How does a flip-flop work? - unknown state

[alexmath]

I'm having trouble understanding how a simple flip flop circuit works. Firstly - can a logic gate work only connected with one wire?
I don't actually understand what happen's when you press a button inside the circuit when the circuit is in the unknown state. When the circuit is in a stable position, all is clear, but when we wire up the wires and we set it for the first time then it does not makes sense... I want to know what's happening with the flip flop in the picture below. Sorry for my poor english grammar. Thank you!

 

[.Scott]

The logic gates are connected through several wires.
Aside from the input and outputs that are shown, each one also has a power line and ground.

When the flip flop devices shown are first powered up, two possible stable states are possible. Either the Nand gate can be "on" or the And gate is on. in practice, any given flip flop will almost always power up in the same state every time. But when you design a circuit that uses a flip flop, you cannot depend on its state until you pulse one of its two inputs.

 

[CWatters]

I don't actually understand what happen's when you press a button inside the circuit when the circuit is in the unknown state.

The image you posted shows three circuits that illustrate two slightly different problems...

In the top circuit one input is not connected. That should never be done because the voltage on that input might float to any level. It might float to a voltage that is not a valid logic 1 or 0. In which case the output is said to be undefined. It could be a 1, 0, some value in between or even oscillate . You would have to look at the circuit inside the gate and treat it as an analogue device to know what the output would do. A bit of noise somewhere in the circuit might make it impossible to know for sure.

The other circuits illustrate a slightly different problem. Some time after the power rails are within spec the output of the flip flop will be found to be in one of two states. Which state it will be is undefined. However if you wait long enough it will eventually settle on a valid logic level even if you don't know which logic level that will be.

It's good practice to add another input to latches and flip flops to force the output to a known state on power up. This is usually connected to a signal called "Reset" or "Power ON Reset" which can be generated using an RC circuit like this...

http://www.renesas.com/media/support/faqs/faq_results/Q1000000-Q9999999/por1.gif

If you don't do this then you have to check that the output every latch or flip flop doesn't matter after a power failure. You wouldn't want something like a saw mill to start up on it's own after a brief power cut!

See also..

http://en.wikipedia.org/wiki/Power-on_reset

 

[donpacino]

the diagram you included represents a latch, not a flip flop. a flip flop is comprised of multiple latches. In the latch, typically one input is referred to as the reset line, and one is the set line. when using digital circuitry you ALLWAYS reset the circuit before using it. this gets rid of the unknown state problem.

 

[rezeew]

A flip-flop is a type of digital circuit that is commonly used in computer memory and other electronic devices. It is essentially a bi-stable circuit, meaning that it has two stable states: 0 and 1. The circuit maintains its current state until it receives a signal to change it.

The basic operation of a flip-flop involves two inputs: a "set" input and a "reset" input. When the set input is activated, the flip-flop changes its state to 1. When the reset input is activated, the flip-flop changes its state to 0. These inputs are typically controlled by external buttons or switches.

Now, let's address the issue of the unknown state. When a flip-flop is first powered on, it is in an unknown state. This means that it could be in either the 0 or 1 state, and there is no way to predict which one it will be in. This is due to the fact that the circuit is made up of electronic components, such as transistors, that have inherent variability in their behavior.

In order for a flip-flop to function properly, it needs to be initialized to a known state. This is typically done by using additional logic gates to control the set and reset inputs. For example, a simple flip-flop circuit might use an AND gate to control the set input and an OR gate to control the reset input. This ensures that the flip-flop is always in a known state when it is first powered on.

In summary, the operation of a flip-flop involves controlling the set and reset inputs to change its state, and it needs to be initialized to a known state in order to function properly. I hope this helps clarify the working of a flip-flop circuit for you.