# 74HC390 question: Choosing input and output pins

#### Alan3

Joined Apr 18, 2017
10
Complete beginner of 76 new here, so hello all.

I have been figuring out how a 74HC390 works with its two sections allowing frequency division by 2 or 5 (each having two sections) so I can have division by 2, 4, 5, 10, 20, 25, 50, 100. In all cases there seem to be alternative ways of doing this. I would like to know how the initial input and subsequent overall output pins are chosen. I realise that using several flip flops is a matter of strapping one output to another input, but it is the choice of order that puzzles me. Once I've found someone who can explain how this is done, I'll post more detail if required.

#### MrChips

Joined Oct 2, 2009
21,326
74HC390 is a dual decade ripple counter.
This means that there are two stages of divide-by-2 and divide-by-5 counters.
Let's do the math.
You have a total of division by 2, 2, 5, 5.
How many different divisions can you achieve with these four divisors?

Here is what I come up with:

1 - inherent, i.e. bypass the counter
2
4
5
10
20
25
50
100

What order do you cascade the stages? It really doesn't matter so long as you arrive at the correct divisor.
If you are going to have a switch selectable option of only two, three or four of the possible options, then choosing the right order could make or break your design. Hence choose wisely.

There are two cases where the order does matter.
1) It might be the case that one stage (e.g. divide by 2) has a higher max frequency than the other stage. Hence if one wishes to operate at this frequency, this would be chosen as the first stage. As far as I can see in the datasheet, this is not the case.

2) If it is important to have a symmetrical waveform (i.e. 50% duty cycle), then one would use a divide-by-2 as the final stage.

#### AnalogKid

Joined Aug 1, 2013
8,487
Working from memory ... Going back to the original 7490, there is a difference in the two possible connections. The way the chip pinouts are arranged, it is an easier board layout to divide by 2, then by 5. In that case the output waveform is not symmetrical. If you divide by 5 first, then by 2, the output has a 50% duty cycle. Whether or not that matters depends on the downstream circuit.

ak

#### Alan3

Joined Apr 18, 2017
10
Working from memory ... Going back to the original 7490, there is a difference in the two possible connections. The way the chip pinouts are arranged, it is an easier board layout to divide by 2, then by 5. In that case the output waveform is not symmetrical. If you divide by 5 first, then by 2, the output has a 50% duty cycle. Whether or not that matters depends on the downstream circuit.

ak
Thanks both. I'd interested to see further inputs (no pun intended). I have seen two descriptions of the pins for divide by 100 which are the same, but do not correspond with the pinouts I worked out myself, so I thought there must be some rule. Later today I'll describe more fully...

#### Alan3

Joined Apr 18, 2017
10
OK. Coffee at the ready. The two pinouts that agree (from different sources) for divide by 100 are: Pin 12 in (2CP1), strap pin 9 (2Q3) to pin 15 (2CP0), strap pin 13 (2Q0) to pin 4 (1CP1), strap pin 7 (1Q3) to pin 1 (1CP0), output pin 3 (1Q0). Of course both MRs are to ground, as is pin 8, and pin 16 is vcc. All this is logical and I understand how the output of each flip flop is strapped to the input of the next. Also the sequence is /5, /2, /5, /2 (the3 divide by 2 is last).

But from principles which I thought I understood, and still going /5, /2, /5, /2, I made it pin 4 in, strap pins 7 to 1, strap pins 3 to 12, strap pins 9 to 15. pin 13 out. My versions appear to be OK but the other two (which both agree) also are right. What have I missed? Also, if /2 is required for 50% duty cycle, what about the /5 and /25 versions?

And another question: Why are nQ1 and nQ2 not used but nQ3 is?. And final question (for now): Why is 2CP0 chosen before 1CP0?

Last edited:

#### dl324

Joined Mar 30, 2015
10,946
OK. Coffee at the ready. The two pinouts that agree (from different sources) for divide by 100 are: Pin 12 in (2CP1), strap pin 9 (2Q3) to pin 15 (2CP0), strap pin 13 (2Q0) to pin 4 (1CP1), strap pin 7 (1Q3) to pin 1 (1CP0), output pin 3 (1Q0). Of course both MRs are to ground, as is pin 8, and pin 16 is vcc. All this is logical and I understand how the output of each flip flop is strapped to the input of the next. Also the sequence is /5, /2, /5, /2 (the3 divide by 2 is last).

But from principles which I thought I understood, and still going /5, /2, /5, /2, I made it pin 4 in, strap pins 7 to 1, strap pins 3 to 12, strap pins 9 to 15. pin 13 out. My versions appear to be OK but the other two (which both agree) also are right. What have I missed? Also, if /2 is required for 50% duty cycle, what about the /5 and /25 versions?

