Low voltage hobbyist-accessible microcontrollers?

[Tim Shoppa]

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Tim.

 

[Mike Harrison]

On 29 Aug 2004 07:29:12 -0700, shoppa@trailing-edge.com (Tim Shoppa) wrote:

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Tim.

Look at the OnSemi NCP1400 and similar for single-cell stepups.

 

[Fred Bartoli]

"Tim Shoppa" <shoppa@trailing-edge.com> a écrit dans le message de
news:bec993c8.0408290629.44b0f5fd@posting.google.com...

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature
limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM
programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Why stick to single cell ?
Two cells will widen your choice a lot, and unless you're heavly constrained
on weight or size, this is simpler than 1 cell + boost and a booster that
still starts at 0.9V isn't necessary easy to find/build.

At 2 cells, the MSP430 will give you power, ease of use, ISP, and Texas will
kindly sample it to you...


--
Thanks,
Fred.

 

[maxfoo]

On 29 Aug 2004 07:29:12 -0700, shoppa@trailing-edge.com (Tim Shoppa) wrote:

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Tim.

Didn't say what you're trying to accomplice, so I'll suggest also looking at
1.8v xilinx cpld's to supplement the design in mind if you need to stay at the
voltage level...

 

[Dave Garnett]

Some of the newer Atmel AVR's go down to 1.8v. They are much nicer than PICs
to use, as well ...

Dave

"Tim Shoppa" <shoppa@trailing-edge.com> wrote in message
news:bec993c8.0408290629.44b0f5fd@posting.google.com...

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature
limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM
programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Tim.

Posted Via Nuthinbutnews.Com Premium Usenet Newsgroup Services
----------------------------------------------------------
** SPEED ** RETENTION ** COMPLETION ** ANONYMITY **
----------------------------------------------------------
http://www.nuthinbutnews.com

 

[maxfoo]

On Sun, 29 Aug 2004 19:37:35 +0100, "Dave Garnett"
<dave.garnett@metapurple.co.uk> wrote:

Some of the newer Atmel AVR's go down to 1.8v. They are much nicer than PICs
to use, as well ...

Dave

Not really, AVR assembler is harder than Pic assembler for a newbie to learn.

Most people use a c compiler with the AVR and unless you use linux those c
compilers are expensive. The few free ones are buggy as hell too...

The new 6 and 8 pin PICs are low voltage (2V) 10F20x series and 12F675 for
examples.. 35 instructions

 

[Dave VanHorn]

Not really, AVR assembler is harder than Pic assembler for a newbie to
learn.

BAAAAALONEY!


I delivered my first AVR project in six weeks, from "what is an AVR" to up
and running.
It was a relatively complex chunk of software to read UPC/EAN barcodes.
I've also written PIC assembler, in one project, I implemented a bytecode
language interpreter in an F84, along with a bunch of interface routines.

As in any micro, there are things that will trip you up, but the internal
"architecture" of the pic, even the new DSPIC, is very ugly compared to the
AVR.

The reduced number of instructions in the PIC, means that it takes more
instructions to get anything done. The pic also takes four crystal cycles
to execute an instruction. The AVR does most instructions in one cycle..
Vectored interrupts on the AVR, no holes in the memory space on the AVR, no
read-modify-write bug/feature on the AVR....

Mchip's support is better though, I'll give them that.

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR

 

[maxfoo]

On Sun, 29 Aug 2004 14:31:22 -0500, "Dave VanHorn" <dvanhorn@cedar.net> wrote:

The reduced number of instructions in the PIC, means that it takes more
instructions to get anything done. The pic also takes four crystal cycles
to execute an instruction. The AVR does most instructions in one cycle..
Vectored interrupts on the AVR, no holes in the memory space on the AVR, no
read-modify-write bug/feature on the AVR....

Mchip's support is better though, I'll give them that.

There are more assembler code snippets on the web for the pic to do virtually
anything.

