On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
MHz. Unfortunately I'm not in a position make any hardware changes.
The data sheet for the MSP430F1611 seems to indicate that the micro can
only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
issues at 4 MHz. Can I then simply program the clock control register
BCSCTL2 to divide MCLK by 2 and solve the problem?
Thanks in advance.
Running at 3.3V @8 MHz
Started by ●February 15, 2007
Reply by ●February 15, 20072007-02-15
Hi,
I am using 8MHz in 3.3V VCC and so far so good. I have not read from datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can be run at VCC greater than 2.8 Volts.
mago
markcrowptm wrote:
On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
MHz. Unfortunately I'm not in a position make any hardware changes.
The data sheet for the MSP430F1611 seems to indicate that the micro can
only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
issues at 4 MHz. Can I then simply program the clock control register
BCSCTL2 to divide MCLK by 2 and solve the problem?
Thanks in advance.
---------------------------------
Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
I am using 8MHz in 3.3V VCC and so far so good. I have not read from datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can be run at VCC greater than 2.8 Volts.
mago
markcrowptm wrote:
On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
MHz. Unfortunately I'm not in a position make any hardware changes.
The data sheet for the MSP430F1611 seems to indicate that the micro can
only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
issues at 4 MHz. Can I then simply program the clock control register
BCSCTL2 to divide MCLK by 2 and solve the problem?
Thanks in advance.
---------------------------------
Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
Reply by ●February 15, 20072007-02-15
The graph on the '1611 is clear regarding running at 8MHz - the supply
voltage needs to
be 3.6VDC. If you want to run out of spec - that's up to you - but if your project requires
running in spec - then your Vcc needs to be 3.6VDC.
Having said that - when hunting for speed paths - I've run the '1611 (rev. B) at 10MHz and 12MHz, but
my Vcc was always 3.6VDC. The part will work with different voltages/frequencies - but it all depends
if you want to be in spec - or if the project specifications require (to be in spec...).
Dividing MCLK by 2 as you have asked will work - that means the core will be clocked at 4MHz - but
there's still the question of the divider circuit and the I/F circuitry to the oscillator running at 8MHz - which
is still out of spec.
Regards,
John W.
----- Original Message -----
From: mago Umandam
To: m...
Sent: Thursday, February 15, 2007 4:43 PM
Subject: Re: [msp430] Running at 3.3V @8 MHz
Hi,
I am using 8MHz in 3.3V VCC and so far so good. I have not read from datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can be run at VCC greater than 2.8 Volts.
mago
markcrowptm wrote:
On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
MHz. Unfortunately I'm not in a position make any hardware changes.
The data sheet for the MSP430F1611 seems to indicate that the micro can
only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
issues at 4 MHz. Can I then simply program the clock control register
BCSCTL2 to divide MCLK by 2 and solve the problem?
Thanks in advance.
---------------------------------
Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
be 3.6VDC. If you want to run out of spec - that's up to you - but if your project requires
running in spec - then your Vcc needs to be 3.6VDC.
Having said that - when hunting for speed paths - I've run the '1611 (rev. B) at 10MHz and 12MHz, but
my Vcc was always 3.6VDC. The part will work with different voltages/frequencies - but it all depends
if you want to be in spec - or if the project specifications require (to be in spec...).
Dividing MCLK by 2 as you have asked will work - that means the core will be clocked at 4MHz - but
there's still the question of the divider circuit and the I/F circuitry to the oscillator running at 8MHz - which
is still out of spec.
Regards,
John W.
----- Original Message -----
From: mago Umandam
To: m...
Sent: Thursday, February 15, 2007 4:43 PM
Subject: Re: [msp430] Running at 3.3V @8 MHz
Hi,
I am using 8MHz in 3.3V VCC and so far so good. I have not read from datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can be run at VCC greater than 2.8 Volts.
mago
markcrowptm wrote:
On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
MHz. Unfortunately I'm not in a position make any hardware changes.
The data sheet for the MSP430F1611 seems to indicate that the micro can
only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
issues at 4 MHz. Can I then simply program the clock control register
BCSCTL2 to divide MCLK by 2 and solve the problem?
