW up to 1900 MHz, or I miss something here ?

    5- "THS4509 voltage supply is from 3V to 5V recommended, or 6V max. I believe you are violating the supplies which is ..........."

           

            I  totally agree with you, so after I submit  previous post I already replaced THS4509 with THS4503 using supply voltage Vs+=5V, VS-=gnd and BW              up to 370 MHz

        

    So finally after these two changes

    • Supply voltage for OPA355, VS+ = 3.3V and VS-= 8V
    • Using THS4503 with power supply VS+ = 5V and VS-= gnd

      and data, BB voltage signal for TRF372017 equal to 0.3Vrms, BW for THS4503 equal to 370

    Do I still need other modifications?

    6-" I can recommend a different circuit if your voltage supplies are flexible. Can you work with these supplies: +/-2.5V and you would have to consider your waveform generator's required single 2.3 V to 5.5 V power supply. Or, we can work with single +5V supply for your entire design including your waveform generator"

       

       Thanks for your proposal, it will be helpful to consider new ideas from your side, i have +5, +3.3,±1.8

    thank you for your patience

    Regards

    Maged

  • Hello Maged,

      Thank you for your additional comments! I will base the proposed design on your available supply voltages and the TRF372017 input requirements. I should be able to get back to you by end of the week.

    Best Regards,

    Sima