2

I've been trying to work this out for days and still can't do it. I have to convert the top equation to NAND only. I've worked out the second line by using Demorgans theorem however doing this would never convert the equation to NAND only.


My working:

$$\overline{ A\oplus(B+C) }\\\overline{\overline A \cdot(B+C)+A\left(\overline{B+C}\right)}$$

Ali Caglayan
  • 5,726

3 Answers3

1

You should of course use the identity:

$$X+Y=\overline{\overline{X}\cdot\overline{Y}}$$

And an overly complex NOT gate might also be useful:

$$\overline{X}=\overline{X\cdot X}$$

  • I've been in library for 3 days and still don't get what you mean. Could you elaborate. – user2437672 Nov 30 '13 at 16:48
  • @user2437672 First you get rid of all the OR parts of your equation using the first identity I gave you, then you use the second identity to convert NOTs to NANDs, and where your ANDs lack the NOT part you double-negate, and then use the second identity to make one of the NOTs into a NAND. – aaaaaaaaaaaa Nov 30 '13 at 16:58
1

We are basically dealing with a XNOR/coincidence logic gate, whose expression in terms of NAND gates can be found here (see picture to the right), with the observation that the “B” from the image corresponds to your $B+C$, which can be rewritten as $\overline{\bar B\cdot\bar C}$

XNOR (Coincidence) Build with NAND

In boolean algebra form, each NAND gate of inputs X and Y corresponds to $\overline{X\cdot Y}$, so the image above can be expressed as $\overline{\overline{\overline{\overline{A\cdot B'}\cdot A}\cdot\overline{\overline{A\cdot B'}\cdot B'}}\cdot1}$ , where B' is $B+C$. If you are not allowed to use $1$, then the expression becomes $\overline{\overline{\overline{\overline{A\cdot B'}\cdot A}\cdot\overline{\overline{A\cdot B'}\cdot B'}}\cdot\overline{\overline{\overline{A\cdot B'}\cdot A}\cdot\overline{\overline{A\cdot B'}\cdot B'}}}$ .

Lucian
  • 48,334
  • 2
  • 83
  • 154
1

The set $\{\uparrow\}$ is functionally complete. This means that the remaining logical connectives can be expressed in terms of alternative denial. It is easy to verify that $\phi \vee \psi \equiv (\phi \uparrow \phi) \uparrow (\psi \uparrow \psi)$ and $\neg \phi \equiv \phi \uparrow \phi$, from which the remaining connectives can be constructed via the usual equivalences.

emi
  • 1,640