EMF Probe Project

Discussion in 'The Projects Forum' started by Butterworth, May 27, 2011.

  1. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    So I have been working of this project for some time, and finally have a working prototype of the finished porject, but I have run into some issues with inconsistant readings on the panel meter, and audio output / volume. Attached is the circuit layout I have created based on a schematic I found.
    *This is not my invention but rather my take on it.*

    I am looking for any possible reason why I am getting radical and inconsistant readings form my panel meter and audio, based on the circuit I posted here. If anyone has any suggestions please let me know. The more info the merrier.

    Thanks eveyone! :)
     
    Last edited: May 31, 2011
  2. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    *bump*

    No body in the house has any input?

    Please I really need opinions on this.

    Thx.
     
  3. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    What do you expect to pick up?

    It will pickup 50Hz or 60Hz mains hum.
    Its opamps have a voltage gain of only 3 and a max output of only 4V in the middle of the AM broadcast band. It will have NO voltage gain above 3MHz.

    The output capacitor value 220uF is huge but an opamp cannot drive a speaker, it can drive 2k ohms minimum. You need a power amp like an LM386 to drive a speaker to a low power level.
     
  4. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    Thank you for the reply Audioguru,

    The circuit worked on my breadboard originally, picked up both 50 / 60 cycle hums, as well as some "white noise" from radio frequencies, also any magnetic fields... but when I created the circuit layout and applied the components, the outcome was not the same. I am trying to find the faults on the circuit that would cause the eratic readings and inconsistance audio.

    The audio circuit drives small 32ohm earbuds. I do not intend on using a larger speaker, maybe a relay but nothing larger than the ear buds.
     
  5. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    The only circuit problem I see is that your input amplifier has DC feedback that will amplify the op amp's input offset voltage by 470k/R(L1), where R(L1) is the DC resistance of your inductor. For example, if R(L1)=1Ω, and the input offset voltage is is greater than ±20 microvolts(!), your op amp will saturate at one of the supply rails. Typical input offset voltage is ±3 millivolts.
    You have to AC couple the inductor. Unfortunately, 1mH is a very low impedance at 50Hz, so you will need a big capacitor. I ran some simulations. 1000uF works well in the sim. The 3.3Ω resistor damps the resonance of the LC combination.

    You have several schematic errors, which I have circled in red. They don't appear to be in the layout.

    I would add a 10pF cap across R1 to compensate for the self-resonance of the inductor. I did not add this to the schematic.
     
  6. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    Thank you for the info!

    I will couple R1 with a 10pF cap for resonance, but I don't think I can fit a 1000uF cap in the project box I am using. I have a 1000uF cap but it measures about 1.5" tall x 1/2" diameter...

    The R5 & R6 were labled wrong on my schematic, it should be reversed where R5 is grounded instead.

    Based on the pcb layout I made, would you consider anything there in error? What do you think the obvious causes for interference would be?
     
    Last edited: May 30, 2011
  7. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    It won't work without the big cap in series with the inductor.
    There were other schematic errors, which you should fix and repost. If you want more help, a schematic with errors is a big turnoff.
     
  8. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    Thank you Ron H, I have made the revisions you suggested and fixed the errors on the schematic, it is uploaded as 'schematic R1' on the original post.

    Its odd but the original schematic did work on my breadboard, but not when I put it on the pcb (due to the obvious errors) but now with the errors corrected, should this schematic work without interference?
     
  9. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    If you used a different IC for the op amp, and I mean a physically different IC, not just another type, then you could have gotten very lucky with the input offset voltage of that particular part.
    If your probe inductor's DC resistance is higher than the 1 ohm I postulated, then the circuit would be less sensitive to input offset voltage.
    Do you have an ohmmeter?
     
  10. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    I have a multimeter. The original author of this schematic designed the circuit around the TL084 quad opamp, which I cannot understand how if it was not supposed to work, it did. When I had the prototype on the breadboard, it picked up 50/60 cycle hums and RF in the form of white noise. The RF was more sensitive to hear than the mains hum. I picked up the RF (in my printer no less :confused:) from about 8-10" away, while the mains were only picked up from about 3-5" away for both audible & meter readings.

    Seeing as it is twitchy now, what do you suggest I replace the Opamp with, if anything? Or can I keep the TL084?
     
  11. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    TL074 is a low noise version of TL084.
    My question about the multimeter was intended to spur you to measure the DC resistance of your inductor, and report it here. I guess I was too subtle.:)
     
  12. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3

    Haha, gotcha. Well I can do that later on this afternoon since I am at work right now. But I can take the measurement and post it tonight. Should the measurement be taken with or without the lead wire? (The inductor is on a 1/8" mono headphone jack)

    The TL084, I thought, was already low noise? I would really like to keep the IC I bought already, try to keep the new parts to a minimum if possible.
     
  13. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Your choice. I have no idea how much the op amp's internal noise will affect the output.
     
  14. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3
    Hmm, well I think I will use the TL084... seeing as I have 2 of them! ;)

    I will get that measurement with the lead wire around 5pm EST. (GMT -5)
     
  15. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Nobody uses a noisy (hiss or white noise) TL084 opamp when a low noise TL074 audio opamp costs the same or less.

    You are just guessing at the amount of noise from a TL084 because an entire production run might all be low noise or might all be noisy. A TL074 is guaranteed to be low noise and you should use one or a better one.

    EDIT: Why do you want to pickup the interference from the power supply in your printer?????
     
  16. Butterworth

    Thread Starter Active Member

    May 6, 2009
    135
    3

    Ok, well I can pick up the TL074 if it is guaranteed low noise, I just didn't want to waste the parts I already have you know? But if it will improve the performance, then I will pick one up.

    I do not want to pick up interference from my printer... I never said that, but the probe picked up an RF signal in the form of white noise from the printer... I do not know why.


    @ Ron H - I have measured the DC resistance of the inductor with the lead wire and it measures 3.2Ω on the 200Ω scale.
     
  17. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    OK, now think about this:
    Referring to your original circuit (not mine), imagine the input offset voltage of the op amp is 1 millivolt. This means that the feedback will only be satisfied if there is 1mV between the +in and -in pins, i.e., the output will try to go to a voltage that forces exactly enough current through the feedback resistor (470k) to cause a 1mV drop across the 3.2 ohm DC resistance of your probe. This means that you would need 1mV/3.2Ω=312.5uA through the 470k resistor. This means that the output would have to go to 312.5uA*470k=147V above (or below, bepending on the sign of the offset) V/2!. Obviously, the op amp will saturate at one of the rails. This is why you need to AC couple the sensor, as I posted earlier.
     
  18. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Please post the link to the original circuit and concept that this thing is based on. :confused:
     
  19. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
  20. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
Loading...