Experience with LV logic such as 74AUC?

[Joerg]

Hello All,

Battery voltages required are ever going down, even to single cell
applications. Does anybody have experience in using 74AUC or similar low
voltage logic? What do you think of it?

They look pretty enticing, except that their price tag is way higher
than the CD series logic I am used to. That puts a damper on some ideas.
But unfortunately CD won't go below 3V and even there the performance
becomes quite sluggish. Besides the price the other thing I am concerned
about is soldering these tiny S-PQFP packages while prototyping.

Regards, Joerg

 

[Fred Bloggs]

Hello All,

Battery voltages required are ever going down, even to single cell
applications. Does anybody have experience in using 74AUC or similar low
voltage logic? What do you think of it?

They look pretty enticing, except that their price tag is way higher
than the CD series logic I am used to. That puts a damper on some ideas.
But unfortunately CD won't go below 3V and even there the performance
becomes quite sluggish. Besides the price the other thing I am concerned
about is soldering these tiny S-PQFP packages while prototyping.

Regards, Joerg

If that's your concern then stick with the logic you trust and look into
the ultra low power, low noise, reasonably efficient, switched capacitor
battery voltage boost converters from manufacturers like Linear Technology.

 

[Joerg]

Hi Fred,

If that's your concern then stick with the logic you trust and look
into the ultra low power, low noise, reasonably efficient, switched
capacitor battery voltage boost converters from manufacturers like
Linear Technology.

Well, the price tag is really the main concern. I can solder it
allright, biting my tongue in order not to let any bad words slip out
after I bridged a few contacts again.

The converter is sometimes the solution and I have done that, mostly
good old PWM design with an inductor or transformer though. But whatever
step-up scheme you use, it becomes really tough when the unit has to
work without any switches, not even a power switch, yet has to have a
battery life in excess of a year. Usually there are several sensors and
other things that need to run all the time with next to nothing in power
consumption. If a switch cap booster would continuously run it would
consume quite a few microamps and that is sometimes too much.

Regards, Joerg

 

[Tim Shoppa]

Hello All,

Battery voltages required are ever going down, even to single cell
applications. Does anybody have experience in using 74AUC or similar low
voltage logic? What do you think of it?

They look pretty enticing, except that their price tag is way higher
than the CD series logic I am used to. That puts a damper on some ideas.
But unfortunately CD won't go below 3V and even there the performance
becomes quite sluggish.

I feel the same way... witness my request for 0.8V microcontroller
recommendations a week or two ago.

Some 74HC parts work (although slugglishly, like 50x the propogation
times at 5V) down to 1.2 or 1.3V, although I've never seen them specced
below 2V. They're almost all specced to 2V at least.

Besides the price the other thing I am concerned
about is soldering these tiny S-PQFP packages while prototyping.

I thought that some 74AUC parts were available in TinyLogic-type
packages (i.e. one NAND in a SOT-23-5 etc.) Those aren't so bad, the
pin spacing is really just a little bit closer than a TO-92.

SOT-70's are doable but take some effort for me.

But those BGA packages used for the denser functions, I can't ever
imagine prototyping with those.

It'd be nice to see "traditional" (e.g. 4 NAND's, 4-bit counter, etc.)
74AUC series parts in 14-pin SOIC or DIP packages but seeing where new
logic families have gone over the past decade, I don't ever expect to see
this with 74AUC's. There will be tinylogic style packages for the
low-density single-gate stuff and there will be BGA's for everything
else (mostly 8-bit or 16-bit or 32-bit bus parts).

Tim.

 

[Tim Shoppa]

Hi Tim,

That always puzzled me as well. There is a substantial market for uCs
for single cell operation but the industry either doesn't see it or
doesn't think it's big enough.

I suspect that there is a big market, yes, but it's for consumer gadgets
made by the millions, not little guys like you and me. Thus the Winbond
parts (masked ROM only) for example.

There may be something fundamentally difficult about flash-programmed
microcontrollers below 2.0V, I don't understand the physics/fabrication
well enough.

Yes, 2V is the limit there. Thing is, even if some work down to single
cell we can't use that fact in a design because it would be in violation
of the data sheet.

Samsung? Hitachi? Goldstar? I sear that
somebody used to spec their 74HC parts down to 1.5V, although I don't think
that's low enough to go to single-cell operation unless you're talking
about 2V or greater cells.

Tim.

 

[Joerg]

Hi Tim,

I suspect that there is a big market, yes, but it's for consumer gadgets
made by the millions, not little guys like you and me. Thus the Winbond
parts (masked ROM only) for example.

That is certainly one problem. But sometimes it's ok since the design
goes into mass production in the end. There is one other caveat. You
can't really switch anything other than a bipolar transistor at single
cell voltages. So a 1.2V uC would need some kind of charge pump,
preferably on board, that can be used when something needs swicthing. Of
course it can't be something that needs to stay on for long or the
advantage of micropower design would go by the wayside.

There may be something fundamentally difficult about flash-programmed
microcontrollers below 2.0V, I don't understand the physics/fabrication
well enough.

Don't know either. But applying a higher voltage for programming would
be fine in production. I had seen one uC that needed this.

Samsung? Hitachi? Goldstar? I sear that
somebody used to spec their 74HC parts down to 1.5V, although I don't think
that's low enough to go to single-cell operation unless you're talking
about 2V or greater cells.

1.5V is kind of marginal since all reasonably priced cells are just at
that voltage. It also would create a purchasing nightmare. I have seen
too often that a sales guy convinced the purchaser that this chip is
identical even though from another manufacturer and it would be just
fine. Well, often it wasn't.

74HC has been reported to work even lower, at slightly less than 1V.
Amazing. It's just not guaranteed.

Regards, Joerg

 

[Tim Shoppa]

Samsung? Hitachi? Goldstar? I swear that
somebody used to spec their 74HC parts down to 1.5V

After doing some research (meaning searching through a big pile
of databooks), maybe I was deluded about 74HC. However, TI CD74AC logic
is generally specced to 1.5V, maybe that's what I was thinking of.

Not TI SN74AC, that's only specced to 2.0V.

TI parts that start with CD are ex-Harris (and potentially ex-RCA),
am I getting the history right?

Not the holy grail of 0.8V parts in DIP or SOIC though...

Tim.