Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Moving from 74 series to 4000 series

Status
Not open for further replies.

Froskoy

New Member
Hi,

I've always used 74 series logic chips but my supplier has stopped supplying XOR gates in this series so I'm moving on to 4000 series as I don't want to switch supplier.

Do they behave in exactly the same way. I don't care about speed for what I am doing and I am aware of differences in supply voltage. I assume the pinouts are different but do you still have to do everything like putting a 1k resistor between inputs and ground?

I don't do a lot of electronics, even though I really enjoy it, because of time and this is the last project I am doing before I want to start playing with PIC and microcontrollers.

Thanks very much,

Froskoy.
 
Hi,

I've always used 74 series logic chips but my supplier has stopped supplying XOR gates in this series so I'm moving on to 4000 series as I don't want to switch supplier.

Do they behave in exactly the same way. I don't care about speed for what I am doing and I am aware of differences in supply voltage. I assume the pinouts are different but do you still have to do everything like putting a 1k resistor between inputs and ground?

I don't do a lot of electronics, even though I really enjoy it, because of time and this is the last project I am doing before I want to start playing with PIC and microcontrollers.

Thanks very much,

Froskoy.

hi,:)
Look over this link.



There are a number of 74 types available, 74HC 74HCT.

This is a handy reference.
http://www.interfacebus.com/voltage_threshold.html
 
Last edited:
An important difference is that you must never leave an unused CMOS gate floating as you can (but shouldn't) with TTL. A floating CMOS gate may go to a logic 1 or logic 0 or somewhere in between (which can significantly increase the current used by the IC).
 
The input impedance of the 4000 series is much higher than the TTL chips you have been using up to this point in time.
The 1k input resistors you are talking about can be increased to 10k, 100k or even 1M, depending on the length of wiring between the input line and the device connected to the input. For instance, if you have a switch on 30 metres of wire, the possibility of the wire picking up a stray signal is fairly high and a 10k to 47k resistor will be suitable.
If the circuit is closely connected to a previous stage, a 1M may be suitable. In normal circumstances, a resistor will not be needed if the input is connected to the output of another gate.
In general, you can consider the input impedance of CMOS is over 1,000 times greater than TTL. And accordingly, the chips take 1,000 times less current.
 
Thanks very much for the replies. ericgibbs I found your links very useful and thanks also for the advice crutschow.

@colin55: There's probably around 5-10cm between the gate and the ground. All my inputs come from the 5V line via a SPDT switch. So should I use a 1M resistor? Thanks very much. Your reply was really useful.
 
pardon my ignorance here, but moving from one ancient logic chip series to another seems analogous to moving from the stone wheel to the wood wheel living in the 20th (not even to mention the 21st) century; everyone else is using rubber tires.

did I miss something?
 
@ericgibbs: Thank you very much for your reply. Those links were very useful.

@crutschow: Thank you also for your reply - very interesting.

@colin55: Thank you very much for this. The distance between the chip and ground rail is around 10cm. The input is from the 5V rail via a SPDT switch. Should I use a 1M resistor? What would be the effect of sticking with a 1k instead of using a 1M? Thanks again very much.

@JustDIY: You have a very valid point there, but I should like to do this for the sake of doing it. As I stated in my original post, I wish to have a play with microcontrollers for my next project.
 
colin55: I'm talking about substituting old with old ... 74xx and 4xxx series are both older than I am, and I no longer qualify as a "spring chicken".

As Eric pointed out, the 74xx has been superseded by the 74HCXX, and many other types, F, VF, VHC, etc. The 4000 series is just as dead as the original 74xx series imho. Some of the more eccentric 4000 chips have been moved over to the 74HC series, check out TI's master logic parts catalog ... you'll have 74HC4xxx and such down at the end buried under a pile of dust.
 
justDIY You really don't know what you are talking about. You should leave your comments off the site as they are totally incorrect.
 
I would have to agree with JustDIY. 74xx / 40xx is handy for occasional glue logic but if you have a grip on basic logic I would move on to microcontrollers.
It's like having to learn to ride a horse before you can drive a car.
 
justDIY You really don't know what you are talking about. You should leave your comments off the site as they are totally incorrect.
I beg to differ. He seems to have a better handle on the situation than most. In fact I agree that using such ancient MSI logic chips is equivalent to using knives and stone axes.
 
Last edited:
Sure. Microcontrollers. Saw an ESR tester schematic just tonight and rather than using good old analog and/or digital chips, it used a PIC (and they didn't include any code listings). Just what I need. A circuit that I can't build because I have no PICs, no programmers, no software, etc. On the other hand, I have hundreds and hundreds of TTL, CMOS and analog chips less than ten feet away. Why not use the stone and wooden wheels I have on hand rather than make a trip to Moe, Manny and Jack's for steel wheels and rubber tires (and the equipment needed to mount them)?

Dean
 
I think the following graph illustrates DIY's point.

**broken link removed**
 
I've got hundreds of old TTL ICs in a drawer somewhere, it's a pretty dusty drawer. I also have a few rails of PICs that empty out now and then. As for the programmer well I do sell em but IMHO I would consider a programmer an essential tool for a digital designer.
Do you have a link to the PIC ESR tester? It's something I always wanted to play with.
 