And another question: Why are nQ1 and nQ2 not used but nQ3 is?. And final question (for now): Why is 2CP0 chosen before 1CP0?
You're new, so someone needs to tell you this.

When trying to talk about circuits, you need to provide a schematic.

#### ScottWang

Joined Aug 23, 2012
6,912
You just missed the 8(1Q3) and more when it is 1000H(high low low low).
When you can using 8 then you can assemble more combinations.
The basic elements will be like these:
Two independence counter.
1. -- 1, 2
2. -- 1,2,4,5

Two counters in series.
3. -- 1,2,4,5,8

So when you using all 74HC390 then how many combinations you can get?

#### Alan3

Joined Apr 18, 2017
10
You're new, so someone needs to tell you this.

When trying to talk about circuits, you need to provide a schematic.
Thanks Dennis, but there is no circuit. I assume (from the answers given) that the chip is well known

#### dl324

Joined Mar 30, 2015
10,946
Thanks Dennis, but there is no circuit.
The connections you described is the circuit.
I assume (from the answers given) that the chip is well known
It's a well known chip, but we can visualize the circuit better with a schematic.

This is what you described. The circuits are equivalent.

#### AnalogKid

Joined Aug 1, 2013
8,487
Paraphrasing Rear Admiral Joshua Painter in The Hunt For Red October (1990):

An Engineer don't take a dump, son, without a schematic.

ak

#### dl324

Joined Mar 30, 2015
10,946
Paraphrasing Rear Admiral Joshua Painter in The Hunt For Red October (1990):

An Engineer don't take a dump, son, without a schematic.
Many of us are familiar with bi-quinary counters and their application, but few of us would waste brain cells trying to memorize pinouts...

Also few would bother reading descriptions of how things are connected. It's not worth the effort.

The only reason I drew the schematic that was described in words was to understand what the OP was talking about.

#### Alan3

Joined Apr 18, 2017
10
I was trying to get an insight into why a particular flip-flop was called first, and some rules of doing it. Looks like the answer is actually 'Don't know'!

#### dl324

Joined Mar 30, 2015
10,946
Looks like the answer is actually 'Don't know'!
It's more like "don't care".

#### Alan3

Joined Apr 18, 2017
10
What am I, a well-meaning human, to make of that reply from a 'distinguished member'? Unless of course you are telling me that it does not matter which order I do the division in... which doesn't seem to accord with the examples I have seen

#### dl324

Joined Mar 30, 2015
10,946
What am I, a well-meaning human, to make of that reply from a 'distinguished member'? Unless of course you are telling me that it does not matter which order I do the division in... which doesn't seem to accord with the examples I have seen
It's the latter.

It's not that we "don't know", it's that we "don't care" what order they need to be connected in to get two stages of divide by 10.

"Don't care" is a common phrase in electronics for situations where we don't care about something. "Don't know" isn't, and could be taken as an insult when commenting on someone's knowledge; or lack there of.

#### Alan3

Joined Apr 18, 2017
10
It's the latter.

It's not that we "don't know", it's that we "don't care" what order they need to be connected in to get two stages of divide by 10.

"Don't care" is a common phrase in electronics for situations where we don't care about something. "Don't know" isn't, and could be taken as an insult when commenting on someone's knowledge; or lack there of.
OK. got that. But when I saw two independent examples of pin connections for divide by 100, but mine wa different, doesn't it indicate that they were following some rule which is at present unknown to me? I'd sure like to know that rule!

#### dl324

Joined Mar 30, 2015
10,946
doesn't it indicate that they were following some rule which is at present unknown to me?
There is no rule. Personally, I would have wired it as /2 /5 instead of /5 /2 because it's easier to wire the former; as @AnalogKid mentioned in post #2.

As I said in post #9, the two circuits you described are equivalent. In electronics, as with many things, there are many ways to accomplish the same thing.

#### Alan3

Joined Apr 18, 2017
10
There is no rule. Personally, I would have wired it as /2 /5 instead of /5 /2 because it's easier to wire the former; as @AnalogKid mentioned in post #2.

As I said in post #9, the two circuits you described are equivalent. In electronics, as with many things, there are many ways to accomplish the same thing.
Thanks, and thanks for putting up with me. I am getting an entry level DSO for my 77th birthday so will be interested to see all this in action.

#### dl324

Joined Mar 30, 2015
10,946
I am getting an entry level DSO
There are some recent posts seeking advice regarding which scope to get.

Personally, I prefer Tektronix analog scopes from the 70's. I have a couple different DSO's, but I rarely need one.

#### AnalogKid

Joined Aug 1, 2013
8,487
Looks like the answer is actually 'Don't know'!
No. We do know, and the consequences of the different connections are spelled out in post #3.

ak