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

Interfacing antenna to ADC

  1. Jan 11, 2010 #1
    greetings all,

    I am trying to start some undersampling RF to ADC hobby stuff. I am looking at this development board:
    http://www.digikey.com/scripts/dksearch/dksus.dll?KeywordSearch?Keywords=ADC11C170HFEB&vendor=14

    I am having a little trouble understanding how to connect the receiver antenna to the board. the board already has DC blocking caps, 1:1 balun transformers, and a termination circuit (to the differential inputs of the ADC) (see the image below)

    obviously I need the antenna, a SAW/BAW filter, and probably a Low Noise Amplifier (LNA). however, the input of the ADC is +/-2.5v (approx, depends on reference). how do I know what voltage I will have at the antenna, and how much the LNA needs to amplify. obviously I want the signal close to the peak for the best dynamic range, but I don't want to go over the max. do I need an adjustable gain amp with clipping detection or something? if so, can anyone give me some keywords for searching, or point me in the right direction?

    Thanks
    (here is the image of the input circuit on the development board)
    http://img138.imageshack.us/img138/6084/adccircuit.jpg [Broken]
     
    Last edited by a moderator: May 4, 2017
  2. jcsd
  3. Jan 12, 2010 #2

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    Your question is a bit like "how long is a piece of string?"
    You would need to know the frequency, the field strength of the received signal and the kind of antenna. However, you would be dealing with less than 1mV for a usable 150MHz off air signal (say 1mV/m), received with a half wave dipole.
    There's a calculator here _ http://www.giangrandi.ch/electronics/anttool/field.html [Broken] which will help you, once you know the frequency and field strength of your signal. (The receiver input volts can be calculated assuming a 50 Ohm input and using
    P = V2/R)

    Your front end amplification will need to be something like 60dB (VoltsX1000), I think, so you will need some effective filtering to reject interfering signals in adjacent bands.
     
    Last edited by a moderator: May 4, 2017
  4. Jan 12, 2010 #3
    sorry if I didn't give all the specifications. I was looking for sort of a general answer, for the sake of those who would reply. I don't want to make people have to crunch a bunch of numbers just to answer the question.

    however, if it helps, we are talking 1090mhz, with a desired distance from 1 to 30 miles, so about 95 to 125db free space loss (unless I calculated wrong). transmitted power is in the neighborhood of 51dBm, depending on transmitter cable loss and antenna gain. I don't know what receiver antenna I will use yet, but I guess it would range from 2dBi to 6dBi.

    the above gives me a receive signal on the order of 1mv. although, I am still not sure how to properly amplify the signal. if the transmitter is close, I could get as much as 10mV but since I am not sure what remaining degradations are in my components (BPF/LNA/ADC/et cetera), I am not really sure of how weak of a signal I can sense at the ADC (through all the introduced component/board noise.). once I have the final configuration of the system figured out, I could get a much better idea of my losses, but I have a lot of unknowns at this stage.

    so, since I don't expect much over 10mV, should I shoot for a constant gain of 250 and just accept that farther transmitters will give me poorer dynamic range? or should i try to design with a variable gain amp?

    Thanks
     
  5. Jan 12, 2010 #4
    Did you consider mixing down the signal and then amplifying it?
     
  6. Jan 12, 2010 #5
    Waht said: "Did you consider mixing down the signal and then amplifying it?"

    yeah, mixing it down would probably be the easiest way to do it. however, the point of the project is to teach myself how to undersample high frequency as close to the antenna as I can get. therefore, it is more of a software define radio. because, in theory, I would just need to change the sampling rate (maybe) and filter/amp circuit at the antenna and I could demod anything up to the low ghz range.

    so, if you have any help on my questions, i would appreciate it.

    thanks much
     
  7. Jan 12, 2010 #6
  8. Jan 12, 2010 #7
    what is your reason for a chain of amps? I can find plenty of single RF amps with 25, 30, 35, and even 40db gain that extend into ghz range. the advantages that I see are (1) I could possibly combine two with very low noise figures to have the same gain but lower noise figure when compared to a single amp, and (2) that I could mix and match amps to get the exact gain I want.

    my biggest problem is knowing what gain I should choose. obviously with the ADC, you want the highest dynamic range possible so, in theory, the peak of my signal would be at 2.5v all the time (as long as it's not AM modulated :P ). my problem is picking a gain so that I don't rail my amps at high power, but also high enough that I can still demod lower power. I suppose I could always conservatively design for a dipole antenna, and if my gain is low, switch to a Yagi or something :P

    so, is it worth going to a variable gain amplifier (VGA), which usually comes with a high noise figure? or should I just make my best guess on a low-noise static gain?

    [begin rant], if I am adding a VGA, I should probably re-spin the development board based on the application note, rather than adding such a complicated board to sit between the ADC board and the antenna. that leads to another headache. I can design boards just fine, but why open another can of worms in an already complicated project.[end rant]

    Thanks!

    p.s. does anyone know anything about crystal oscillators for undersampling? most oscillators are rated for frequency stability, but I think that just refers to how accurate the frequency is. for my purposes, I care less about how accurate it is (over the long term), and more about jitter (short term variations). do you think it is worth shelling out $180 for a TCXO with a frequency stability as slow as 10 or 20 parts per billion (ppb)
     
    Last edited: Jan 12, 2010
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Interfacing antenna to ADC
  1. ADC and DAC (Replies: 4)

  2. Current ADC (Replies: 5)

Loading...