CD4060 question

Hey guys I'm using a CD4060 CMOS ripple counter to save myself some programming on an arduino project. Do you know if it's safe to short some of the Q pins to get a certain short pulse that happen during one full cycle? I know the chip limits current internally to 15 mA. What do you think? Otherwise I will have to possibly make some smoke when my order comes in... but can't wait to test.

No.

To get the shorter pulses you should use gates (and / Or etc).

The older CMOS was a lot less robust than the newer parts types and will not take shorting very long.

John

Do you think you could be more specific and elaborate your proposed plan, in terms of exact decription of what you want to short and where ?

Do you know if it's safe to short some of the Q pins

What do you mean ? You mean short one Q pin to another ?

As a general rule shorting any output pins to anything is not a good idea.

Do you think you could start your post by explaining your objective instead of explaining your plan to
fix a problem that you haven't explained ?

What is your final objective ?
Have you read the CD4060 Datasheet ? (I shouldn't have to ask that but just to be safe...)

Yes I've read the data sheet... I do know this chip is pretty robust and internally controls pin current to about 15mA. Here's a simplified diagram

U want to short Q2 to Q3 so that Q1 will only give 1 (1/8th) pulse in one Q3 cycle instead of two... Does that make more sense?

If I put an LED anode on Q1 and short Q2 to Q3, it should work if it doesn't release smoke :slight_smile:

I will have to use a gate like proposed... shorting creates more issues for other pins and make the divider unstable.

So "shorting Q2 to Q3" is another way of saying "Feedback Q3 output to Q2" ?

How does this help ?

it actually does and that is what I'm trying achieve

I'm still going to try it when my chips come in... I have an abundance of them. :smiley:

If I put an LED anode on Q1 and short Q2 to Q3, it should work.

As you can see from your own diagram the various outputs are sometimes at 0 and sometimes at 1. If you short Q2 to Q3 what output will it be when Q2 is 0 and Q3 is 1? (Other than outputting smoke that is).

You're right it's the cathode to Q1. But it's a bad idea... but one I plan to test.

0
0
0
1

Nevermind I figured it out... don't need the gates or shorting pins... these dividers can get complicated! Very fun!

Nevermind I figured it out

I don't know if it has occurred to you yet, but there is a reason why Arduino is open source and why this is a forum. The whole point of posting on a forum that can be viewed almost anywhere in the world (except maybe North Korea) is to SHARE the information and knowledge.

Telling the whole world "nevermind, I figured it out..." isn't going to help anyone else trying to do what you are trying to do (which seems impossible to find out since you have still not told us your project or
what you are trying to do (exactly) and why.
How about elaborating on HOW you figured it out and WHAT (exactly) you "figured out" so the next guy
won't have to post for the same reason ?
Is that a reasonable request ?

You're right it's the cathode to Q1.

I'm not convinced you understood my question, or the point of my question. It had nothing to do with the LED and everything to do with the outputs.

Rasch,

The issue I was posting about what the idea of shorting the pins together to get a sort of feedback gate. That's a NO NO as per our discussion. I don't have the verbal acumen to even try to explain the silly patterns I'm trying to get with a 14 stage bit counter. It was a hypothetical. I'm trying to build some flashing units for various model train projects for a friend of mine. Instead of programming all the patterns I decided to use an old CMOS chip which I'm really starting to like.

The arduino will use it to call up certain patterns and it's still mostly in my head... I haven't decided, but I may use multiple chips for different patterns.

So a particular pattern really has no importance. It's the concept of using CMOS counters with the Arduino that's interesting. Kind of like using a shift register to control a display... I'm sorry the project is incomplete but I do value the input from guys more knowledgeable than I am...

To share somethings I've learned:

  • the 4060 has a oscillator built in so RC circuit can be used to control the oscillation
  • built in current limiter does not need resistors on LED unless using external power to run multiple LED's (I will need this for some of the project) using small logic level mosfet to power multiple LED's in series and or parallel depending on how many I need) again it's the concept not the final design that matters here.
  • extremely efficient - talking uA's
  • CD version tolerates 5-20v without regulator
  • the RC timer is kind of challenging to figure out
  • can be timed with crystal for super accurate timers (defeats the purpose to me might as well use 4040)
  • 14 bit timer is crazy amount of divisions!
  • and guys here know a lot more than I do about this :smiley:

some of this is really getting off the Arduino topic tho. Because it's sort of like a super 555 to me where I can replace Arduinos with this chip in some situations.

