[sdiy] Funky flip-flops

[Scott Gravenhorst]

Karl Ekdahl <_nial_ at yahoo.com> wrote:
>I looked for shorts and outputs going directly to VCC
>or GND but didn't find anything. I thought that if any
>of the two boards i'm making (it will be 3 boards
>eventually) is having some kind of short-circuit or
>likewise, it should draw a tremendous amount of power
>so i measuerd the mA usage of each board. Apparently
>the 3 x 2 digit LED displays draw 30mA together, the
>first board around 220mA and the second 210mA, does
>this sounds reasonable? If so, i guess this baby will
>draw around 1A when i'm done, thus i can totally
>forget of making this one battery operated as my first
>thought was. Maybe i should just drop this and do it
>in PIC, though then i won't get rid of any 74 chips.

Or CMOS.  I too have a pile of old 74 series chips.  I just
can't see the current draw against the function they provide,
especially since I don't do anything beyond 2 MHz, most stuff
is audio rate.  So speed of the device isn't an issue.  CMOS
used as logic devices can easily run on batteries and since
they have a wide supply voltage range, usually 3 to 18 volts,
you can use almost any battery combination you like.

I used to work on DEC minicomputers as a customer engineer. 
These things had as many as 5 power supplies per drawer, each
supply was 5 volts at 25 to 35 amperes.  The VAX series, also
TTL, used 100 ampere 5 volt supplies.  So you can see that
TTL is expected to draw heavy current.

CMOS has a lower quiescent current as well as a lower current
during output state change.  CMOS still needs bypass, but
it's not as critical as TTL.

Regarding your oscilloscope angst blowing the supply fuse, I
would check the probe's resistance to ground to make sure you
don't have a short or shunt in the probe or input.  Or like
Harry said, maybe it's a 50 ohm input?  This sounds
outlandish to me, scope inputs are usually rather high input
impedance.  IMHO, a 50 ohm input should be clearly marked as
such.


