I have for the last two or three months now studied cpu design and have read several books on making simple processors that start at the gate level.

I now wish to put my knowledge into practice!

I noticed that 74ls series chips have high speed switching capabilities and so I ordered some quad 2 input and, or, and xor chips.

I decided as a first project to build an 8 bit adder with them.

But I now have a slight problem...

I can not get them to do what I want them to do with the exception of the quad 2 input xor chip.

I cannot get the outputs of the chips to source any current at all as I have measured the voltage with a multimeter and they just will not source current for anything.

However I have been able to work the logic in reverse by sinking the current.

The "xor" chip works fine and I have no problems with it however the "and" chip logic functions like an "or" gate instead of an "and" gate. The or chip also does not work right.

My power supply is 5v regulated and the chip numbers are the following:

74LS08 - OR
74LS32 - AND
74LS136 - XOR

Any suggestions as to what's going wrong??

 

Well, you should post a schematic so we can see what you are trying to do but in the interim, you should know that TTL of any species SINKS current way more than it sources. Its just the way it is. Look at the input characteristics. A '1' requires uAmps. A '0' requires sinking anything from 1.6mA (for old 74xx stuff) to lower, but still significant, values for LS, ALS and newer stuff. Since TTL is designed to drive other TTL, it does not need to source much to drive other inputs so.. it doesn't.

 

I cannot get the outputs of the chips to source any current at all as I have measured the voltage with a multimeter and they just will not source current for anything.

Then you are doing something wrong. You have to provide us with your full experimental set up. A photograph might help.

If you monitor the voltage at the output pin of a 74LS08 or 74LS32 gate you should be able to see that the output voltage corresponds to the boolean function of the gate.

 

Thanks John, I am still curious though as to why the "and" chips and the "or" chips are not performing their correct logic functions... Any ideas as to why that is? - I am currently working on schematics for this thing.

 

Do you have the correct pinouts?
Are you applying power and ground to the Vcc and GND pins?

 

Here is the Circuit - bottom rail is power - top rail is ground

 

Attach the picture as a .png file. Use go advanced, attachment, find the file, up load it.

 

I did have the pinouts right as the 7 pin was ground and the 14 pin was power. I read the data sheet for each of these chips which was what they required. I also have the power supply 5v regulated...

 

Sorry about the messy photo - top rail is ground- bottom rail is power

This is 5v regulated and I do have some experience with components as I have done this sort of thing in the past.

 

Some of my learned compadres here mighta-could deal with that but you should draw a sketch i.e. a schematic of what you have hooked up including power, ground and what you have connected to unused inputs. Draw the skiz, snap a photo and post it.

From your photo, its hard to see what's going on but I don't see power/gnd on pins 14/7 in any orientation?? Looks like your regulator is backwards as well??

EDIT: looking again, the logo on the breadboard is backwards.. is the picture flipped?

 

Here is the photo flipped. The power looks correct (assuming the DIPs are plugged in with the notch towards the right) but hard to see any logic being done...

But the part numbers are not visible so, how about a schematic?

As an aside.. its a good idea to orient your chips so that power is on top and ground is on the bottom like the databooks show the part and how the schematics read. Its just easier to see what's going on... Yours is backwards with power on bottom etc.. Makes it harder for you (and others reading your stuff) to read it.

 

Remove the LED and measure the voltage at the output of the gate with a DVM.

You cannot drive the LED directly as you have done.

In the next post we will show you how to drive an LED.

 

You do know that you have to ground the inputs of TTL circuits to give a logic low (?). If the inputs are left open they generally register as a logic high.

 

You do know that you have to ground the inputs of TTL circuits to give a logic low (?). If the inputs are left open they generally register as a logic high.

Good point. I forgot about that one too.

 

Here is the schematic.

 

Ok, so if I am sinking instead of sourcing the output, then wouldn't grounding the inputs technically be a logic high?

 

The 7432 is an OR gate but how you've wired it, all outputs should be '0' i.e. sinking current with the LED on.

The LED has no current limiting resistor so you may have blown the output of the chip and/or the LED.

Grounding inputs makes them a logic LOW by convention. 0-0 input to an OR gate gets '0' out, sinking current and turning the LED ON (after adding the resistor). Any input = 1 makes the output 1, turning the LED OFF.

 

Ok, I forgot to mention that I disconnected the led so you could see the wiring in the picture.(It looks connected but its not) The led works fine and I haven't blown anything... So if I wanted to have a logic 1 on any of the inputs (and assuming both of the inputs are grounded) then i just simply disconnect it from ground to make it high or should I apply power?

 

So if I wanted to have a logic 1 on any of the inputs (and assuming both of the inputs are grounded) then i just simply disconnect it from ground to make it high or should I apply power?

Don't leave the inputs floating i.e. not connected. Use a pullup resistor from 5V to the pin. Then you can ground the pin for a logic 0 and remove the ground for a logic 1, the resistor will then pull the input up to 5V, a good logic level. Unconnected TTL inputs will indeed float themselves to a logic 1 but its not a good idea for many reasons, noise susceptibility for openers. Since the logic 1 input current is small, you can use a reasonably big resistor - 10K - 68K or such.. The bigger the R, the less current needs to be sunk to pull a logic low..

Many designs connect unused inputs directly to power and that's an option but its recommended to collect all unused inputs and pull them up through a common resistor.

As an exercise, measure the voltage on an unconnected input then compare it to the minimum logic 1 level in the datasheet. You won't find it to be a very solid 1 vs pulling it up or driving the input from another TTL output.

The led works fine and I haven't blown anything...

You still need a current limiting resistor. You are asking the TTL output to provide current limiting, something it was not designed to do.

 

If you want a logic 0 input, connect the input to GND.
If you want a logic 1 input, connect the input to Vcc.

(Edit: or do as JohnInTX says, with a 1kΩ pullup resistor)

As indicated earlier, you cannot use the LED directly to monitor the output state as you have shown.

Use this circuit to monitor inputs or outputs: