# Why does this antenna (piece of wire) work?

Discussion in 'The Projects Forum' started by fraterai, May 20, 2016.

1. ### fraterai Thread Starter New Member

May 20, 2016
1
0
I built this "AC Mains hum detector" circuit and the yellow bar is just a piece of copper wire (about 20cm). When you put the antenna near a mains ac source line the led lights up. Question is why does this work? What I am confused about is that the first transistor is obviously 'on' when the EM field interacts with the antenna, but I cant measure the voltage at the base without disrupting the thing and it turns off. Is it the electro-part inducing a voltage or the magnetic part making a current (or both?). The other thing that is bothering me is that the base is "floating" and i don't really have a way to describe this. Thanks for the help

2. ### wayneh Expert

Sep 9, 2010
15,002
5,454
IF you had a perfect meter, by which I mean one with infinite input impedance, you could measure the same signal that the copper wire is seeing. But your meter is not perfect and is likely a 1MΩ impedance, maybe 10M if you've paid more for it. Either way, it places a relatively large load on the signal being developed in the wire and it can no longer develop enough base-to-emitter voltage across the first transistor to turn it on. Nothing happens.

I believe the copper wire, a conductor, develops an emf in response to a fluctuating magnetic field.

3. ### #12 Expert

Nov 30, 2010
18,093
9,679
Imagine a series string of capacitance from the power wire to the antenna, to the detector, to you, to the floor. I calculate 5pf would be the amount of capacitance in that series string which would allow 316 nanoamps to be amplified enough to light the LED.

So there is your current path and gain calculation.

Sinus23 and nsaspook like this.
4. ### Marley Member

Apr 4, 2016
252
61
This works due to capacitance. A very small value of capacitance but sufficient due to the high impedance input. It is not really an "antenna".
Actually, the first transistor has no base bias so theoretically should not work. But leakage current is helping here.

Aug 23, 2012
6,422
995
6. ### wayneh Expert

Sep 9, 2010
15,002
5,454
Are you guys saying that an emf appears in the wire antenna in response to an electric field, not a magnetic one?

7. ### nsaspook AAC Fanatic!

Aug 27, 2009
4,929
5,367
Yes. It's a capacitance voltage divider. The 'antenna' is the middle plate inserted between the upper AC power wire plate and lower earth ground plate capacitor system similar to the body in the diagram.

8. ### DGElder Member

Apr 3, 2016
351
87
I agree this is e-field coupling but I wonder how close the probe and circuit needs to get to a 120V line for this to work? I would think the probe would have to be pretty close and not so much for the circuit. In particular because he is using a BJT instead of a FET first stage and the negative potential appears to be floating - with a battery source. Getting sufficient voltage potential across the BE junction and supplying the requisite base current seems tenuous and finicky to me.

Last edited: May 21, 2016
9. ### wayneh Expert

Sep 9, 2010
15,002
5,454
Is a regular antenna (AM radio, for instance) any different?

10. ### nsaspook AAC Fanatic!

Aug 27, 2009
4,929
5,367
Sure, a AM antenna is a far field EM energy device that normally converts one or the other far field radiating EM wave component (usually the magnetic field with a electrically large AM ferrite loopstick antenna design) into an electrical current for detection. The mains electric field detected with this 'hum' capacitive coupling device is a reactive near field that surrounds the wire but does not radiate (the actual amount of radiation is pretty close to zero) 'radio' waves' because the wire is much too electrically short at the 50/60hz power frequency.

http://www.edn.com/design/sensors/4...ds-and-antenna-radiation-takes-almost-no-math

Last edited: May 21, 2016
11. ### #12 Expert

Nov 30, 2010
18,093
9,679
The fact that you think it's tenuous and finicky, and the fact that this circuit works as advertised, are in contradiction.
Please adjust your beliefs.

The pick-up range is only a few inches. Leakage from the collector to the base contributes to the tendency to conduct.
Icbo is about 3 na at room temperature.
Vbe@316na is about 0.32 volts.

Yes, a j-fet would be more sensitive, but j-fets also use idle current. The bjt version will have better battery life.

File size:
57.5 KB
Views:
2
12. ### DGElder Member

Apr 3, 2016
351
87
"The fact that you think it's tenuous and finicky, and the fact that this circuit works as advertised, are in contradiction.
Please adjust your beliefs."

I didn't say it didn't "work" and I didn't see any advertised specs beyond: it works.

13. ### #12 Expert

Nov 30, 2010
18,093
9,679
That's OK. You can keep believing this circuit is suspect and I will keep believing that the legs of a horse are too thin to carry that much weight.