0

So, a few weeks ago I got into digital logic, and thought I'd buy some logic ICs. However, instead of doing actual research about which ones are good and which ones are bad, I just bought whichever had the most in stock on mouser.com, which just happened to be these:

  • CD4093BE, a NAND gate.
  • SN74HC86NE4, an XOR gate.
  • CD4071BE, an OR gate.
  • CD4081BE, an AND gate.

However, it seems that none of them work. Some of them just don't produce output at all, and some of them ignore the input completely and always output 1. They also seem to cause bad connections or something, as when using them the lightest nudge could cause it to stop working. I'm sure I'm using the right pins for input and output. I know that I'm in the range of correct voltages as well (using a 9V battery). I have noticed that all of them use "Schmitt triggers," but I'm not even sure what that means (again, I was stupid and didn't do my research).

try-catch-finally
  • 1,266
  • 3
  • 19
  • 37
undo_all
  • 11
  • 1

3 Answers3

3

Ok, let's start off with what you should have bought.

It's better (in my opinion) to start with the 74 series as they are less prone to static than the 4000 series. The HC subset is the "norm" nowadays unless you have specific requirements. So, some good ones to start with:

  • 74HC00 - Quad NAND gates (4 2-input NAND gates)
  • 74HC04 - Hex inverter (6 NOT gates)
  • 74HC08 - Quad AND gates (4 2-input AND gates)
  • 74HC32 - Quad OR gates (4 2-input OR gates)
  • 74HC73 - Dual J-K flip flop
  • 74HC79 - Dual D-type latch

Other useful ones to have around:

  • 74HC138 - 3-to-8 line decoder
  • 74HC245 - Octal bus transceiver with tri-state outputs
  • 74HC585 - 8-bit shift register with output latches

Now, why aren't yours working? Well, it is highly unlikely that they are all dead unless you are very unlucky, or were doing electronics with the cat sat on your lap. So it must be how you are wiring them up.

Since you don't show how you are wiring I will have to take a wild stab in the dark at it.

My guess is that there is 2 important things missing from your circuit (though this is only a guess):

  1. Decoupling capacitors. All \$V_{CC}\$ pins should have a 100nF capacitor to ground.
  2. Pullup / pulldown resistors on inputs - just wiring an input pin to \$V_{CC}\$ through a button will not work.
Ricardo
  • 6,164
  • 20
  • 53
  • 89
Majenko
  • 56,223
  • 9
  • 105
  • 189
  • Seriously, like Majenko says, EACH chip needs its OWN bypass capacitor, with the leads of the capacitor short length and close to the chip (within about 10mm-20mm). Ceramic 100nF (0.1uF). These integrated circuits switch faster than you may expect, and if you don't have a good low-inductance power and ground system, you won't get well-behaved digital signals. – MarkU Jul 05 '14 at 04:34
2

Those parts are all CMOS - which means they have very high impedance inputs.

All unused inputs MUST be connected to either Vcc (power) or to ground, otherwise they may float to a "maybe" state, and cause the part to draw excessive current.

Also, the used inputs must be connected to power or ground depending on the input level you want.

As mentioned in a comment the 74HC family has a maximum Vcc of 6 volts. The 4000 series should be OK with 12 volts or so - get the datasheets for the parts you are using to check on maximum voltage ratings, output current ratings, etc.

Peter Bennett
  • 59,212
  • 1
  • 49
  • 132
0

  1. When working with CMOS you must never let any inputs float. If they're not being used they must be tied either to the positive rail or to ground.

  2. For testing the functionality of the gates, you don't really need to be concerned with a particular voltage level other than to consider the positive rail a logical "1" and ground a logical "0". That means, for instance, if you want to see whether your 4081 is working, you make a truth table, like this:

     A  B  Y
---------
 0  0  0
 0  1  0
 1  0  0
 1  1  1

where "A" and "B" are the inputs to one of the gates and "Y" is the output of the same gate. Then you exercise the gate by connecting both A and B to 0V (ground) and examining the Y output, which should be 0V.

Next, you cycle the inputs through all their possible permutations, which should yield 0V on the output until you get to the one where A and B are both 1's, (9v) and the output should then go to 9V. If it all goes as planned, then that gate is good and you can go on to the next one.

Bear in mind that the gates that aren't being tested should have their inputs tied to either 9V or 0V.

The only chip you've mentioned with Schmitt trigger inputs is the 4093, and what the Schmitt trigger does is provide an input with hyseresis, which means that, say, if you have a chip working off of a 9V supply, the threshold for an input sensing a "1" on the way up to 9V might be at 5V while sensing a "0" on the way back down from having been at "1" might be at 4V.

Read the data sheets...

EM Fields
  • 17,527
  • 2
  • 19
  • 23