Thanks in advance.
---------------------------------
Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
Reply by ●February 16, 20072007-02-16
The datasheet seem to say that you need need at lest 1.8V
for the CPU to run at MCLK=4MHz. But you need at least 2.8V
for 8MHz crystal.
(Read Note 2 under Recommended Operating Conditions.)
Thus I think XTCLK=8MHz, MCLK=XTCLK/2=4MHz is within spec
if Vcc is at least 2.8V.
--- In m..., "John C. Westmoreland, P.E."
wrote:
>
> The graph on the '1611 is clear regarding running at 8MHz - the
supply voltage needs to
> be 3.6VDC. If you want to run out of spec - that's up to you - but
if your project requires
> running in spec - then your Vcc needs to be 3.6VDC.
>
> Having said that - when hunting for speed paths - I've run
the '1611 (rev. B) at 10MHz and 12MHz, but
> my Vcc was always 3.6VDC. The part will work with different
voltages/frequencies - but it all depends
> if you want to be in spec - or if the project specifications
require (to be in spec...).
>
> Dividing MCLK by 2 as you have asked will work - that means the
core will be clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F
circuitry to the oscillator running at 8MHz - which
> is still out of spec.
>
> Regards,
> John W.
> ----- Original Message -----
> From: mago Umandam
> To: m...
> Sent: Thursday, February 15, 2007 4:43 PM
> Subject: Re: [msp430] Running at 3.3V @8 MHz
> Hi,
>
> I am using 8MHz in 3.3V VCC and so far so good. I have not read
from datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says
that 8 MHz can be run at VCC greater than 2.8 Volts.
>
> mago
> markcrowptm wrote:
> On my project, VCC can run from 1.8V to 3.3V and my oscillator is
8
> MHz. Unfortunately I'm not in a position make any hardware
changes.
> The data sheet for the MSP430F1611 seems to indicate that the
micro can
> only run at 8 MHz if VCC is 3.6V. It indicates I can run with my
VCC
> issues at 4 MHz. Can I then simply program the clock control
register
> BCSCTL2 to divide MCLK by 2 and solve the problem?
>
> Thanks in advance.
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers
users.
>
>
>
>
>
>
>
for the CPU to run at MCLK=4MHz. But you need at least 2.8V
for 8MHz crystal.
(Read Note 2 under Recommended Operating Conditions.)
Thus I think XTCLK=8MHz, MCLK=XTCLK/2=4MHz is within spec
if Vcc is at least 2.8V.
--- In m..., "John C. Westmoreland, P.E."
wrote:
>
> The graph on the '1611 is clear regarding running at 8MHz - the
supply voltage needs to
> be 3.6VDC. If you want to run out of spec - that's up to you - but
if your project requires
> running in spec - then your Vcc needs to be 3.6VDC.
>
> Having said that - when hunting for speed paths - I've run
the '1611 (rev. B) at 10MHz and 12MHz, but
> my Vcc was always 3.6VDC. The part will work with different
voltages/frequencies - but it all depends
> if you want to be in spec - or if the project specifications
require (to be in spec...).
>
> Dividing MCLK by 2 as you have asked will work - that means the
core will be clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F
circuitry to the oscillator running at 8MHz - which
> is still out of spec.
>
> Regards,
> John W.
> ----- Original Message -----
> From: mago Umandam
> To: m...
> Sent: Thursday, February 15, 2007 4:43 PM
> Subject: Re: [msp430] Running at 3.3V @8 MHz
> Hi,
>
> I am using 8MHz in 3.3V VCC and so far so good. I have not read
from datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says
that 8 MHz can be run at VCC greater than 2.8 Volts.
>
> mago
> markcrowptm wrote:
> On my project, VCC can run from 1.8V to 3.3V and my oscillator is
8
> MHz. Unfortunately I'm not in a position make any hardware
changes.
> The data sheet for the MSP430F1611 seems to indicate that the
micro can
> only run at 8 MHz if VCC is 3.6V. It indicates I can run with my
VCC
> issues at 4 MHz. Can I then simply program the clock control
register
> BCSCTL2 to divide MCLK by 2 and solve the problem?
