Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 615 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: Setup and results

James Prince
Prodigy 80 points
Part Number: THS4303

My setup:

Using THS4303EVM with a Rigol DSA-815 Spectrum analyzer and Tracking generator (TG) sweeping from 1Mhz to 1.5 Ghz.  The THS4303EVM input is connected to the TG and the output is connected  to the spectrum analyzer input. The power supply is +- 2.5V. The power supply current is 40 mA.

Does anyone have ( or can someone produce ) the expected output?

I expected the output amplitude to be flat beyond 1.5 GHZ.

I am seeing 100 to 300 mV with a 25mV input. This make no sense to me. I expect a gain of 2.5 V/V. Gain = .5 * 10 * .5, input 50 ohm "divider" X amplifier gain 10 X .5 output "divider" = 2.5.

Jim 

  • 0
    Guru 140200 points
    Hi James,

    have you confirmed that the tracking generator is working properly? Is the output amplitude correct? Is the frequency response flat? Do you use 50R cables? Are the input and output impedances set to 50R?

    Kai
  • 0
    TI__Genius 15555 points

    Hey James,

    Here is a simulation of the expected small signal AC response, sweeping from 1MHz to 1.5GHz on TINA.

    James.TSC

    Best,

    Hasan Babiker

  • 0 in reply to Hasan Babiker31
    Prodigy 80 points

    Hi,

    Thanks for you response.

    I am really looking for a measurement of the THS4303EVM. One would think that, someone  would have taken a measurement of an evaluation board before release for sale.

    I would think that setting one up and taking a measurement would be routine in a lab.

    Since this is a first for me, I would like to know what to expect.

    Simulations are good but there are many issues that could come up with a hardware build such that it will not match the simulation results.

    It is possible that my setup is wrong, so again I am looking for a reference measurement.

    Jim Prince

  • 0 in reply to James Prince
    TI__Genius 15555 points
    Hello Jim,

    Out EVM boards are setup in the test configuration used in the datasheet. For testing results please refer to the figures in the THS4303 datasheet. Note that these responses account for impedance matching and don't show the 0.5 gain in the input and output.

    Best,
    Hasan Babiker
  • 0 in reply to Hasan Babiker31
    Prodigy 80 points

    Clearly those are not measurements.

    We are not talking about the design, We are talking about a board that I paid $50 for.

    Can I expect someone to measure an EVM board as requested ?

    Jim Prince

  • 0 in reply to Hasan Babiker31
    Prodigy 80 points
    We are talking around the issue. Where are the measurements ?
  • 0 in reply to James Prince
    TI__Genius 15555 points

    Hey Jim,

    To clarify, the figures on the datasheet are measured responses of the device not simulations. Can you share the results achieved from your setup? You mention expecting a flat response beyond 1.5GHz, are you referring to the 1.8GHz bandwidth in the datasheet? The response should actually begin to lower around 500-600MHZ with 1.8GHz being the -3dB point. I can run a test on an EVM and will get back to you with the results, but if you are looking for a reference on what you should expect both the TINA sims and the datasheet figures are reliable indicators.

    Best,
    Hasan Babiker

  • 0 in reply to Hasan Babiker31
    Prodigy 80 points

    Hi Hasan,

    I have run the TINA sims and have reviewed the datasheet. Since my measurements do not reflect the expected results from the simulation something is not right. I suspect that something is wrong with my test setup. But it is possible that the EVM is not performing as expected. Since I do not have confidence in my equipment / setup, I can not resolve if the issue is the DUT or test setup. Therefore I am asking for this help.

    When I said that the the datasheet graphs are not measurements, what I meant is that the graphs that do not show the results of the front and back divide by 2 are not measurements of the EVM circuit. 

    I am requesting the EVM measurement (and setup notes) to resolve whether it is the EVM or my measurements. Although I am a power supply guy; not really an RF guy; I still need to run the issue to ground.

    In that my measurement setup has to be right before I can test the prototype of the product circuit which will introduce more unknowns.

    I currently do not have a good way to share my measurements. if it a setup error i fear that a graph would not give new insight without the reference graph that I am asking for.

    If you would like, we can take up a conversation about the product that I am developing privately. Direct emails and/or phone conversation would be fine.

    Thank you for your offer to take the measurements. I am much more interested in the Voltage gain, not so much the power gain but I can convert.

    Jim Prince

  • 0 in reply to James Prince
    Guru 140200 points
    Hi James,

    all such measurements are taken with these 50R divider resistances, because there's no other way to deal with the cablings than to use the technique of matching of characteristic impedance of cable by using 50R termination resistors. In a second step the frequence response is corrected for the 6dB losses.

    Doing measurements in the GHz range is very demanding. You need excellent (and expensive!) 50R cables!

    Kai
  • 0 in reply to kai klaas69
    TI__Genius 15555 points
    Hey James,

    Kai is correct, you will also find on page 14 of the datasheet, "Figure 43 is the noninverting gain configuration used to demonstrate the typical performance curves. Most of the curves were characterized using signal sources with 50-Ω source impedance, and with measurement equipment presenting a 50-Ω load impedance. In Figure 43, the 49.9-Ω shunt resistor at the VIN terminal matches the source impedance of the test generator. The 50-Ω series resistor at the VO terminal in addition to the 50-Ω load impedance of the test."

    Our characterization curves are done using the EVM. The 6dB loss from the input and output are just corrected on these curves to show the performance of the device. Shifting the AC response in the datasheet down by 6dB will give you the curve you are looking for. If you would like to go into detail on your measurement setup but can't on this forum, you can email me at h-babiker@ti.com so we can figure out where the issue may lie.

    Best,
    Hasan Babiker