• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 48 subscribers
  • Views 502 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

UAF42 output amplitude change with scope sampling rate

Khalil Zahar
Prodigy 30 points

Other Parts Discussed in Thread: UAF42

Hi,

I am currently testing the UAF42 as a 60 Hz notch filter for a prototype design but there is something that I misunderstand: Although the filter works fine, so it does filter the 60 Hz component quite well, the output (in this case I am only measuring the noise in the system) gets larger when I increase the oscilloscope sampling frequency, and it gets lower when I reduce it. Now I don't know if it is a filter characteristics or that I need an anti-alias filter (although I would think that recent digital oscilloscopes have a built-in anti-alias filter but I may be wrong); However this ONLY happens when I filter the signal with the UAF42. The signal magnitude stays the same at all frequencies if the filter is not in the circuit. 

Can you advise on this?

Thank you,

Khalil

  • 0
    TI__Guru* 93105 points

    Hi Khalil,

    The UAF42 gain and noise should not be affected by the oscilloscope sampling frequency setting; baring anything unexpected and unusual about the instrument's behavior.

    You mention that the signal magnitude stays the same at all frequencies when the filter is not in the circuit. Do note that your generator is most likely designed to drive a 50 Ohm load. If it is connected via a 50 Ohm cable directly to the oscilloscope input - set to the 50 Ohms termination impedance setting - the amplitude should remain nearly constant as you change the and scope settings. The UAF42 has a very high input impedance and its input should be terminated with a 50 Ohm terminator when making these kinds of measurements.

    Also, the UAF42 output cannot drive 50 Ohms directly so the oscilloscope input should be set to the high-impedance setting, usually 1 M-Ohm. The cable connection between the UAF42 output and the oscilloscope input should be as short as possible, a few inches or tens-of centimeters at most. Alternately, a resistive divider can be set up at the output where the 50 Ohm oscilloscope input impedance is the lower resistor of the divider. A drawback to this approach is the output is attenuated by the divider and low level signals can get buried in the noise.

    Do you have access to an analog oscilloscope? If so, checking the circuit with it might help confirm, or dispel, any questions about the digital oscilloscope's measurements. I keep an old Tek 2465 in the lab for times when I am just not exactly sure about what I am seeing with the DSO.

    Regards, Thomas

    PA - Linear Applications Engineering

     

  • 0 in reply to Thomas Kuehl
    Prodigy 30 points

    Thomas, thank you for your quick answer. I will look more into it.

  • 0 in reply to Khalil Zahar
    Prodigy 30 points

    Hi Thomas,

    It works fine now. The problem that I have now is really the 60 Hz harmonics. Is there a way to tune the UAf42 to filter these harmonics as much as possible besides just putting new filters? Also, for best results, would it be better to put the filter directly at the power output? or it does not change anything since the filter applies to the whole system?...

    Thank you

    Khalil

  • 0 in reply to Khalil Zahar
    TI__Guru* 93105 points

    Hi Khalil,

    The UAF42 filter function is intended for a specific low-pass, high-pass, band-pass, or band-reject (notch) response designed in relation to a specific frequency. I don't see any way to obtain harmonic filtering (low-pass) in addition to the notch function with one UAF42. It is probably best to use a separate low-pass function to do the harmonic filtering. Another UAF42 could be used for that purpose, or a different filter employing an op-amp in the Sallen-Key, or Multiple-Feedback topology could be used. TI's FilterPro tool can be used to synthesize these types of active filters.

    http://www.ti.com/tool/filterpro

    If the noise performance of the UAF42 filter stage is lower than that of a stage that precedes it in the signal chain, then it would be advantageous to place it first in the line-up.

    Regards, Thomas

    PA - Linear Applications Engineering