>
> Thanks in advance.
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers
users.
>
>
>
>
>
>
>
Reply by ●February 16, 20072007-02-16
> *Subject:* Re: [msp430] Running at 3.3V @8 MHz
> *From:* "John C. Westmoreland, P.E."
>
> *To:*
> *Date:* Thu, 15 Feb 2007 17:49:31 -0800
> Dividing MCLK by 2 as you have asked will work - that means the
> core will be clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F
> circuitry to the oscillator running at 8MHz - which
> is still out of spec.
I have a feeling that someone asked TI about this and was told it wasn't
a problem... though I admit that doesn't make it in-spec, and it doesn't
tell you exactly how high you COULD go with the input divider.
I would have little conscience about over-running the input divider,
provided I'd tried it at a 50% greater sped than I was shipping it. I
suspect it's 100K transistors in the core that has issues with timing,
not one odd f/f
David
> *From:* "John C. Westmoreland, P.E."
>
> *To:*
> *Date:* Thu, 15 Feb 2007 17:49:31 -0800
> Dividing MCLK by 2 as you have asked will work - that means the
> core will be clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F
> circuitry to the oscillator running at 8MHz - which
> is still out of spec.
I have a feeling that someone asked TI about this and was told it wasn't
a problem... though I admit that doesn't make it in-spec, and it doesn't
tell you exactly how high you COULD go with the input divider.
I would have little conscience about over-running the input divider,
provided I'd tried it at a 50% greater sped than I was shipping it. I
suspect it's 100K transistors in the core that has issues with timing,
not one odd f/f
David
Reply by ●February 16, 20072007-02-16
We asked TI about this and they stated that the frequency restriction only
applies to the cpu core and not the crystal and associated divider circuits.
Therefore use of 8MHz is within spec provided that the core and peripherals are
clocked - as suggested - at 4MHz.
Hugh
--- "John C. Westmoreland, P.E." wrote:
> The graph on the '1611 is clear regarding running at 8MHz - the supply
> voltage needs to
> be 3.6VDC. If you want to run out of spec - that's up to you - but if your
> project requires
> running in spec - then your Vcc needs to be 3.6VDC.
>
> Having said that - when hunting for speed paths - I've run the '1611 (rev. B)
> at 10MHz and 12MHz, but
> my Vcc was always 3.6VDC. The part will work with different
> voltages/frequencies - but it all depends
> if you want to be in spec - or if the project specifications require (to be
> in spec...).
>
> Dividing MCLK by 2 as you have asked will work - that means the core will be
> clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F circuitry to
> the oscillator running at 8MHz - which
> is still out of spec.
>
> Regards,
> John W.
> ----- Original Message -----
> From: mago Umandam
> To: m...
> Sent: Thursday, February 15, 2007 4:43 PM
> Subject: Re: [msp430] Running at 3.3V @8 MHz
> Hi,
>
> I am using 8MHz in 3.3V VCC and so far so good. I have not read from
> datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can
> be run at VCC greater than 2.8 Volts.
>
> mago
> markcrowptm wrote:
> On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
> MHz. Unfortunately I'm not in a position make any hardware changes.
> The data sheet for the MSP430F1611 seems to indicate that the micro can
> only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
> issues at 4 MHz. Can I then simply program the clock control register
> BCSCTL2 to divide MCLK by 2 and solve the problem?
>
> Thanks in advance.
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
>
>
>
>
>
>
>
>
> Yahoo! Groups Links
applies to the cpu core and not the crystal and associated divider circuits.
Therefore use of 8MHz is within spec provided that the core and peripherals are
clocked - as suggested - at 4MHz.
Hugh
--- "John C. Westmoreland, P.E." wrote:
> The graph on the '1611 is clear regarding running at 8MHz - the supply
> voltage needs to
> be 3.6VDC. If you want to run out of spec - that's up to you - but if your
> project requires
> running in spec - then your Vcc needs to be 3.6VDC.
