[sdiy] 4514 w/4516

[Magnus Danielson]

From: Tom Arnold <xyzzy at sysabend.org>
Subject: Re: [sdiy] 4514 w/4516
Date: Sun, 16 Jan 2005 06:02:20 -0800
Message-ID: <20050116140219.GK41567 at moo.sysabend.org>

> On Sat, Jan 15, 2005 at 01:05:47PM -0800, Tom Arnold wrote:
> > I figured expanding this 16step sequencer board to 32 steps was going to be
> > as easy as chaining a second 4516 off the first 4516, but then my brain
> > finally kicked in yesterday and I realised that by chaining them they'll
> > count up to 8 bits as a pair unstead of one counting up 4 bits and then the
> > second counting up 4 bits.
> 
> So, I'm replying to myself again.
> 
> First with a rant.  Frys sucks.  I mean, this isnt a huge surprise to anyone
> but the real surprise is that they can indeed continue to get worse.  Its
> bad enough that they only carry NTE parts, whats worse is the aisle is
> slowly shrinking as they sell out of what they have.  I should offer them
> $100 for the entire aisle and see if they take it.
> 
> Anyway...  If you use a 4516 to slave from another, there is a small but
> noticable delay when it carries the bit.  The delay depends on the width of
> your clock pulse also which makes it more fun to deal with.  Survey says?
> 4514 aint good for this application.  I could work around the delay with
> some logic and in fact I started to, then I realise that much more quickly I
> could code up the previous mentioned scenix chip to act as a pair of 4514's
> and fairly cleanly wedge it into the existing circuit.  So thats what I've
> done still feeding a pair of 4514's.  Still some kinks to work out, but I'll
> nail them after some sleep...

The 4516 are synchronous counters, but you only experience the synchronicity
if you hook them up properly. To wire a pair of 4516 to do synchronous counting
you hook both to the same clock signal, at CP. You then hook TC of the lower
chip to the CE of the next chip, so only when the lower chip says "carry" will
the upper chip make a count in that clock-cycle. The UP/DOWN pins should be
wired together naturally as well as the PL pins and the MR pins, thus the
combo has common CP, UP/DOWN, PL and MR. The CMOS Cookbook (Don Lancaster)
has the CE pin named CI for Carry In and the TC pin named CO for Carry Out
with a complete little description similar to the above.

> It just feels like overkill to use a microcontroller for this, on the other
> hand it will save close to half a dozen bits of logic and associated wiring
> and board space, so I guess its a win in the end...

Microcontroller? Can't we avoid it and do it properly instead? >:o)

Cheers,
Magnus - fighting the microcontrolleroverbeleif decease on a regular basis