When I attended an Atmel seminar back in june, the atmel reps said that the avr
arch is optimized for C and they do not recommend using an assembler.

 

[Roy Smith]

maxfoo <maxfoo@punkassSPAM.com> wrote:

the atmel reps said that the avr arch is optimized for C a

Ah, so it's a pdp-11 clone?

 

[Roger Gt]

"Roy Smith" <roy@panix.com> wrote in message
news:roy-1C4475.15480029082004@reader1.panix.com...
: maxfoo <maxfoo@punkassSPAM.com> wrote:
: > the atmel reps said that the avr arch is optimized for C a
:
: Ah, so it's a pdp-11 clone?

I've written code in Assembly and forth for both!
The Rep is full of it!

By the way, Forth was written for the PDP11.
It's been ported to many other machines.

Assembly is the native code for any machine.
You write the boot and many other hardware interfaces in assembly
if your C compiler is not up to the task.

 

[Joerg]

Hi Tim,

Considering that single cells can dip to below 1.5V this might become a
tough call. It might make sense to consider power consumption as well
and here the MSP430 really sticks out in terms of economy. If, for
example, you could run it in sleep mode and then only use it for brief
bursts of operation a voltage conversion could become pretty efficient.
As long as there is enough energy storage for the time when the uC has
to run something.

Regards, Joerg

http://www.analogconsultants.com

 

[David L. Jones]

"Antti Lukats" <antti@case2000.com> wrote in message news:<cgssar$kip$07$1@news.t-online.com>...

"Tim Shoppa" <shoppa@trailing-edge.com> wrote in message
news:bec993c8.0408290629.44b0f5fd@posting.google.com...
In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature
limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

www.cypressmicro.com
runs from single cell, but has step up built in the micro !

antti

Which one?
The selection table shows 2.4V minimum.

Dave

 

[David L. Jones]

shoppa@trailing-edge.com (Tim Shoppa) wrote in message news:<bec993c8.0408290629.44b0f5fd@posting.google.com>...

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Tim.

What about using a 3V Lithium cell?, that's still "single cell".
If you want to run on a single Alkaline cell then you will need a
step-up converter like the MAX866, works to less than 0.8V.

Dave

 

[Antti Lukats]

"Tim Shoppa" <shoppa@trailing-edge.com> wrote in message
news:bec993c8.0408290629.44b0f5fd@posting.google.com...

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature
limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

www.cypressmicro.com
runs from single cell, but has step up built in the micro !

antti

 

[Rich Webb]

On Sun, 29 Aug 2004 19:19:29 GMT, maxfoo <maxfoo@punkassSPAM.com> wrote:

On Sun, 29 Aug 2004 19:37:35 +0100, "Dave Garnett"
dave.garnett@metapurple.co.uk> wrote:

Some of the newer Atmel AVR's go down to 1.8v. They are much nicer than PICs
to use, as well ...

Dave

Not really, AVR assembler is harder than Pic assembler for a newbie to learn.

IMHO, this is incorrect. I've developed several products using only
assembler on both architectures and my opinion is that the AVR layout is
MUCH more developer-friendly (even for a "newbie"). PICs certainly have
their place and I use whichever makes sense for a given project, but
trying to squeeze everything through that single W-register is a pain in
the arse.

Most people use a c compiler with the AVR

Well, don't "most people" also use a C compiler with PICs, at least for
projects where a higher-level language is indicated?

and unless you use linux those c
compilers are expensive. The few free ones are buggy as hell too...

Incorrect. There is a well supported MS Windows port of gcc for the AVRs
which is hardly "buggy as hell". The project is up on SourceForge so
anyone who is interested can access the bug tracker.
http://sourceforge.net/projects/winavr

The commercial C compilers are not all "expensive", either. I've been
using Imagecraft's ICC-AVR for a few years. Good product, NOT expensive.

BTW, to the best of my knowledge, a gcc cross-compiler for PICs has been
started a few times but none are released.