>
> Having said that - when hunting for speed paths - I've run the '1611 (rev. B)
> at 10MHz and 12MHz, but
> my Vcc was always 3.6VDC. The part will work with different
> voltages/frequencies - but it all depends
> if you want to be in spec - or if the project specifications require (to be
> in spec...).
>
> Dividing MCLK by 2 as you have asked will work - that means the core will be
> clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F circuitry to
> the oscillator running at 8MHz - which
> is still out of spec.
>
> Regards,
> John W.
> ----- Original Message -----
> From: mago Umandam
> To: m...
> Sent: Thursday, February 15, 2007 4:43 PM
> Subject: Re: [msp430] Running at 3.3V @8 MHz
> Hi,
>
> I am using 8MHz in 3.3V VCC and so far so good. I have not read from
> datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can
> be run at VCC greater than 2.8 Volts.
>
> mago
> markcrowptm wrote:
> On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
> MHz. Unfortunately I'm not in a position make any hardware changes.
> The data sheet for the MSP430F1611 seems to indicate that the micro can
> only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
> issues at 4 MHz. Can I then simply program the clock control register
> BCSCTL2 to divide MCLK by 2 and solve the problem?
>
> Thanks in advance.
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
>
>
>
>
>
>
>
>
> Yahoo! Groups Links
Reply by ●February 16, 20072007-02-16
The design team on a project I worked on a couple of years ago went round and
round on this. Talked to TI several times, etc. This could be part specific
maybe; we were using the '1611 - since it had to pass FDA - it was
determined that we had to have Vcc be 3.6V.
Sent via BlackBerry from T-Mobile
-----Original Message-----
From: Hugh Molesworth
Date: Fri, 16 Feb 2007 10:45:49
To:m...
Subject: Re: [msp430] Running at 3.3V @8 MHz
We asked TI about this and they stated that the frequency restriction only
applies to the cpu core and not the crystal and associated divider circuits.
Therefore use of 8MHz is within spec provided that the core and peripherals are
clocked - as suggested - at 4MHz.
Hugh
--- "John C. Westmoreland, P.E." gineering.com> wrote:
> The graph on the '1611 is clear regarding running at 8MHz - the supply
> voltage needs to
> be 3.6VDC. If you want to run out of spec - that's up to you - but if your
> project requires
> running in spec - then your Vcc needs to be 3.6VDC.
>
> Having said that - when hunting for speed paths - I've run the '1611 (rev. B)
> at 10MHz and 12MHz, but
> my Vcc was always 3.6VDC. The part will work with different
> voltages/frequencies - but it all depends
> if you want to be in spec - or if the project specifications require (to be
> in spec...).
>
> Dividing MCLK by 2 as you have asked will work - that means the core will be
> clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F circuitry to
> the oscillator running at 8MHz - which
> is still out of spec.
>
> Regards,
> John W.
>
>
> ----- Original Message -----
> From: mago Umandam
> To: msp430@yahoogroups.: com
> Sent: Thursday, February 15, 2007 4:43 PM
> Subject: Re: [msp430] Running at 3.3V @8 MHz
>
>
> Hi,
>
> I am using 8MHz in 3.3V VCC and so far so good. I have not read from
> datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can
> be run at VCC greater than 2.8 Volts.
>
> mago
>
>
> markcrowptm > wrote:
> On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
> MHz. Unfortunately I'm not in a position make any hardware changes.
> The data sheet for the MSP430F1611 seems to indicate that the micro can
> only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
> issues at 4 MHz. Can I then simply program the clock control register
> BCSCTL2 to divide MCLK by 2 and solve the problem?
>
> Thanks in advance.
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
>
>
>
>
>
>
>
>
>
>
>
> > egroups.com
>
>
>
Sent via BlackBerry from T-Mobile
-----Original Message-----
From: Hugh Molesworth
Date: Fri, 16 Feb 2007 10:45:49
To:m...
Subject: Re: [msp430] Running at 3.3V @8 MHz
We asked TI about this and they stated that the frequency restriction only
applies to the cpu core and not the crystal and associated divider circuits.
Therefore use of 8MHz is within spec provided that the core and peripherals are
clocked - as suggested - at 4MHz.
Hugh
--- "John C. Westmoreland, P.E." gineering.com> wrote:
> The graph on the '1611 is clear regarding running at 8MHz - the supply
> voltage needs to
> be 3.6VDC. If you want to run out of spec - that's up to you - but if your
> project requires
> running in spec - then your Vcc needs to be 3.6VDC.
>
> Having said that - when hunting for speed paths - I've run the '1611 (rev. B)
> at 10MHz and 12MHz, but
> my Vcc was always 3.6VDC. The part will work with different
> voltages/frequencies - but it all depends
> if you want to be in spec - or if the project specifications require (to be
> in spec...).
>
> Dividing MCLK by 2 as you have asked will work - that means the core will be
> clocked at 4MHz - but
> there's still the question of the divider circuit and the I/F circuitry to
> the oscillator running at 8MHz - which
> is still out of spec.
>
> Regards,
> John W.
>
>
> ----- Original Message -----
> From: mago Umandam
> To: msp430@yahoogroups.: com
> Sent: Thursday, February 15, 2007 4:43 PM
> Subject: Re: [msp430] Running at 3.3V @8 MHz
>
>
> Hi,
>
> I am using 8MHz in 3.3V VCC and so far so good. I have not read from
> datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8 MHz can
> be run at VCC greater than 2.8 Volts.
>
> mago
>
>
> markcrowptm > wrote:
> On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
> MHz. Unfortunately I'm not in a position make any hardware changes.
> The data sheet for the MSP430F1611 seems to indicate that the micro can
> only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
> issues at 4 MHz. Can I then simply program the clock control register
> BCSCTL2 to divide MCLK by 2 and solve the problem?
>
> Thanks in advance.
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
>
>
>
>
>
>
>
>
>
>
>
> > egroups.com
>
>
>
Reply by ●February 20, 20072007-02-20
Hi folks,
not too much surprisingly even TI uses its chip beyond clock specifications.
If you look at the eZ430U development board you'll find an MSP430F16x
running at *12 MHz* at 3.6V! (It uses the 12 MHz because of the USB chip.)
I guess they had been very conservative to save on test budget.
By the way, if you want to be strictly on specs keep in mind that you have
to run above 3.6V to have enough room for regulator IC overall tolerance.
Cheers,
Bruno
-----Messaggio Originale-----
Da: "John C. Westmoreland, P.E."
A:
Data invio: venerd16 febbraio 2007 20.12
Oggetto: Re: [msp430] Running at 3.3V @8 MHz
| The design team on a project I worked on a couple of years ago went round
and round on this. Talked to TI several times, etc. This could be part
specific maybe; we were using the '1611 - since it had to pass FDA - it was
determined that we had to have Vcc be 3.6V.
|
| Sent via BlackBerry from T-Mobile
|
| -----Original Message-----
| From: Hugh Molesworth
| Date: Fri, 16 Feb 2007 10:45:49
| To:m...
| Subject: Re: [msp430] Running at 3.3V @8 MHz
|
| We asked TI about this and they stated that the frequency restriction only
| applies to the cpu core and not the crystal and associated divider
circuits.
| Therefore use of 8MHz is within spec provided that the core and
peripherals are
| clocked - as suggested - at 4MHz.
|
| Hugh
|
| --- "John C. Westmoreland, P.E." gineering.com> wrote:
|
| > The graph on the '1611 is clear regarding running at 8MHz - the supply
| > voltage needs to
| > be 3.6VDC. If you want to run out of spec - that's up to you - but if
your
| > project requires
| > running in spec - then your Vcc needs to be 3.6VDC.
| >
| > Having said that - when hunting for speed paths - I've run the '1611
(rev. B)
| > at 10MHz and 12MHz, but
| > my Vcc was always 3.6VDC. The part will work with different
| > voltages/frequencies - but it all depends
| > if you want to be in spec - or if the project specifications require (to
be
| > in spec...).
| >
| > Dividing MCLK by 2 as you have asked will work - that means the core
will be
| > clocked at 4MHz - but
| > there's still the question of the divider circuit and the I/F circuitry
to
| > the oscillator running at 8MHz - which
| > is still out of spec.
