7400 Chip

That's OK. I'm just amazed that you can actually FIND a 7400 around.

slightly confused as to the operation of the 7400 quad 2-input NAND gates chip. I need to connect pin 7 to 0V, but does pin 14 (Vcc) need to be supplied with 5V? And then the inputs to the NANDs should be supplied with

5V for high I presume... And for low do they need to be connected to 0V, rather than left disconnected?

OK.

In TTL logic, Vcc is 5 volts.

Usually, anything under around 0.8 volt was considered a LOW and anything over 1.2 volts was considered high.

In practice, LOW was on the order of 0.2 volt and high was from 3 to 4 volts. With OC (open collector) stuff, the high could well be 5 volts.

with chips in the flesh you realise how little you know.

Have fun!

You should know, however, that even in the early 70s, 7400 (quad NAND) was just "glue" that was used to tie stuff like latches and counters together.

But there were some fun tricks with 7400 stuff:

Tie the output to one of the inputs and when the other input goes from low to high, the output is just a narrow spike.

Take two NAND gates and take the outputs of each one to one input of the other and you have your basis R-S flip/flop. Sometimes used to "debounce" push button switches.

With a little controlled feedback, you can turn 7400 gates into amplifiers or oscillators. As the old joke says: when the only tool you have is a hammer all problems look like nails.

If you get to using counters and such you may find out the hard way that decoding a value from a counter can generate glitches.

Let us know what you are doing. Several of us would be glad to help just for "old times sake."

You are quite welcome.

trying to build a circuit I've designed with NAND gates, but am slightly confused as to the operation of the 7400 quad 2-input NAND gates chip. I need to connect pin 7 to 0V, but does pin 14 (Vcc) need to be supplied with 5V? And then the inputs to the NANDs should be supplied with 5V for high I presume... And for low do they need to be connected to 0V, rather than left disconnected?

chips in the flesh you realise how little you know.

John gave you some good stuff in his post - be sure to note it. Most often you can find stuff on line with Google, by putting the part number and the word "datasheet" (without the " marks) in the Google search box. Here's one helpful site that search found:

Enjoy!

Ed

I have a few drawers full of 74xx parts (I think), complete with '73ish date codes. ;-)

At the receiver VL 2.4V. At the driver, I believe it's .4V and 2.8V. An input of 1.2V is right in the threshold (two Vbe above gnd).

Yes, the no-load Voh was about two diode drops below Vcc (3.6ish volts).

Thanks John.

What I'm actually up to at the moment is stepping through a 'Principles of Computer Hardware' book, primarily for coursework but secondarily for my own interest (hence my slight deviation into actually acquiring some chips and constructing something).

I was quite surprised to find the 7400s available on the Maplin site. They seem to have a good selection. At the moment, I've planned my BCD to 7 seg display decoder, and with your help I've been trying to put it together on a breadboard. I've since decided that I'd better get a PCB made. I've been looking at various software solutions for designing and printing the mask, and was hoping to find one that would a) support 7400s specifically and b) take a circuit designed within the program and auto-route everything. Since such a program doesn't seem forthcoming, I'm beginning to resign myself to having to do all the routing by hand.

The book I'm stepping through took the basic gates first, and then moves on to sequential logic, so I'm currently reading about RS', JKs and suchlike. Just flicking through, it seems as though I've got all the counter stuff and suchlike coming up in not too long.

I hadn't read about routing one of the inputs of a NAND to the output. I'll give that a little go now.

I'll post again here when I finally get a PCB made, which will probably take a few weeks, as I won't have access to the facilities until I'm back at school.

Thanks for your help and info.

--------------= Posted using GrabIt =----------------

------= Binary Usenet downloading made easy =---------

-= Get GrabIt for free from

=-

take a few weeks, as I won't have access to the facilities until I'm back at school.

Why bother with a PCB?

You might try using hot glue to place the ICs with leads up on a metal plate (pie pan?). Then just solder wires from point to point. If you "ground" pin 7 right at each IC you will be better off from a noise view than then a single layer PCB.

That way you can have fun right away.

trying to build a circuit >I've designed with NAND gates, but am slightly confused as to the operation of the 7400 quad 2-input >NAND gates chip.

I worked with these things in 1966 ~ 1970 designing ICL 1900 mainframes. The 7400 were £5 a go in very large quantities then, but it was a super technology compared with silicon transistor boards. CMOS might be my choice nowadays I think. There were two schools of thought on some of your questions: that of the people using them to 'make things' and that of the circuit design/quality team who were more formal.

yes

yes. We also put a 0.1 microfarad capacitor between 5v and ground for every 3 or 4 chips to keep the supply clean.

Well, our circuit design boys didn't like that because if the supply voltage went a bit high they claimed it would fry them. We had a dropper resistor pair giving about 2.4v I think. However, lots of designs get away with connecting directly to the power rail. I later used one of our computers for 10 years - in that time the only electronic failure in the processor was one of those dropper resistor chips!

Oh yes, definitely, but remember that will force the output high irrespective of the other input. If you leave them open circuit they float up, giving a high. We could rely on them to stay this way and just follow the other input, but the circuit boys said it would slow them down a bit. With clock cycle times of 300ns (wow!) we could not afford that.

With 7400 NAND (actual 7400, not the family!) though, if you really want an unused input (ie you are using it as an inverter) you can connect the two inputs together - no need to connect to high or low.

Voltage limits on these things are 5v + or - 0.25v, but I was never happy with a system unless I weeded out anything which didn't work down to the lower limit of the supply at 4.2v.

yes, it's a steep learning curve.

The devices were almost indestructible. No static problems; shorts in the wiring to ground, power, each other never seemed to harm them. That's until a colleague poked about in the wiring with a main hand lamp and managed to connect +24volts to a reset line visiting lots of places!!

Oh the nostalgia!

Phil

I agree with Phil. I like CMOS better than TTL. The power requirements are a bit less demanding too.

My spa controller is all 4000 CMOS and SSRs. It has survived hurricanes, thunderstorms and a fire. I don't know what would break it. I suggest wirewrap sockets for your prototype. Changes are easier.