• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 2506 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

THS4303: Operating with outthe 50 Ohm load to the non-inverting input.

Joseph Fox
Prodigy 80 points
Part Number: THS4303

I am interested in use this op-amp for a Negitive Impedance Converter (NIC). I have removed the terminating R to gnd . Does the non-invert need a R to GND to operate. I do not get the expected results from the THS4303EVM, compared to a simulation In Genesys.

  • 0
    TI__Genius 15975 points

    Hi Joseph,

    Would it be possible for you to attach a schematic of the circuit you are trying to debug? Also, any waveform or results of the current circuit would help.

    Best Regards,

    Rohit

  • 0 in reply to Rohit Bhat
    Prodigy 80 points
    Rohit, I am using the evaluation BRD THS4303EVM. For my final circuit I can not have the 50 Ohm load @ the non-inverting input. I have measurements of S11, S22, and S21 but can not attach them. With the 50 Ohm to GND the op-amp operates as expected. Without the 50 Ohm there is a significant loss in Gain. Pin -20dBm Pout -40dBm.
  • 0 in reply to Joseph Fox
    TI__Genius 15975 points

    Joseph, 

    It is possible that without the 50-ohm resistor to the non-inverting input, you might be distorting the signal from the network analyzer causing it to interpret a lower gain due to the drastic mismatch. At what frequency are you reporting the gain to be lower? The THS4303 non-inverting input impedance is 1.6Mohm||1pF. So, when the network analyzer output sees this impedance, you would be getting total signal reflection as it is ideally expecting 50-ohms and thus, resulting in the gain interpreted wrongly. 

    Would it be possible to describe in more detail as to what application are you evaluating the THS4303EVM for? Maybe, only for evaluation purposes, you could use the 50-ohm resistor to GND and not needed for your end application. It would greatly help to debug further if you could attach circuit description of your end application or screen-shots of S11, S22 and S21.

    Best Regards,

    Rohit 

  • 0 in reply to Rohit Bhat
    Prodigy 80 points

    THS4303EVM.pdfTHS4303_NIC_W_50OHM to GND.pdf

    Rohit,

    See attached plot/schematic. The goal is to use the THS4303 as a negative impedance converter (NIC) utilizing a non-inverting feed back resistor. The goal is to achive a -50Ohm impedance. Adding a 100Ohm resistor in series with the non-inverting input the response in theory should be around 50ohm. The 100Ohm will not be used in the final layout but just used to realize the -50 Ohm impedance.

  • 0 in reply to Joseph Fox
    TI__Genius 15975 points

    Joseph,

    Something you can try quickly is to use the "port-Z conversion" feature on the network analyzer by setting the input port Z0 to 1.6Mohm, and see whether the S21 response is corrected without the 50-ohm resistor to GND. Also, what supply voltage are you operating the THS4303 on? Is it +/-2.5V or single +5V?

    I have ordered a THS4303EVM board to see whether I can correlate your response without the 50-ohm termination to GND. I will keep you updated on it. I don't think you should see such a drop in S21 unless the part is powered down. I would expect somewhat of an RC low pass filter characteristic with the S21 gain holding up at least at the lower frequency end of 10MHz.

    Best Regards,

    Rohit 

  • 0 in reply to Rohit Bhat
    Prodigy 80 points
    Rohit,

    I am operating with +/- 2V supply. I will try the port conversion. Thank you for you help.
  • 0 in reply to Joseph Fox
    TI__Genius 15975 points

    Hi Joseph,

    I replicated the issue you are seeing in the lab. I can correlate to what you are seeing on the bench with regards to the S21 having significant loss by removing the non-inverting input 50-ohm resistor.

    It turns out that the issue is related to the input common mode voltage not being centered close to GND, because the network analyzer is inherently ac-coupled. This causes the output common mode voltage to clip close to the positive rail, thereby giving no signal at the output. You can correct this by using a Bias-T, like ZFBT-4R2GW, where the RF port is connected to the input port from the network analyzer, the DC port is set to GND and the RF&DC port is connected to the THS4303EVM input. I think this should provide a flat S21 gain response similar to the one obtained using a 50-ohm at the non-inverting input. I have verified this on the bench in my lab.

    Best Regards,

    Rohit

  • 0 in reply to Rohit Bhat
    Prodigy 80 points

    Rohit,

    Thank you. I will try that approach and get back to you with the results. Hopefully by the end of this week.

  • 0 in reply to Joseph Fox
    Prodigy 80 points

    Rohit

    Yes using a bias-tee with the network analyzer resolves the issue with the appearance of high losses operating without the 50 load @ the non-inverting input. I appreciate the extra steps you took. Thank you for all of your help.