I built a simple amplifier circuit last night, which consisted of 3 BJT NPN transistors, 3 resistors, an LED, and some coils of wire. It's was made to act as a high sensitive touch circuit, this worked well, but then I decided to connect the base of the 1st transistor to a custom air core inductor... then I found that it increased its sensitivity a little, so I made a slightly larger one with few coils and a large cross-sectional area (roughly 2 inches in diameter made with regular 22 gauge insulated wire) this made it act as a close range proximity sensor of some type; when I placed my finger a few mm away the LED illuminated, but when I put my finger through the middle of it... it functioned even better, it worked with metal objects too. I eventually figured out that the larger the cross-sectional area, the more sensitive it was, but too small and you'll have to physically touch it lightly, and too large actually made it stay on without any presence. Was my finger emitting some kind of field that the inductor picked up, or was it some kind of capacitive properties that the coils of wire enhanced, or was the the inductor emmiting some electromagnetic field (although I don't see how it would, because it was an incomplete circuit hanging out like an antenna)? Can anyone explain exactly what this circuit was sensing? The inductor was hooked up just as shown in the schematic, and I was using 2N3904...or 06 (which ever is NPN, I can't remember) transistors. Also, I placed a piece of paper, then glass (actually a mirror) over it to see if it could sense through it, and it performed slightly better even.

 

Can anyone explain exactly what this circuit was sensing?

The only thing a circuit such as this is likely to sense is the ambient EMF emanating from the mains (60 Hz or 50 Hz, depending on your location. Bringing any part of your body near the circuit increases the amount of this signal coupled to it via capacitance, turning on the transistors and lighting your LED.

EDIT: There isn't any "inductive sensing" happening here; the only thing your inductor is doing is serving as a sort of antenna. You'd get the same effect with a length of wire or (even better, most likely) a metal plate.

 

That's a cute circuit! What comes after Darlington? Triplington?
This is easier to do with a j-fet and there are ICs designed for touch switch dimmers which have an internal oscillator which loads down when you touch the metal plate. Think, "Lutron" light dimmers in your local home repair store.

Not much to add to what OBWO said except we used to call this, "finger hum" which was used to test audio amplifiers. You are an antenna. You pick up power line frequency and you bring this energy close to the sensor coil. It's that simple.

 

...we used to call this, "finger hum" which was used to test audio amplifiers. You are an antenna. You pick up power line frequency and you bring this energy close to the sensor coil. It's that simple.

Yup.

Here it is, on a scope, several volts peak-to-peak. The scope probe is set on 10X (10 MΩ input impedance) and connected to a 3-foot length of wire just hanging loose:



Touching the wire roughly doubled the voltage. Turning on the room lights, all CFLs, adds a bunch of high-frequency "hash" (from the CFLs' switching regulators) onto the 60 Hz signal:

 

The only thing a circuit such as this is likely to sense is the ambient EMF emanating from the mains (60 Hz or 50 Hz, depending on your location. Bringing any part of your body near the circuit increases the amount of this signal coupled to it via capacitance, turning on the transistors and lighting your LED.

EDIT: There isn't any "inductive sensing" happening here; the only thing your inductor is doing is serving as a sort of antenna. You'd get the same effect with a length of wire or (even better, most likely) a metal plate.

But doesn't it have to pulsate to do that

 

Yup.

Here it is, on a scope, several volts peak-to-peak. The scope probe is set on 10X (10 MΩ input impedance) and connected to a 3-foot length of wire just hanging loose:

View attachment 136612

Touching the wire roughly doubled the voltage. Turning on the room lights, all CFLs, adds a bunch of high-frequency "hash" (from the CFLs' switching regulators) onto the 60 Hz signal:

View attachment 136614

Ok, thanks for that info

 

But doesn't it have to pulsate to do that

I don't understand what you mean by this.

 

I don't understand what you mean by this.

I mean there's no oscillating filter in that circuit, so I don't understand how it can continuously supply current to the LED, so shouldn't the LED technically be pulsating... or is it just so fast the human eye can't tell?

 

I mean there's no oscillating filter in that circuit, so I don't understand how it can continuously supply current to the LED, so shouldn't the LED technically be pulsating... or is it just so fast the human eye can't tell?

Correct. If you were to look at the voltage across the LED with an oscilloscope, you would see it rapidly turning on and off at the mains frequency. Whether 50 Hz or 60 Hz, that's above the human eye's flicker fusion frequency so we perceive the LED as being on continuously.

 

Correct. If you were to look at the voltage across the LED with an oscilloscope, you would see it rapidly turning on and off at the mains frequency. Whether 50 Hz or 60 Hz, that's above the human eye's flicker fusion frequency so we perceive the LED as being on continuously.

That makes perfect sense! What happens if that frequency is greater than the transistor's limit listed in the data sheet though?

 

That makes perfect sense! What happens if that frequency is greater than the transistor's limit listed in the data sheet though?

What will most likely happen at high frequencies is that if the electric field is sufficiently strong, the high frequency AC gets rectified into DC by the transistor's base-emitter junction, resulting in the LED being continuously on.

 

What will most likely happen at high frequencies is that if the electric field is sufficiently strong, the high frequency AC gets rectified into DC by the transistor's base-emitter junction, resulting in the LED being continuously on.

Ok cool, thanks for your kind replies.