• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 49 subscribers
  • Views 592 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.

IP3 of OPA842

Karsten Hansen
Prodigy 50 points
Other Parts Discussed in Thread: OPA842

I am using the OPA842 in a traditional trans-impedance configuration with a BF862 as input buffer (included in the feedback-loop). Gain is 20dB up to around 10MHz.

I have measured IP3 on this amplifier with f1:1.1MHz and f2:1.11Mhz and found that the slope of the third-order products is far less than 3 times the slope of the fundamentals. In fact the slope was only 1.3 times (in dB), which means that the resulting IP3 is highly depending of the level of the input signal.

Is this a normal behaviour?

If so, it seems to make IP3 a meaningless figure or at least you have to specify the test conditions.

Looking forward to seeing your comments on this. Thanks in advance.

Best regards

Karsten 

  • 0
    TI__Genius 12631 points
    Hello Karsten,
    Can you please send a schematic of the circuit you are describing.
    -Samir
  • 0 in reply to Samir Cherian
    Prodigy 50 points

    Hi Samir,

    this is the amplifier I am working on. When measuring the IP3 I added a 50 ohm shunt resistor in front of the 470ohm input resistor to achieve a good input return loss. 

    Thanks in advance for your effort.

    Best regards

    Karsten

  • 0 in reply to Karsten Hansen
    TI__Genius 12631 points
    Hello Karsten,
    Its possible that you are running into the measurement limitations of the spectrum analyzer at these low frequencies. Can you please send me your measured values so that I can do some baseline measurements here to make a comparison. Also, if it is indeed a measurement limitation of the instrument, bumping the fundamental frequency to 15MHz should help to make the measurement easier. I spoke to the designer of the part and he believes that down at 1MHz the part should be very well behaved.
    -Samir
  • 0 in reply to Samir Cherian
    Prodigy 50 points

    Hi Samir,

    I used a ZNB 8 network analyzer for the test as described in the attached file. It goes down the 9kHz so the issue is not a bandwidth limitation of the instrument.

    I have just noticed that at high input port to the amplifier the slope seems more "normal", unfortunately I dont have the instrument available any more, so I cannot decide if it is a test error.

    Once again thanks for your assistance.

    Best regards

    KarstenTest of IP2 and 3 using the ZNB 8.xlsx

  • 0 in reply to Karsten Hansen
    TI__Genius 12631 points
    Hello Karsten,
    I think I see the issue. The problem is not with the frequency range of your instrument but with the dynamic range. Spectrum and Network analyzers typically demonstrate a dynamic range from 80dBc to 100dBc. It looks like your analyzers DR may be around 84dBc. This is limiting your measurement at lower powers but at higher powers the measurement is being lifted off the noise floor. We usually run into similar issues when measuring Harmonic Distortion as well. Hope this makes sense. Let me know if you have any further questions.
    -Samir