Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Opamp for MF/RF ranges (1.0-1.2MHz)

  1. May 2, 2017 #1
    Hello!

    I have a bit of trouble finding general op-amps that can amplify at medium frequency ranges, and with a good enough gain of 200 or 46dB. My max desired operating frequency is around 1.2MHz so I don't really care if it no longer amplifies much at frequencies higher than this.

    I am aware the good way to go would be RF amplifiers, but they're hard to find in my area. For example, part #s like ADs, THS and CLCs, I'll have to order overseas but I'm constrained by time. I'm already looking at some datasheets, specifically checking their Max Output and Gain v Frequency but haven't found satisfactory ones yet.

    Also, I already read one thread regarding this, the one posted by amanno. But his/her range was around 2000MHz, so in his/her case, I believe RF amps really are necessary. However, mine is only at around 1.2MHz so I was hoping I could still use the non-RF ones. Is it possible to find LM/LF that can operate on this range?

    I'm still a noob with RF and amps so I'm hoping someone could guide me on this. Thanks!
     
  2. jcsd
  3. May 2, 2017 #2

    berkeman

    User Avatar

    Staff: Mentor

    Welcome to the PF.

    Have you looked at the amplifiers available from Mini-Circuits? https://www.minicircuits.com/

    They have lots of higher-frequency modules, but they have lower frequency amplifiers in your frequency range as well.

    Beyond that, can you say more about your application? A gain of 200 at 1MHz is non-trivial, but very do-able with discretes and/or a canned amplifier...
     
  4. May 3, 2017 #3
    Hello! Unfortunately, I am not from the US. Also, what do you mean by canned amplifier?
     
  5. May 3, 2017 #4

    berkeman

    User Avatar

    Staff: Mentor

    You didn't say yet what your application is...
    What does that mean? Does it mean you cannot purchase items from Mini-Circuits? You should still be able to find what you want at their website and then use Google to find similar items from suppliers in your country.
    A pre-assembled amplifier module. Like what you can buy from Mini-Circuits or from many other suppliers...

    http://electronicstudio.net/wideband tv amp copy.jpg
    wideband%20tv%20amp%20copy.jpg
     
  6. May 3, 2017 #5

    Baluncore

    User Avatar
    Science Advisor

    Current feedback amplifiers will solve your problem. They are asymmetric and have gain*BW products that will meet your requirements. Ask more questions when you have studied CFAs and can tell us your application.
    Start by looking at … https://en.wikipedia.org/wiki/Current-feedback_operational_amplifier
    Google ' current feedback amplifier '
    Many manufacturers provide data, tutorials and application notes.
    Go to suppliers websites and use the parameter search to identify available devices. digikey, farnell, mouser ...
     
  7. May 3, 2017 #6
  8. May 3, 2017 #7

    vk6kro

    User Avatar
    Science Advisor

    Since this is for a narrow bandwidth, you could ( and should) use tuned amplifiers.
    If you follow one amplifier with another, and both have a gain of 15, the combined gain will be about 15 times 15 or 225.

    I found an article I wrote a long time ago which may help.
    http://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/radcom/1991/07/page28b/
    Note that the FETs used were depletion mode N channel types.
    Winding the inductors would be tedious, but if you could find some old style transistor radios which used discrete components, you could salvage the oscillator transformer from some of them and use these.

    I built many amplifiers like this. They were always low noise and high gain. And stable.

    But, as has been mentioned, you need to describe your application.

    Where are you?
     
  9. May 4, 2017 #8
    Hello again! Sorry. I really am just using it for a simple RF amplifier block on an AM transmission circuit. So I need to amplify a carrier signal at around 1.2MHz (1.22MHz to be even more specific). I was just hoping if I could use op-amps rather than transistor-based, considering op-amps are much simpler to analyse and construct than FETs.

