CD4013 Vdd?

Question, the CMOS 4013 flip-flop datasheet says Vdd can be -0.5 to
+18v, and recommended operating range is from 3.0v to 15v. But will it
operate at 2.0v? Speed is not an issue.

http://www.megatronics.lk/data/4000/hcf4013.pdf

-M

 

Probably. All you can do is try it. My experience is that CMOS will
usually operate right down to a supply equal to two thresholds, or
about 1.5V. It _will_ be _slow_.

...Jim Thompson

 

It's not gauranteed. The small print below the absolute maximum ratings
says -

This is a stress rating only and functional operation of the device at these
or any other conditions above those indicated in the operational sections of
this specification is not implied.

 

Don't expect any two to operate similarly, though you might find some
that work. I suggest you start with a chip based on a lower voltage
process like the HC, HCT, ACT, AHC family.

 

That's the ticket, just get a 74ALVC74, although I'm sure that the
74HC74 design I reworked for ON Semi will work at 2V.

...Jim Thompson

 

The only problem being that if you get it into production and it doesn't
work the way your samples did the manufacturer will say "tough s---,
you're running it out of spec".

 

Of course. However I doubt that anyone inquiring about a 4013 could
have production in mind ;-)

...Jim Thompson

 

Something from my memory raises a danger flag.

The CD4013 uses "analog switches" to route the signals to create its
flip-flops.

 

It makes a good device for rememebering whether the "power is on" or not
in a system that leaves some of its logic always powered. The CD4XXX
stuff draws almost nothing. I know of at least one product with a
membrane keypad that powers on by flipping a CD4013 flip-flop.

Someday I hope some bright company sees this sort of application and makes
a super low power, wide VCC range highish density CPLD or FPGA.

BTW: What I'm designing now draws lots of power so right now, I don't
need that.

 

No-one else said it so "just use a PIC". They work down to 2.0V at slow
clock speeds.

If you care about the low static Idd, most HC parts are OK in that regard
but you have to spec. the suppliers because others aren't.

 

Yep. But they be P- and N-types in parallel, thus my 2*VT rule still
applies.

...Jim Thompson

 

Yes, now that I've though about it I remember that the analog switches are
driven by inverters and drive inverters. This makes it no worse than a
NAND or NOR driving another gate. There are two devices (N or P) in
series doing the ull UP or DOWN.

 

The 4013 _is_ a MOSFET design, so there's no free lunch there. There
are 3.3V "Logic" MOSFETs out there, so there's hope.

Use a P-channel MOSFET to turn it on, with an NPN "logic" transistor
from the PIC to keep it on. Connect the gate of the MOSFET to the
transistor collector and the switch. Make sure to scan the state of the
switch on power up, or delay for a bit when you turn on -- I've had
power supply decoupling caps bounce up in voltage when the current drain
is removed, turning the uP back on.

 

Dunno, but you can get the consumption of a 16xxx down to around 10-15uA
at 32kHz on 3v without trying too hard simply by disabling all
unnecessary bits - and much less with cunning use of sleep mode.

Regards,

 

How about this to get more gate drive?:

Buffer
!\
-----! >---+--------->!-------+------+------------ Mosfetgate
!/ / ! !
\ --- /
/ --- \
! !\ ! /
+----! >------------ !
! !/ GND

 

Neat idea that!.
Like the way it can extend as a ladder form.
regards
john

 

Especially with clocked flip-flops going into a region where
rise and fall times are glacially slow is a bad thing.

It's not so bad if you can do your function without a clock line
(a set-reset flip flop).

There are D-type flip-flops that work fine under 2V; the 74AUC74
for example (if you know anywhere to buy it...) has versions that
are supposed to go down to 0.8V.

Tim.

 

If possible, feed those slow rising/falling signals into a
schmitt trigger and use the output of that to clock your
flip-flop. You can use a logic gate that already has a
schmitt trigger on the input, or you can make one with a
comparator or opamp using positive feedback.

 

Hi Mark,

It looks like you are the OP (can't see the whole thread). Have you
considered the 74HC series? It is meant to operate between 2V and 6V so
it may be a nice fit here. They are much cheaper than newer very low
voltage logic and sometimes cost even less than CD4000 chips. Also, I'd
look into a Schmitt device here.

Regards, Joerg

 

Or a logic gate with pfb

NT