>Karl
>
>
> --- harrybissell <harrybissell at prodigy.net> skrev: 
>> Whenever something happens with a scope it is a
>> clue.
>> I'd say there is no way the scope should knock out
>> your
>> power supply.  I'd suspect that you have the scope
>> connected
>> to ground (third pin if you're USA - are you ?)...
>> and when you
>> hook up the scope ground you are making an unwanted
>> connection ?
>> Or maybe the scope is mistakenly a 50ohm input (rare
>> but possible) ?
>> 
>> LEDs can eat amazing current... you can calculate
>> from the series
>> resistor, the power supply voltage and figure a 1.5V
>> drop in the
>> LED. Three seven segment displays could be more than
>> 200mA
>> alone.
>> 
>> I'd look into that scope thing carefully...
>> 
>> H^) harry
>> 
>> Karl Ekdahl wrote:
>> 
>> > So, i did add the 0.1uF decoupling caps and fired
>> the
>> > circuit, unfortuanley it behaved even more strange
>> > than before so i tried plugging in my scope
>> looking
>> > for noise in the powerlines. Even more unfortunate
>> is
>> > the fact that every time i plug in my scope (an
>> old
>> > Tek scope) the fuse in my powersupply brakes.
>> > Apparently, the 74xxx does draw a lot of current,
>> but
>> > is it really so that a couple of 15 74xx's + 3 LED
>> > displays should blow a 500mA fuse? It's becoming
>> clear
>> > to me that i have NO idea whatsoever how much
>> current
>> > things actually draw. And by the way, yes, the
>> 74xxx's
>> > are free - i've got shitloads and i want to get
>> rid of
>> > them so i decided to build an overcomplicated
>> > drummachine sequencer. I've also been thinking of
>> > building a *huge* pocket calculator only using
>> simple
>> > logic 74xxxx's.
>> >
>> > Karl
>> >
>> >  --- harrybissell <harrybissell at prodigy.net>
>> skrev:
>> > > Yes Jay...
>> > >
>> > > I'd like to add... decoupling caps. TTL is
>> > > desperately in need of good power supply
>> > > decoupling.   I'd suggest a .1uF cap across
>> EVERY
>> > > single TTL chip.   I follow this rule even
>> > > with 4000 series CMOS... which could use
>> probably
>> > > half that amount.
>> > >
>> > > About the RC on the ... clock line ??   I would
>> not
>> > > use a filter on the clock line... TTL needs
>> > > a good fast pulse to trigger well... although I
>> have
>> > > seen some times where a tiny tiny RC
>> > > helps in stopping a bounce (and mistrigger). 
>> Better
>> > > to fix the real problem than try to mask
>> > > it.
>> > >
>> > > A modest proposal... switch to 4000 series CMOS.
>>  It
>> > > is way, way, way easier to use than TTL,
>> > > uses a fraction of the power...
>> > >
>> > > ...unless the TTL stuff was free, of course...
>> > >
>> > > H^) harry
>> > >
>> > > jays at aracnet.com wrote:
>> > >
>> > > > Reset sounds funky.
>> > > >
>> > > > 1) Holding the reset lines low has the FF in a
>> > > constant state of reset. They are active low and
>> > > should be tied high (reset to +V) with a 1K-3K
>> ish
>> > > resistor. Or better yet put a cap in there
>> between
>> > > reset and ground (with the resistor in there) to
>> > > form a RC that slowly charges up to V+ providing
>> a
>> > > power on reset. Also if you aren't using the
>> resets
>> > > individually you should just tie them
>> altogether.
>> > > >
>> > > > If I were to guess I'd say there might be some
>> > > interaction between the PS and the 10K reset
>> > > resistors not providing a 'weak ground'.
>> > > >
>> > > > 2) If you tie stuff to ground don't use a
>> > > resistor.
>> > > >
>> > > > 3) Put some sort of RC on the switches going
>> to
>> > > CP.
>> > > >
>> > > > 4) Put a pull up (1K-2K) on the ~set lines.
>> It's
>> > > active low to.
>> > > >
>> > > > Jay S.
>> > > >
>> > > > Karl Ekdahl <_nial_ at yahoo.com> wrote :
>> > > >
>> > > > > Hi all, i've got a strange problem. I have 8
>> > > > > flip-flops (4 * 7474) which all share a
>> common
>> > > ~SET
>> > > > > line and each have it's ~RESET tied to
>> ground
>> > > trough a
>> > > > > 10k resistor. Each flip-flop is connected to
>> > > itself
>> > > > > via ~Q -> D and each is set/reset trough
>> a
>> > > push button
>> > > > > connected to CP. All Qs are connected to one
>> LED
>> > > each.
>> > > > > So, pushing the first button sets/resets the
>> > > first
>> > > > > flip-flop. Pushing the 8th button while the
>> > > first
>> > > > > flip-flop is set resets the FIRST one and
>> sets
>> > > the 8th
>> > > > > one. Pushing the 8th resets the 8th
>> flip-flop
>> > > and then
>> > > > > pressing the 1st sets BOTH flip-flops. All
>> the
>> > > > > flip-flops reacts like this with eachother
>> in
>> > > all
>> > > > > kinds of strange patterns. The only way they
>> > > possibly
>> > > > > COULD interact with eachother should as i
>> > > understand
>> > > > > it be
>> > > > >
>> > > > > a) the comon SET
>> > > > > b) trough the RESET resistors to ground (?)
>> > > > > c) trough the 74245 that delivers the
>> signals
>> > > from the
>> > > > > pushbuttons.
>> > > > > d) powersupply noise (?)
>> > > > >
>> > > > > a) could probably be ruled out since they
>> bot
>> > > SET and
>> > > > > RESET eachother. b) and d) is the same i
>> guess
>> > > and i
>> > > > > have a pretty huge powersupply feeding only
>> a
>> > > couple
>> > > > > of TTL chips, so that seems unlikley. I had
>> > > > > powersupply problems earlier but fixed that.
>> c)
>> > > the
>> > > > > 74245 seems to give perfect result while
>> looking
>> > > at it
>> > > > > with my multimeter, so i don't think so...
>> > > > >
>> > > > > I'm stranded, anyone?
>> > > > >
>> > > > > btw. i might as well thank everyone for
>> their
>> > > help
>> > > > > with various things the last couple of
>> weeks,
>> > > i'll put
>> > > > > some pictures and stuff of what i'm doing
>> > > somewhere in
>> > > > > some time..
>> > > > >
>> > > > > Karl
>> 
>=== message truncated === 
>

---------------------------------------------------------
- Where merit is not rewarded, excellence fades.
- Hydrogen is pointless without solar.
- What good are laws that only lawyers understand?
- The media's credibility should always be questioned.
- The only good terrorist is a dead terrorist.
- Governments do nothing well, save collect taxes.

-- Scott Gravenhorst | LegoManiac / Lego Trains / RIS 1.5
-- Linux Rex         | RedWebMail by RedStarWare
-- FatMan: home1.gte.net/res0658s/FatMan/
-- NonFatMan: home1.gte.net/res0658s/electronics/
-- Autodidactic Master of Arcane and Hidden Knowledge.