    [EDIT] Also, I tried testing with an LF353 opamp earlier (its one of the few opamps I got so far right now), and got an approximate gain of 20 on one stage, i.e. given a 10mV to around just 50mV output at 1MHz despite employing a non-inverting config that should give me around 200 gain.

    I checked its datasheet, and looking at the graphs, it should still be able to at least give a max Vout of 2.5V at around 1MHz. So, given a 10mV input, an output of 2V for a 200 gain should still be within range...? Moreover, even with an Av = 20, cascading it with an amp config that can give me Av = 10 for a total of Av = 200; somehow, however, probing at the second stage gives me an output of near 0 instead. Also, I've tested only the RF block itself, with no other circuit connected, so loading shouldn't be a concern.

    I'd be glad for any insights on this? Thanks again!
     
    Last edited: May 4, 2017
  10. May 4, 2017 #9

    berkeman

    User Avatar

    Staff: Mentor

    Is that in a licensed band or an unlicensed band? What power are you trying to transmit? Who will be receiving your transmissions?
     
  11. May 4, 2017 #10
    Actually, this is just for a class project to test and incorporate some RF topics discussed. We were given different specifications for carrier frequencies given an audio message signal. So, I guess it's just within the class... nothing really high-end. I already had most blocks done, except the RF amp for the things mentioned above.
     
  12. May 4, 2017 #11

    vk6kro

    User Avatar
    Science Advisor

    A problem you may have is that, in an opamp, you will have gain, but it also affects the DC operating conditions of the opamp.

    So, you will find it very easy to get an output which is equal to one of the supply voltages.

    You then get, say, +12 volts output but no AC component at 1.2 MHz.

    You may be able to do it with cascaded opamps.
    You drive one amplifier with another via capacitive coupling.
    If they both have a gain of 15, you would get a gain of 225. This makes the design of the amplifiers easier.

    More likely, though, it would be unstable and you would get uncontrollable oscillation.
    This is because the opamps will be amplifying a very wide bandwidth and unwanted phase shifts can easily cause oscillation.

    Unless your assignment insists that you use opamps, you would make your life easier by using discrete components like FETs or other transistors.
     
  13. May 5, 2017 #12
    EF184 pentode can do it with a single device and provide 100 RMS volts output!
     
    Last edited by a moderator: May 5, 2017
  14. May 5, 2017 #13

    vk6kro

    User Avatar
    Science Advisor

    Yes, but do you really want to go there?

    You would need a 6.3 volt heater supply, a 350 volt DC power supply, stringent safety precautions and some way of avoiding RF interference from the resulting device which would be a fairly potent transmitter.

    The opamp would have no way of producing 100 volts RMS, so it is probably not needed in this case.

    Having said all that, it would be a fun, nostalgic project. :)
     
  15. May 5, 2017 #14
    Just to be Devil's Advocate, in the 50s we might have said that transistor solutions would require difficult, high current power supplies, maybe involving numerous components, would be non linear without NFB, have low input impedance, provide small power output with limited voltage swing, would require a higher component count, be vulnerable to overload, supply reversal and thermal effects, have limited frequency response and might be noisy!
     
  16. May 5, 2017 #15

    Baluncore

    User Avatar
    Science Advisor

    Maybe you should start with some specifications. Are these correct?
    The centre frequency; Fixed 1.22 MHz. Bandwidth; +/– 20 kHz. Q = 30.
    Input impedance; 50 ohm. Output impedance; 50 ohm.
    Gain required; 46 dB. Voltage gain; 200.
    Minimum input signal; 10 mVrms.
    Maximum output signal; 2 Vrms.
    LC tune the output to eliminate harmonics.
     
  17. May 5, 2017 #16

    vk6kro

    User Avatar
    Science Advisor

    Yes, that would cover it.

    It would be easy with a simple FET amplifier with tuned step-up input and output circuits. Followed by an emitter follower, probably.

    But trying to make a RC amplifier with an opamp or a vacuum tube to do this seems to be asking for trouble.

    Note VK7.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted