ARM microcontroller with 2.0-3.2V supply range

Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
though many have an internal regulator for 1.8V for the core.

Are there any ARM microcontrollers that can run directly on a
lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
a boost regulator or charge pump?

There are a lot of 8-bit microcontrollers that will do that, so
I'm somewhat surprised not to find a suitable ARM.

Thanks,
Eric

Eric Smith wrote:
> Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
> though many have an internal regulator for 1.8V for the core.
>
> Are there any ARM microcontrollers that can run directly on a
> lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
> a boost regulator or charge pump?
>
> There are a lot of 8-bit microcontrollers that will do that, so
> I'm somewhat surprised not to find a suitable ARM.

That is a very interesting question.  Looking at the data sheet for the
AT91SAM7S family, I see that the various supply voltages are either 1.8
or 3.3 volts.  However, the only supply that must be 3.3 volts is the
Flash supply for programming.  The I/Os are rated for operation at
either 1.8 or 3.3 volts.  I'm not sure why they don't rate the I/Os for
operation over the full range of voltage, but it may have to do with
powering the internal 1.8 volt regulator.  Either you provide 3.3 volts
which can internally derive 1.8 volts or you provide 1.8 volts to all
Vdds.  I guess if you use an external LDO to generate 1.8 volts you
might be able to power the IO supply from the battery voltage, or you
can use the 1.8 volts for all.  The current will be lower than if you
connected directly to the battery voltage so using a tiny LDO can be a
win-win.

This would be a good question to pose to Atmel support.  You can either
contact them directly or you can try posting your question to the
AT91.com web site.

Eric Smith wrote:
> Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
> though many have an internal regulator for 1.8V for the core.
>
> Are there any ARM microcontrollers that can run directly on a
> lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
> a boost regulator or charge pump?

yes

http://www.standardics.nxp.com/news/lpc2800/~LPC2888/#LPC2888

steve wrote:
> Eric Smith wrote:
> > Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
> > though many have an internal regulator for 1.8V for the core.
> >
> > Are there any ARM microcontrollers that can run directly on a
> > lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
> > a boost regulator or charge pump?
>
> yes
>
> http://www.standardics.nxp.com/news/lpc2800/~LPC2888/#LPC2888

This seems to be an unusual part.  The data sheet is preliminary and
the press release does not give many details, but it seems like the
chip includes a DC/DC switching converter.  So a coil and bulk caps are
still required and is typically the largest components of a switching
converter.  It is very odd that the data sheet has so little info on
this circuit.  It is not clear if it is a buck, boost, or a buck/boost
converter.  

I guess we will have to wait...

rickman wrote:
> steve wrote:
> > Eric Smith wrote:
> > > Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
> > > though many have an internal regulator for 1.8V for the core.
> > >
> > > Are there any ARM microcontrollers that can run directly on a
> > > lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
> > > a boost regulator or charge pump?
> >
> > yes
> >
> > http://www.standardics.nxp.com/news/lpc2800/~LPC2888/#LPC2888
>
> This seems to be an unusual part.  The data sheet is preliminary and
> the press release does not give many details, but it seems like the
> chip includes a DC/DC switching converter.  So a coil and bulk caps are
> still required and is typically the largest components of a switching
> converter.  It is very odd that the data sheet has so little info on
> this circuit.  It is not clear if it is a buck, boost, or a buck/boost
> converter.
>
> I guess we will have to wait...

yes it's a DC/DC converter, see users manual chapter 7

"steve" <bungalow_steve@yahoo.com> skrev i meddelandet 
news:1162693704.760605.61840@b28g2000cwb.googlegroups.com...
>
> Eric Smith wrote:
>> Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
>> though many have an internal regulator for 1.8V for the core.
>>
>> Are there any ARM microcontrollers that can run directly on a
>> lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
>> a boost regulator or charge pump?
>


AT91M40800 will run downto 1.8V, but
does not have a lot of fancy features.

-- 
Best Regards,
Ulf Samuelsson
This is intended to be my personal opinion which may,
or may not be shared by my employer Atmel Nordic AB 

Ulf Samuelsson wrote:
> "steve" <bungalow_steve@yahoo.com> skrev i meddelandet
> news:1162693704.760605.61840@b28g2000cwb.googlegroups.com...
> >
> > Eric Smith wrote:
> >> Most of the ARM microcontrollers seem to require 3.0-3.6V supply,
> >> though many have an internal regulator for 1.8V for the core.
> >>
> >> Are there any ARM microcontrollers that can run directly on a
> >> lithium cell (e.g., 2.0 to 3.2V supply), avoiding the need for
> >> a boost regulator or charge pump?
> >
>
>
> AT91M40800 will run downto 1.8V, but
> does not have a lot of fancy features.

It still seems to me that it would be better to add the very small LDO
and run from 1.8 volts regardless of battery voltage.  That should
minimize the current at all voltages and extend battery life over a
design that simply connects the part to the battery and draws higher
currents at the higher voltages.  Unless there is some clear advantage
to using a part like the AT91M40800, I would expect one of the SAM7S
parts running at 1.8 volts for both core and I/O to be a much better
choice.

"Ulf Samuelsson" <ulf@a-t-m-e-l.com> writes:
> AT91M40800 will run downto 1.8V, but
> does not have a lot of fancy features.

No flash memory, though.

I don't require that the flash memory be programmed at 1.8V (though that
would be nice), but I need a part that can run from Flash memory.

"rickman" <gnuarm@gmail.com> writes:
> Unless there is some clear advantage
> to using a part like the AT91M40800, I would expect one of the SAM7S
> parts running at 1.8 volts for both core and I/O to be a much better
> choice.

The datasheet indicates that the AT91SAM7S parts require 3.0-3.6V for the
flash memory.  Even just to read it.  Otherwise it would be OK.

Eric Smith wrote:
> "rickman" <gnuarm@gmail.com> writes:
> > Unless there is some clear advantage
> > to using a part like the AT91M40800, I would expect one of the SAM7S
> > parts running at 1.8 volts for both core and I/O to be a much better
> > choice.
>
> The datasheet indicates that the AT91SAM7S parts require 3.0-3.6V for the
> flash memory.  Even just to read it.  Otherwise it would be OK.

Yes, you are right.  They also have a separate VDDIN pin to power the
LDO.  Seems rather silly that they only offer 1.8 and 3.3 volt IO
standards given that you have to have 3.3 volts for the part to work
from Flash.  I guess there is some market for 1.8 volt IOs though.