| >
| > Regards,
| > John W.
| >
| >
| > ----- Original Message -----
| > From: mago Umandam
| > To: msp430@yahoogroups.: com
| > Sent: Thursday, February 15, 2007 4:43 PM
| > Subject: Re: [msp430] Running at 3.3V @8 MHz
| >
| >
| > Hi,
| >
| > I am using 8MHz in 3.3V VCC and so far so good. I have not read from
| > datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8
MHz can
| > be run at VCC greater than 2.8 Volts.
| >
| > mago
| >
| >
| > markcrowptm > wrote:
| > On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
| > MHz. Unfortunately I'm not in a position make any hardware changes.
| > The data sheet for the MSP430F1611 seems to indicate that the micro can
| > only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
| > issues at 4 MHz. Can I then simply program the clock control register
| > BCSCTL2 to divide MCLK by 2 and solve the problem?
| >
| > Thanks in advance.
| >
| > ---------------------------------
| > Need Mail bonding?
| > Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| > >
egroups.com
| >
| >
| >
not too much surprisingly even TI uses its chip beyond clock specifications.
If you look at the eZ430U development board you'll find an MSP430F16x
running at *12 MHz* at 3.6V! (It uses the 12 MHz because of the USB chip.)
I guess they had been very conservative to save on test budget.
By the way, if you want to be strictly on specs keep in mind that you have
to run above 3.6V to have enough room for regulator IC overall tolerance.
Cheers,
Bruno
-----Messaggio Originale-----
Da: "John C. Westmoreland, P.E."
A:
Data invio: venerd16 febbraio 2007 20.12
Oggetto: Re: [msp430] Running at 3.3V @8 MHz
| The design team on a project I worked on a couple of years ago went round
and round on this. Talked to TI several times, etc. This could be part
specific maybe; we were using the '1611 - since it had to pass FDA - it was
determined that we had to have Vcc be 3.6V.
|
| Sent via BlackBerry from T-Mobile
|
| -----Original Message-----
| From: Hugh Molesworth
| Date: Fri, 16 Feb 2007 10:45:49
| To:m...
| Subject: Re: [msp430] Running at 3.3V @8 MHz
|
| We asked TI about this and they stated that the frequency restriction only
| applies to the cpu core and not the crystal and associated divider
circuits.
| Therefore use of 8MHz is within spec provided that the core and
peripherals are
| clocked - as suggested - at 4MHz.
|
| Hugh
|
| --- "John C. Westmoreland, P.E." gineering.com> wrote:
|
| > The graph on the '1611 is clear regarding running at 8MHz - the supply
| > voltage needs to
| > be 3.6VDC. If you want to run out of spec - that's up to you - but if
your
| > project requires
| > running in spec - then your Vcc needs to be 3.6VDC.
| >
| > Having said that - when hunting for speed paths - I've run the '1611
(rev. B)
| > at 10MHz and 12MHz, but
| > my Vcc was always 3.6VDC. The part will work with different
| > voltages/frequencies - but it all depends
| > if you want to be in spec - or if the project specifications require (to
be
| > in spec...).
| >
| > Dividing MCLK by 2 as you have asked will work - that means the core
will be
| > clocked at 4MHz - but
| > there's still the question of the divider circuit and the I/F circuitry
to
| > the oscillator running at 8MHz - which
| > is still out of spec.
| >
| > Regards,
| > John W.
| >
| >
| > ----- Original Message -----
| > From: mago Umandam
| > To: msp430@yahoogroups.: com
| > Sent: Thursday, February 15, 2007 4:43 PM
| > Subject: Re: [msp430] Running at 3.3V @8 MHz
| >
| >
| > Hi,
| >
| > I am using 8MHz in 3.3V VCC and so far so good. I have not read from
| > datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8
MHz can
| > be run at VCC greater than 2.8 Volts.
| >
| > mago
| >
| >
| > markcrowptm > wrote:
| > On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
| > MHz. Unfortunately I'm not in a position make any hardware changes.
| > The data sheet for the MSP430F1611 seems to indicate that the micro can
| > only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
| > issues at 4 MHz. Can I then simply program the clock control register
| > BCSCTL2 to divide MCLK by 2 and solve the problem?
| >
| > Thanks in advance.
| >
| > ---------------------------------
| > Need Mail bonding?
| > Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| > >
egroups.com
| >
| >
| >
Reply by ●February 20, 20072007-02-20
Good observation Bruno :-)
Sometimes characterisation or specs can border the ridiculous..
