0
\$\begingroup\$

Im working on a junior project in which we need to buffer an input signal from a RX antenna, pass the signal through 3 filters in parallel, and pipe the outputted signal into a RTL-SDR. It serves very little practical purpose as i know that the filtering could be accomplished digitally, but for the project we need to design an analog component.

My analog component is essentially going to comprise of: -> Recieve signal from antenna -> sma connector to pcb board -> input buffer op amp(Ive read these are important at buffering our input) -> active parallel filters(1 x band-pass, 1 x high-pass, 1 x low-pass) -> op-amp summer -> output sma connector (will be connected via short sma cable to SDR)

The frequency range we are going to be sampling is 700KHz - 1300Khz (part of the AM radio band).

This is my first exposure into looking at op-amps and researching the right ones to purchase for this project. In our lab i found a few LF357N op-amps and saw on the data sheet that they had a wide band-width of 20MHz. I figured i would try to breadboard a quick circuit to test the feasibility of this design. All i was trying to achieve was unity-gain using a simple non-inverting design. ** I understand that there are huge problems once we start getting into high frequencies in terms of signal distortion and loss on lines, especially bread boarding. This is why i wanted to make a mock up design before eventually trying my hand at creating a pcb for the final design of this project.

After i was having issues with this, i went to Multisim and started simulating the most simple circuit i could with the LF357N.

What ive seen is that while unity gain seems to work at very low frequency(10Khz), once i bumped my frequency up to 1MHz, the gain skyrocketed to tens of thousands. Then i threw in 2 resistors to give myself a gain of 2, which worked very well. But as i increased the frequency in 1MHz increments, i noticed the gain was affected adversely again.

Now, my question is why? Is the LF357N not unity gain stable at high frequencies? Even though the datasheet said the bandwidth goes up to 20MHz? What should i look for in op-amps for buffer amps?

Will the same op-amps work for buffering and summing? Or will i need to find a different style op-amp for both of those operations. My understanding is that i will just need 1 kind, as long as the Unity-Gain-Bandwidth is high enough.

Thank you for any help you can give me! I know this may be fairly trivial information, but Ive been searching the web for a few weeks trying to figure all this out and figured i would finally ask a question on it.

Low Freq High Freq High Freq- Gain2 High Freq- 2MHz - Gain2

\$\endgroup\$
5
  • \$\begingroup\$ For better simulation, you should add series 50 ohm resistor to the source, then probably a matching 50 ohm resistor (parallel) on the output of the opamp. But the real matching is accomplished with LC network, it also serves as bandpass filter. \$\endgroup\$ Commented Feb 26, 2020 at 20:26
  • \$\begingroup\$ It's gain drops at high frequency - at 1700 kHz, only has open-loop gain of about 20. And it is noisier than a single transistor. Please describe your antenna - it would help us to know what impedance it would present to this buffer stage. \$\endgroup\$
    – glen_geek
    Commented Feb 26, 2020 at 20:40
  • \$\begingroup\$ @MarkoBuršič Gotcha, i will add those resistances in on my next go at this. I currently have said goodbye to the 357 and chosen a new op amp - the TL084CN(there arent too many options here on campus). I have simulated my circuit again with much better results. Unity Gain is stable at 1MHz. Now im lookin into op amps to use for my gain stages after my filters! \$\endgroup\$
    – Brett Ryan
    Commented Feb 27, 2020 at 18:25
  • \$\begingroup\$ @glen_geek I have not had much time to look at the antenna we are using yet. It is the simple antenna that comes with the RTL-SDR dongles. Would it be difficult to create my own dipole antenna for AM radio, and solder a sma cable onto the end of it? Or would this route cause me more headaches with impedance matching than i care for? Thanks again, Im very interested in these topics and am trying to learn as much as possible. \$\endgroup\$
    – Brett Ryan
    Commented Feb 27, 2020 at 18:28
  • \$\begingroup\$ arrl.org/files/file/Technology/tis/info/pdf/0109031.pdf describes an antenna + buffer appropriate for an RTL-SDR dongle that has been supplemented with a front-end up-converter. The antenna itself is a voltage-probe type: a short vertical whip that should be mounted outside, high up. The buffer stage must reside at the bottom end of the whip. It is a high-performance design. \$\endgroup\$
    – glen_geek
    Commented Feb 27, 2020 at 19:20

1 Answer 1

1
\$\begingroup\$

The 357 is not unity gain stable at any frequency. It requires a gain of at least 5. Simulators are unreliable on this sort of thing.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.