I find all the comparisons of circuit design and logic implementation to stone tools to be more than absurd. Microcontrollers are only yet another tool available to solve design problems. They will never supplant, or make obsolete, all other methods and tools which have been developed, or those that will be developed. When I first began to develop custome ASICS during the early part of the decade, we were using a couple million gates, and a single processor, mostly for house keeping; setting up registers, etc. Some of the last projects I worked on used nearly 20 million gates and about 10 processors. Much of the processing had been offloaded to the processors by that time; however, logic has continued to grow exponentially on these products. Efforts to replace all or most of the logic design with processors is pretty much running out of steam. Processors just can't come close to logic when it comes to speed, parallel capabilities and low power consumption. Even after all these years since the death knell was rang for digital logic, we still do design the old fashioned way: with state machines, coders/decoders, fifos, etc.

On of the last designs I worked on was an update to DOCSIS. The "new and improved" design is processor and memory based. It was supposed to make the core more flexible. My contribution was to synthesize the core and optimize for speed and area. When the numbers came in, the new core was 3X larger than the old one, due to the large memory arrays that were required. The power numbers were even worse. These results nearly cost the project manager her job.

I routinely come across ridiculous suggestions of how to replace logic with microcontrollers. Seems some designers think they can't light an LED without a controller anymore. Beyond that, others will make absurd claims that already simple circuits will be simplified by 1000% with microcontrollers. Those claims are laughable, since it's clear that many think circuit design is just too damn hard to do now in the new age of controllers.

Not long ago, I had to invert and tri-state a signal on a "finished" product that had been a design flaw which got missed before it went to manufacturing. I ordered that a simple logic chip be deadbugged and connected to the original signal pin. I used the signal as the tri-state control as well because it was only important to have the signal on the trace when it was active. During the signal's inactive period, a different chip used the trace. Simple fix. Product shipped. No microcontroller needed. No code to develop, debugg and download. No reset to worry about. No watchdog timer to program, the simple logic always runs, never hangs.

Remember when analog design was supposed to become obsolete? That never happened. Neither will logic design ever be obsolete.
 
Last edited:
4000 CMOS is old, but it is also still very useful because of the very wide supply voltage range of 3v to 15v.
This can be particularly useful when mixing digital and analog circuitry, or when working direct from battery power.

Most of the more modern, and much faster, and higher performance CMOS is limited to a very narrow supply voltage range, which can sometimes be very limiting.
 
I find all the comparisons of circuit design and logic implementation to stone tools to be more than absurd. Microcontrollers are only yet another tool available to solve design problems. They will never supplant, or make obsolete, all other methods and tools which have been developed, or those that will be developed. When I first began to develop custome ASICS during the early part of the decade, we were using a couple million gates, and a single processor, mostly for house keeping; setting up registers, etc. Some of the last projects I worked on used nearly 20 million gates and about 10 processors. Much of the processing had been offloaded to the processors by that time; however, logic has continued to grow exponentially on these products. Efforts to replace all or most of the logic design with processors is pretty much running out of steam. Processors just can't come close to logic when it comes to speed, parallel capabilities and low power consumption. Even after all these years since the death knell was rang for digital logic, we still do design the old fashioned way: with state machines, coders/decoders, fifos, etc.

On of the last designs I worked on was an update to DOCSIS. The "new and improved" design is processor and memory based. It was supposed to make the core more flexible. My contribution was to synthesize the core and optimize for speed and area. When the numbers came in, the new core was 3X larger than the old one, due to the large memory arrays that were required. The power numbers were even worse. These results nearly cost the project manager her job.

I routinely come across ridiculous suggestions of how to replace logic with microcontrollers. Seems some designers think they can't light an LED without a controller anymore. Beyond that, others will make absurd claims that already simple circuits will be simplified by 1000% with microcontrollers. Those claims are laughable, since it's clear that many think circuit design is just too damn hard to do now in the new age of controllers.

Not long ago, I had to invert and tri-state a signal on a "finished" product that had been a design flaw which got missed before it went to manufacturing. I ordered that a simple logic chip be deadbugged and connected to the original signal pin. I used the signal as the tri-state control as well because it was only important to have the signal on the trace when it was active. During the signal's inactive period, a different chip used the trace. Simple fix. Product shipped. No microcontroller needed. No code to develop, debugg and download. No reset to worry about. No watchdog timer to program, the simple logic always runs, never hangs.

Remember when analog design was supposed to become obsolete? That never happened. Neither will logic design ever be obsolete.
Boy did you miss the point entirely. The tools and techniques of logic design were never the point of the discussion. The specific realization of a design with the original TTL parts (74xx) or the original 4000 series parts was the point. I would have no argument with a hobbyist making one of anything with any part he could lay his hands on. Trying to make a bunch of something with obsolete parts is, in my opinion, quite risky. Many of those parts are quickly becoming pure unobtainium not to mention gray market counterfit.
 
Last edited:
Boy did you miss the point entirely. The tools and techniques of logic design were never the point of the discussion. The specific realization of a design with the original TTL parts (74xx) or the original 4000 series parts was the point. I would have no argument with a hobbyist making one of anything with any part he could lay his hands on. Trying to make a bunch of something with obsolete parts is, in my opinion, quite risky. Many of those parts are quickly becoming pure unobtainium not to mention gray market counterfit.

Agreed, I think point was not that logic design is obsolete, rather certain logic families are.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top