I have a feeling I'm missing the obvious, something really, really obvious.

I need to use the CD4026 (datasheet) for a ÷60 and ÷12 counter. The datasheet says there is a way, but neglects to mention how.

CD4026B

When the DISPLAY ENABLE IN is low the seven decoded outputs are forced low regardless of the state of the counter. Activation of the display only when required results in significant power savings. This system also facilitates implimentation of display-character multiplexing.

The CARRY OUT and UNGATED "C SEGMENT" signals are not gated by the DISPLAY ENABLE and therefore are available continuously. This feature is a requirement in implimentation of certain divider functions such as divide-by-60 and divide-by-12.

Any thoughts?

 

The only connection That I see is that the "C" section is the only section that is used in all numbers except for # 2; that somehow 10+2=12, first stage goes around once with a carry, then advances to 2, giving an output = divide by 12 ?? How do we ignore the first C ?? Sorry- not much help.

 

I find it very unusual that I could find no circuits that use CD4026 in a digital clock, or any examples of how to use it to divide by 12 or 60.

 

I'm wondering if there is a sadistic sense of humor by a TI engineer way back when. Guess I'll use the segments.

Still, I have to wonder what the heck they were thinking with the "Ungated C segment".

 

I'm wondering if there is a sadistic sense of humor by a TI engineer way back when. Guess I'll use the segments.

Still, I have to wonder what the heck they were thinking with the "Ungated C segment".

I tried on LTspice to design and simulate a ÷12 using a couple of CD4017s, assuming that the only outputs available for use were 2 (same as complement of ungated c) and carry out. I was able to monitor all states by looking at the other outputs.
I never did figure out how to do it.

 

Datasheet for TI's CD4026B & CD4033B: http://focus.ti.com/lit/ds/symlink/cd4026b.pdf

At the top right on page 3 there's a timing diagram for the 4026. I'll agree that UNGATED "C SEGMENT" is a rather obscure name for the signal, to say the least.
If you wanted to reset the display after '2', then you'd do it on the rising edge of the ungated "C SEGMENT". If you wanted to reset the display after '6', you could look for the rising edge of the 'b' output; as the b output is low for displaying 5 and 6, but high otherwise - with the exception of if display enable goes low.

 

IMHO, this part was not designed to be used in a time clock, or if the designers were intending to include features to allow it to be used as such, they screwed up.
I have included a simulation using CD4017s that counts from 1 through 12 and then repeats. The CD4026 does not have outputs which allow this implementation. The CD4026 is a Johnson counter like a CD4017, but its outputs are decoded to drive a seven segment display.
Note that in my CD4017 implementation, the LSD reset state is called "1", because that's what a clock does. The CD4026 does not allow reassigning counter states like I did here, because the 7 segment decodes are hard-wired, with the reset state of the counter decodes as zero.
Bill, if you figure out how to use the CD4026 in a digital clock, please post your circuit!

 

Maybe look at this.

Digital watch

5 paragraphs above second picture.

 

IMHO, this part was not designed to be used in a time clock, or if the designers were intending to include features to allow it to be used as such, they screwed up.
I have included a simulation using CD4017s that counts from 1 through 12 and then repeats. The CD4026 does not have outputs which allow this implementation. The CD4026 is a Johnson counter like a CD4017, but its outputs are decoded to drive a seven segment display.
Note that in my CD4017 implementation, the LSD reset state is called "1", because that's what a clock does. The CD4026 does not allow reassigning counter states like I did here, because the 7 segment decodes are hard-wired, with the reset state of the counter decodes as zero.
Bill, if you figure out how to use the CD4026 in a digital clock, please post your circuit!

It is process in the Feedback and Suggestions forum, as a new experiment for the AAC book. It is very much a work in progress.

 

Maybe look at this.

Digital watch

5 paragraphs above second picture.

Thank you. That looks quite promising. Something to build on at least.

I will have to breadboard the parts I need, then I will try to figure out the logic. It is quite convoluted, but it looks like the TI guys knew what they were talking about.

 

I am currently recovering from a bad case of hoof in mouth disease.

 

Ron, I googled this for around two days with no luck. I still think it isn't obvious, however it looks like I'll get there. I have a good example of a 12 Hour counter, but I still would like to figure out a 24 Hour version. I've used about 6 feet of wire making short protoboard jumpers for the prototype 60 hour counter/display, with the prospect of using over 20 feet more.

I'll start with a 99 minute counter, then add the mod that makes it 60 hours. I'm still not clear what the type D FF does, but I will figure it out now I have an example in front of me. I wrote the author of the wrist watch article a thanks, he mentioned having different old data book that explained how it was done. Frankly that datasheet that seems to be all that is available on the web sucks.

Google kept referring me back to my own articles. This isn't the first time that happened, Google loves AAC. At least now I can help other people who may have similar problems

 

Hello Bill,

I have been looking around, but had no luck.
I have a copy of the RCA CMOS Integrated Circuits DATABOOK (1983), but the datasheet looks exactly the same as yours.

Bertus

 

OK, first I've done the simple 00-99 counter, it works.

I'm going to measure LED currents, I am still not confident of the lack of resistors. Seems to work OK so far though. After that add the 4013 to convert it to a 00-59 counter.

 

Bill, the D FF is used to reset each ÷60 after being at 59 for one count. He does this by delaying the carry out of the tens counter by one clock cycle using a D FF, which makes the output of the FF go high when the tens count gets to 6. At this time, the ones counter is at 9, so the combination instantly rolls over to 00 when the count hits 60.

 

Here's a video of a clock using 4026's as counters/display drivers where hours go to 24:
http://www.youtube.com/watch?v=SpUf5FIYv0w
just some quick notes:
2hz clock: 4060 -> 4013
Displays: all 4026
4073 IC's (triple 3-input AND).
Resets for 10s seconds & 10s minutes if 6 is reached.
If 24 hours is reached, hours digits reset

 

Here's one more.

URL here

At the very bottom.

Maybe a scheme for reset after 6 clock pulses.

Seen it a couple days ago. Forgot to include
in my previous post.

Maybe helpful, maybe not.

 

This guy is doing it the same way though, you can not use the segments to tell when 24 count has been reached since the LEDs are being driven directly by the 4026. I sent him a quick comment/question asking how he did it, but I don't really expect an answer, so I'll have figure it out.

Here is a revised schematic, though I think I like the NOR solution Pencil dug up. I'll try this, then I'll try the other.

 

Okay one more then I'll stop.

Last one

Doesn't seem to be using Pin 14 (Ungated C) or Pin 5 (Carry Out) for
the divide by 12 or 60. Maybe you can get something from it.

Hope I'm not being a pest.

 

Not at all. You have advance the project substantially, for which I thank you. I don't feel the need to reinvent the wheel in the slightest.