PerryBebbington:
I'm not convinced you understood my question, or the point of my question. It had nothing to do with the LED and everything to do with the outputs.

Perry, It's very possible I didn't understand your question. I posted this as a question if I shorted those out can I get a divided signal out without damaging the chip... some CMOS can handle this sort of abuse.

So theoretically if I shorted out a high with a low it should send 15 mA (internally limited) to the low and give me low - IN THEORY. If the CMOS can handle this current there should be no problem...

But it may also throw the counter off if it doesn't end up burning.

I may damage some chips testing the theory... just waiting for my packages to arrive. When I do the test I will share the results.

Instead of programming all the patterns

I'm not clear if you mean writing a program that generates the required pattern or...

I suggest that for each pattern you create an array with each element containing the desired pattern. For example

uint8_t pattern[] = {0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80};

If you step through the array one element at a time and feed it to 8 output pins it will produce a single running LED. You can devise whatever patterns you like and put them into an appropriate array.

Perry, It's very possible I didn't understand your question.

So theoretically if I shorted out a high with a low it should send 15 mA (internally limited) to the low and give me low

That's what I wanted you to think about. Connecting a high to a low doesn't give a low, it gives some indeterminate output. It might be a low, it might be high, it might be somewhere in between. It might change as the chip heats up.

If you really want to do something like that then for a low when one of the 2 pins is high connect a diode to each pin with the cathode to the pin and join the anodes together. The joined anodes will go low if either pin is low. For a high if one or both pins is high connect the anodes to the pins and join the cathodes together.

In effect that is what I'm doing with an LED.... no?

PerryBebbington:
That's what I wanted you to think about. Connecting a high to a low doesn't give a low, it gives some indeterminate output. It might be a low, it might be high, it might be somewhere in between. It might change as the chip heats up.

If you really want to do something like that then for a low when one of the 2 pins is high connect a diode to each pin with the cathode to the pin and join the anodes together. The joined anodes will go low if either pin is low. For a high if one or both pins is high connect the anodes to the pins and join the cathodes together.

hmmm I'll have to think about that Perry... some chips may be hurt testing this theory.

I don't quite understand how the chip regulates current. If it's limited to 15 mA... it may be enough to dissipate without an issue and it should drop pretty close to zero- and enough to go low on the LED.

on the second part
In effect that is what I'm doing with an LED.... no?

I don't quite understand how the chip regulates current.

Neither do I. I suspect it is in the on resistance of the output FETs, but that's only a guess.

In effect that is what I'm doing with an LED?

Is it? If so I've misunderstood your circuit. A diagram would help. My understanding was you were connecting 2 pins together with a wire link, not a diode or 2.

PerryBebbington:
I'm not clear if you mean writing a program that generates the required pattern or...

I suggest that for each pattern you create an array with each element containing the desired pattern. For example

uint8_t pattern[] = {0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80};

If you step through the array one element at a time and feed it to 8 output pins it will produce a single running LED. You can devise whatever patterns you like and put them into an appropriate array.

the whole point of this is that I can free up pins and memory by using the CMOS counter to flash the patterns. I'm really liking the logic chips... I have plenty of sketches that use patterns on the AVR. I believe this is old school but I'm enjoying the challenge of these chips. The chip is really kind of amazing. I remember being amazed at the binary clocks and now I understand them. You don't need an AVR or programming for everything and it's very efficient.

I have other projects in mind that can free up pins and add more blinking lights using CMOS logic... ones that had me baffled otherwise... charlie plexing is the same sort of thing...

It also gives the 8 bit chips more life as I can rely on other chips to do things that may otherwise eat up my pins.

I remember being amazed at the binary clocks and now I understand them.

Playing with random logic is a great way to understand the kind of things going on inside a processor. My introduction to electronics was building things with TTL and CMOS logic, including light controllers for discos.

PerryBebbington:
Playing with random logic is a great way to understand the kind of things going on inside a processor. My introduction to electronics was building things with TTL and CMOS logic, including light controllers for discos.

Perry,

That's pretty cool... I bet the sequencers and drum machines work the same way!... yeah I can now see where this type of device is being used... I can see the pulses being generated. I was always a little sleepy whenever we got to the logic gate stuff... wow it's amazing.

I have ideas about how to place little delays on them using small capacitors to throw up the timing. I've been thinking non stop about this since last night.