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

FM transmitter sensitivity

  1. Jun 30, 2009 #1
    I am trying to amplify extremely small signal(around 200 nanovolt) from a coil of wire on a ferrite to make a gaussmeter, I've seen some FM transmitter schematics online that I want to connect this signal to, Would a FM transmitter be able to work with a signal this low?
    I tried amplifying with op amps(OP07) but the noise was more than signal.

    thank you
     
  2. jcsd
  3. Jun 30, 2009 #2

    vk6kro

    User Avatar
    Science Advisor

    No, probably not.
    That is 0.2 microvolts. An extremely small signal and difficult to handle. Even the best receivers would have trouble with such a signal if it was at radio frequency.
    Trying to modulate an FM transmitter with this would not give any result.

    Maybe you could consider using a Hall Probe chip. One type I have used is a UGN3503U which is readily available in Australia.
    Output is 2.5 volts with no magnetic field with a 5 volt supply and the output voltage goes up for one magnetic polarity and down for the other.
    One benefit of Hall Probe chips is that they work on static magnetic fields where coils only work on varying magnetic fields.
    Also these chips have amplifiers in them to give reasonable output. They are not super sensitive, though, and may still require more amplification.
    You have to regulate the 5 volt supply to them with a low noise regulator to keep the output noise low.
     
  4. Jun 30, 2009 #3
    You would need to amplify it substantially before you could use it to modulate a carrier. Using a 200 nV signal will be difficult but not impossible. However you need to answer a few questions.

    What frequency is your signal? Over how wide a frequency range do you need to detect the signal? How good a signal to noise ratio do you need in order for it to be useful?

    Opamps are not the way to go. You'll need very low noise transistors and good filtering. This will not be a trivial project.
     
  5. Jun 30, 2009 #4

    Basicaly I'm trying to make a gaussmeter that is VERY sensitive I would like the frequency range to be 100 to to about 4000 hz , for S/N ratio I need the signal amplitude to be atleast 20 times higher than the noise I don't know what db that is. And if you could tell me what transistors and other components to use for least noise possible I would really appreciate it.
    I thought modulating with FM would be a good idea because I saw a GMR sensor(more sensitive than hall device) and on their site it says "for really low field detection noise is the limit and an AC Modulation/Demodulation scheme should be implemented", isn't FM like this? what do they mean by this, the site is NVE,
    go to:
    -analog sensors
    -application notes
    -More sensor application notes from the Sensor Catalog
    -signal conditioning.
     
  6. Jul 1, 2009 #5
    Your range of input frequencies is so wide (over 5 octaves) I recommend a narrowband tunable input filter to reduce noise. Perhaps you could combine this with your pickup coil. As for the transistor, do some searches on the internet for very low noise transistors and see what you can learn about them and low noise amplifiers in general.

    One question I forgot to ask is what is the maximum voltage you expect to see at the input?
     
    Last edited: Jul 1, 2009
  7. Jul 1, 2009 #6
  8. Jul 1, 2009 #7

    vk6kro

    User Avatar
    Science Advisor

    Maybe it would be possible to improve the probe?

    I am trying to amplify extremely small signal(around 200 nanovolt) from a coil of wire on a ferrite

    Could you describe what you are using here? Is it a ferrite toroid or a short bar or something else? How many turns of wire are you using?
     
  9. Jul 1, 2009 #8
    I'm using 7 inch 1 cm ferrite rod 38 gauge magnetic wire 6000 turns. The output is high at close distance.
     
  10. Jul 1, 2009 #9
    approximately 200nV to 500nV with the fields I will be measuring but the mains hum puts about 400mV. I will need lowpass and highpass filters.
     
    Last edited: Jul 1, 2009
  11. Jul 1, 2009 #10
    How will you calibrate it?

    I'm thinking you might switch in various values of capacitors across the coil with a selector switch in order to tune it to various frequencies. Narrowing your bandwidth will increase your signal to noise ratio.

    If your amplifier is extremely sensitive, you may have a problem if your signal is too strong. You may want to have a switchable attenuator for the amplifier.

    Another approach you might want to try is an RF receiver IC. I know this isn't RF but most of them work at low frequencies too. Some are very sensitive and usually have an RSSI (received signal strength indicator) which gives a logarithmic output, not very accurate but are good over a wide range of input signal strengths, often 70 to 90 dB of range.
     
  12. Jul 1, 2009 #11
    I made a simple 3 opamp amplifier and compared it to a gaussmeter I bought, I put a 200 Hz frequency to a headphone and measures from different distances the magnetic field, I noticed that with the exact same sensor coil the one I bought measured:
    150mV at 1 inch
    43mV at 2
    10mV at 3
    3mV at 4

    The one I made measured(adjusted to measure 150 at 1 inch):
    33mV at 2
    2.5mV at 3
    300 microvolt at 4

    If they start with the same amplification is the drop off rate so different
     
  13. Jul 1, 2009 #12
    Plot both sets of values using a log vertical scale. The gaussmeter's output follows a straight line, yours has a break at 2 inches. Check to see if the quiescent point is in the middle of the output range on each opamp.
     
  14. Jul 1, 2009 #13
    This is complicated I think I will just have to hire an electrical engineer
     
  15. Jul 1, 2009 #14
    Questions:

    How did you measure the voltages?

    Were the measurements at the input or output of your 3 opamps?

    What is your power supply voltage?

    What voltage do you see at the outputs of each of your opamps with no signal applied (input shorted to ground)?

    Do you have any diodes, transistors or other potentially non-linear elements in your circuit besides opamps?

    What is the opamp you're using?
     
  16. Jul 1, 2009 #15



    I think the easiest way is to use a gaussmeter that I already have the only problem is that it has a DC output I need to somehow modify the circuit to make it AC output.
    I will draw a schematic from the board and post it and i'll see if anyone can tell me what changes to make to get AC output. Do you think you would be able to make the changes to make it AC ?, the circuit has ,op amps, PNP NPN transistors, diodes, one MC14046BCP and ofcourse resistors and caps.
     
  17. Jul 1, 2009 #16
    A schematic would certainly help. Is the schematic of your amplifier or of the gaussmeter? Why do you need an AC output?
     
  18. Jul 1, 2009 #17
    I'm making a schematic of the gaussmeter, I won't make my own amp if I can make this AC output. AC so I can also get frequency of fields.
     
  19. Jul 2, 2009 #18

    vk6kro

    User Avatar
    Science Advisor

    You haven't said what the source of your magnetic field is.

    If it was in a very confined location you could use a modified toroid to detect it.

    I haven't tried this, but you get a toroid and use an angle grinder (carefully!) to remove a small segment of the ferrite.
    You wind a coil on the remaining section of the toroid and use this to take off a signal.
    Then the main pickup is from the missing segment of the ferrite.

    This may help to remove the hum from mains pickup.

    Putting ferrite material in a magnetic field will modify the field you are trying to measure, though so I guess you are already allowing for this.

    The heads used in tape recorders used a similar principle to detect very small magnetic fields from magnetic tapes.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: FM transmitter sensitivity
  1. FM Transmitter (Replies: 4)

Loading...