Transistor HELP!!

[BManriquez]

I need some advice on a simple circuit I am trying to construct using one PNP transistor (2N3906)and one NPN transistor (2N3904). The attached image shows the circuit I am working on (top image - 1). I am trying to make a pendulum circuit that uses a magnet that passes over a coil of wire. When the magnet passes over the coil (about 10 ohms), an EMF spike activates the base of the NPN transistor switching it on and, in turn, switching on the PNP transistor which sends full voltage to the coil which pushes the magnet through its swing. Once the magnet is away from the coil, the drop in voltage is supposed to turn off both transistors until the magnet returns. However, when I connect the 3V supply, the transistors are always on and there is always voltage on the LED and coil. Thinking the transistors were bad, I changed them out several times, but I keep getting the same results. I even went as far as constructing another circuit (bottom image - 2), and again, the LED is always lit regardless of the resistance on the base of the NPN. I even added a 10 Meg resistor to the base of the NPN and measured a voltage of .56V on the base and yet, the LED is still lit! What the heck?? I though .7V was needed to send a transistor into saturation. What am I doing wrong? Just so I know, when both transistors are on, should the base of the PNP measure zero, because as it stands now, I'm measuring something closer to 2.35V and the LED is still fully lit. Also, I'm having a problem with the coil whining when the magnet passes over it - any suggestions on how to eliminate the coil noise?

**broken link removed**

 

[crutschow]

The image did not. Di you use the "Insert Image" icon at the top of your post window?

 

[BManriquez]

Yes, but it doesn't do anything. The window pops up, but when I hit upload, the window simply disappears.

 

[Grossel]

Hi.

Just upload your image at tinypic dot com or similar services.

 

[crutschow]

What coil?

The circuit is always on because the NPN is always biased on (what is the purpose of the diode?). Even 10meg is enough to turn it on sufficient to turn on the PNP. You measured a lower voltage at the base of the NPN because the input impedance of your meter is also likely 10meg.

The voltage at the base of the PNP will never be lower than about 0.7V below it's emitter voltage unless it is zapped.

 

[BManriquez]

The second circuit is a simplified version of the first (top circuit). The coil in that circuit was always on, so I built a second circuit (the bottom one) using an LED instead of the coil as I tried to figure out where the error was.