Constructing Overflow Detection Circuit with Logic Works

[mathrocks]

My teacher assigned this problem for homework and I'm completely lost in what exactly I'm supposed to do. Btw "logic works" is a program we use to implement the circuits.

Using Logic Works, construct a circuit to detect overflow in 4-bit signed-2’s complement addition. Connect your circuit (in Logic Works) to the 4-bit adder included with your kit.

As discussed in class, overflow detection can be accomplished by comparing the sign of the addition inputs to the sign of the output. If the two inputs share the same sign and the output has a different sign, then overflow has occurred.

Hint: Can overflow occur if the two inputs have a different sign? No.

Use only parts available in your pencil box:
4-bit adder – 74_283
NANDs – 74_00
NOTs – 74_04

Further hint: Tie the carry input (C0) of the 4-bit adder to logic zero (or GND) lest you see a lot of unknown values in your output. Ignore the carry out bit (C4).

Test your circuit using a timing file for several different cases (at least 4), which clearly show examples of overflow and non-overflow. For each case explain what is happening in the circuit (i.e. “we are trying to add minus five to minus four and thus overflow because the circuit can only represent down to minus eight”).

I don't really understand what he wants me to do. Do I just build a circuit and the inputs are two 4 bit values and the adder adds the two binary numbers and somehow I detect to see if overflow occurred?
I don't understand what the output to the circuit would be? If overflow occurs what will happen? And how do I even detect overflow in my circuit in the first place.

I'm sorry if this is too detailed to answer, but I'm really confused about this whole thing.
Thanks a lot!

 

[NoTime]

Do I just build a circuit and the inputs are two 4 bit values and the adder adds the two binary numbers and somehow I detect to see if overflow occurred?

Thats what the task description says.

I don't understand what the output to the circuit would be?

The sum of two numbers.

If overflow occurs what will happen?

The sum will be wrong.

And how do I even detect overflow in my circuit in the first place.

The task description actually tells you. Try reading it again.

 

[mathrocks]

Thats what the task description says.

The sum of two numbers.

The sum will be wrong.

The task description actually tells you. Try reading it again.

I understand the concept of "overflow" but I don't understand how to actually implement it as a circuit. How do I make a circuit that understands how to compare the signs of the two inputs? That's where I'm getting stuck.

 

[mathrocks]

I'm going to attempt to make my question more clearer.

In order for overflow to occur the inputs must have the same signs or the last two carries are supposed to be the same. So in order to make this circuit I will need to compare the signs of the inputs since the adder I'm using ties all the internal carries together. Do I just look at the left most bit of the 4 digit value for both inputs and if they are equal then an overflow has occurred?

 

[don rigby]

wow, when you say sign do you mean "not vs true"? part numbers seem a little short can you post diagrams?
i am intrigued, and trying to learn something. i have never heard of (or don't recall) anything about 4 bit signed 2's compliment addition for overflow detection with the co (?) input tied to gnd. (is co the trigger?).
please provide the answers when taken up in class.
thanks don

 

[mathrocks]

wow, when you say sign do you mean "not vs true"? part numbers seem a little short can you post diagrams?
i am intrigued, and trying to learn something. i have never heard of (or don't recall) anything about 4 bit signed 2's compliment addition for overflow detection with the co (?) input tied to gnd. (is co the trigger?).
please provide the answers when taken up in class.
thanks don

http://www.ece.vt.edu/cel/ece2504/lab-manual-spring-2004.pdf

that's the link for the pdf with the part diagrams. The full adder chip is listed on page 9 while the other ones start on page 7. I believe C0 is the initial carry input.

And by sign I mean positive(0) value for negative(1) value. If both inputs are negative or both are positive then there will be a overflow I believe. I'm just confused on how to make the circuit reflect this.

Thanks!

 

[mathrocks]

http://engineering.dartmouth.edu/~engs031/hansen/outlines/21Arithmetic.pdf

Basically the circuit shown on page 8 is exactly what I need to compare the sign bits, but I'm having trouble understanding how they did this exactly. Like what do the AND and OR gates accomplish and if the output of the overflow is a 0 what does the zero mean?

Thank you

 

[NoTime]

Do you understand that the maximum absolute number value allowed is 7?