--
Rich Webb Norfolk, VA

 

[Dave VanHorn]

When I attended an Atmel seminar back in june, the atmel reps said that
the avr
arch is optimized for C and they do not recommend using an assembler.

It is optimized for C, but that has no bearing on using it with assembler.
That's what I do for a living.

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR

 

[Dave VanHorn]

"Roy Smith" <roy@panix.com> wrote in message
news:roy-1C4475.15480029082004@reader1.panix.com...

maxfoo <maxfoo@punkassSPAM.com> wrote:
the atmel reps said that the avr arch is optimized for C a

Ah, so it's a pdp-11 clone?

So I could get RSTS-E?

--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR

 

[picnic]

It can also depend on what you consider to be a suitable "cell"...

Lithium-manganese-dioxide cell (plain old Lithium button cell) has a
3V cell voltage with 2V end-of-life voltage at -20C. Good for most
embedded CPUs.

Lithium-thionyl-chloride batteries have 3V cell voltage with 2.5V
end-of-life at -40C. Also good for most CPUs, but they are expensive (
Check Digikey 3.6V lithium batteries)

BTW, the LR123 half-AA cells are also of the above class of cells.

Hope this helps!

shoppa@trailing-edge.com (Tim Shoppa) wrote in message news:<bec993c8.0408290629.44b0f5fd@posting.google.com>...

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

PIC16LF62x's go down to 2.0V (with some reasonable temperature limitations)
and are very hobbyist-accessible. (OK, there are some minor points
about verifying the flash at low voltage but those aren't biggies).
2.0V is OK for two-cell operation, but I want better.

The Fairchild ACE 1502 is good down to 1.8V, is serial EEPROM programmable,
etc. Not nearly so much stuff floating around the 'net about using them.
And they are still two-cell operation.

The Winbond W541L200 looks good down to 1.2V, which is good enough for
single-cell use (if still not ideal in terms of using the full discharge
curve). But it's mask-ROM programmable only and a complete non-starter
in terms of hobbyist tinkering.

Is there anything obvious that I've missed out on?

I may end up learning about single-cell-input voltage boosters instead of
searching for more low voltage microcontrollers. The TI TPS61001 series
looks good at a first glance, any alternatives?

Tim.

 

[John S. Dyson]

In article <h_-dnToqWKcaD6_cRVn-ow@comcast.com>,
"Dave VanHorn" <dvanhorn@cedar.net> writes:

"Roy Smith" <roy@panix.com> wrote in message
news:roy-1C4475.15480029082004@reader1.panix.com...
maxfoo <maxfoo@punkassSPAM.com> wrote:
the atmel reps said that the avr arch is optimized for C a

Ah, so it's a pdp-11 clone?

So I could get RSTS-E?

Instead, you could run a PDP11 emulator on your PC, and it would

probably be 10X faster than the fastest real PDP11 ever made.

John

 

[Tim Shoppa]

maxfoo <maxfoo@punkassSPAM.com> wrote in message news:<0cu3j0p7f2c85vfou0i9q8sntht5gcjkgf@4ax.com>...

In the perfect world there would be a flash-based microcontroller,
serially programmable, that could run off a single cell (meaning 0.9V -
1.8V or so).

Didn't say what you're trying to accomplice

[Again, assuming a single cell == single alkaline or carbon-zinc 1.5V
nominal cell]

Correct, I didn't. Mostly I'm looking at all the cheap consumer gizmos
which only use one cell (often a button cell) and asking "how do they do that?".
Then "how can I do that?" Some of these single-cell gizmos have LED's
so they obviously have to boost somehow (possibly using charge pumps),
others are "just" LCD and buttons and don't obviously have any voltage
boost inside (although they may, combined with some clever power-down
stuff).

Some of the LED's are for remote-control purposes and I suppose it's possible
they take advantage of the smaller Vf there to do it without a voltage boost.

Another inspiration is my recent playing around with LM3909-style single-cell
LED blinkers (again charge pump boosters).

Tim.