[sdiy] simple TTL programmable counter question (74LS163)

[Magnus Danielson]

From: Michael Buchstaller <buchi at takeonetech.de>
Subject: [sdiy] simple TTL programmable counter question (74LS163)
Date: Sun, 26 Oct 2003 17:36:38 +0100

> Hello friends,

Dear Michael,

> i could run up the wall and bang my head against it !
> Since 2 days now i am trying to make a simple programmable
> oscillator circuit to drive a wavetable in an EPROM with it.
> 
> The idea is: I take a high frequency oscillator (100 MHz in the final
> poject, 16 MHz in my bench project so i can see the signals on my
> Oscope).
> Tis is followed by 4 74LS163 (later 74F163) asynchromous counters,
> the lowest bit of the last one being used as an output.
> Free-running, this will divide the oscillator frequency by 65536, also
> give roughly 244 Hz.
> So far, this works.
> 
> But, i want to be able to program other frequencies.
> Now, i set the "programming" inputs of the 74LS163 (pins 3...6)
> via a bank of DIP switches and pullup resistors to some value.
> I want to take the Overflow (pin 15) on it´s falling edge to cause that
> the values from the DIP switches will be written to the ´163, so they
> will count up from that value instead run until 0 and over.
> (Say, i want 128 KHz, so i will have to divide the 16 MHz down by a factor
> of 125, so i will set the switches to 65536-126 = 65411. I would expect the
> circuit to overflow, then set back to 65411 and count upwards to the next
> overflow, and so on)
> 
> Looking at the last counter´s output, a signal is there at pin 15.
> I feed that into an 47121 Monoflop to stretch out the pulse width to some
> length and get a defined pulse from the low-going edge of the signal.
> This works, too. I get a fine 1 uS negative pulse every time the input
> overflow goes down to L state.
> 
> Now, i feed this pulse to all pin "9" of the 74LS163, but nothing happens.

Now, I took the oppertunity to bring out my Don Lancaster "TTL Cookbook".
I never really use it, but I got it (and it's CMOS equivalent) for old memories
sake. One of the things Don does is spend time on divide-by-N curcuits. He does
not use the '163s but he does treat the '161s (it is the synchronous or
asynchronous clear signals which differ, but for the moment we should hook it
to +5V anyway). He does basically what you do, but with the exception that the
carry out is sent through a inverter ('04) before feeding the LOAD inputs.

So, try toss your 47121 out of the loop (I don't see why it should be needed)
and simply replace it by a simple inverter.

Cheers,
Magnus