As an example, in 1996 I used a PIC16LC71 in a mixed signal design where the circuit had to run
down to 3.0 Volts.
Since this was a very high volume design and the manufacturer buggered up production badly
(British Aerospace if you must know - there you go, it's long ago enough :-) - therefore faced
massive litigation, the defense tried to criticise my design by pointing out that the 16LC71 was
only rated down to 3.5 Volts.. (cheap shot, the alignment cores on RF ferrites had been cracked
with metal screwdrivers, amongst a host of others atrocities :-)
Strictly speaking the whole 16LC7xx family is rated down to 3.0 Volts...except for the LC71. (Not
that this affected my core + RF design anyway :-)
The reason ? Stupid :
The A/D converter in that family had been eventually tested down to 3.0 Volts but *never* had been
characterised in the _original_ 16LC71.
So, although this was the IDENTICAL A/D, Microchip didnt want to (bother to) rate that ADC down
to 3.0 Volts... I was told exactly this by one of the design people in Chandler (and that it was
fine at 3.0 V).
I then asked for an official statement (while on the phone) that the LC71 indeed runs down to 3.0
Volts (core), but that A/D wasn't characterised at 3.0 Volts (my design clocked the PIC at 100 kHz
btw !). I was assured I'd get one...
The next day I was Emailed an "official statement" :
"The LC71 as per datasheet is rated down to 3.5 Volts" ...
Geez, well you can't win 'em all I guess - anyhow, I was fully exonerated.
Best Regards,
Kris
-----Original Message-----
From: m... [mailto:m...] On Behalf Of Bruno Galli
Sent: Tuesday, 20 February 2007 9:41 PM
To: m...
Subject: Re: [msp430] Running at 3.3V @8 MHz
Hi folks,
not too much surprisingly even TI uses its chip beyond clock specifications.
If you look at the eZ430U development board you'll find an MSP430F16x
running at *12 MHz* at 3.6V! (It uses the 12 MHz because of the USB chip.)
I guess they had been very conservative to save on test budget.
By the way, if you want to be strictly on specs keep in mind that you have
to run above 3.6V to have enough room for regulator IC overall tolerance.
Cheers,
Bruno
-----Messaggio Originale-----
Da: "John C. Westmoreland, P.E."
A:
Data invio: venerd16 febbraio 2007 20.12
Oggetto: Re: [msp430] Running at 3.3V @8 MHz
| The design team on a project I worked on a couple of years ago went round
and round on this. Talked to TI several times, etc. This could be part
specific maybe; we were using the '1611 - since it had to pass FDA - it was
determined that we had to have Vcc be 3.6V.
|
| Sent via BlackBerry from T-Mobile
|
| -----Original Message-----
| From: Hugh Molesworth
| Date: Fri, 16 Feb 2007 10:45:49
| To:m...
| Subject: Re: [msp430] Running at 3.3V @8 MHz
|
| We asked TI about this and they stated that the frequency restriction only
| applies to the cpu core and not the crystal and associated divider
circuits.
| Therefore use of 8MHz is within spec provided that the core and
peripherals are
| clocked - as suggested - at 4MHz.
|
| Hugh
|
| --- "John C. Westmoreland, P.E." gineering.com> wrote:
|
| > The graph on the '1611 is clear regarding running at 8MHz - the supply
| > voltage needs to
| > be 3.6VDC. If you want to run out of spec - that's up to you - but if
your
| > project requires
| > running in spec - then your Vcc needs to be 3.6VDC.
| >
| > Having said that - when hunting for speed paths - I've run the '1611
(rev. B)
| > at 10MHz and 12MHz, but
| > my Vcc was always 3.6VDC. The part will work with different
| > voltages/frequencies - but it all depends
| > if you want to be in spec - or if the project specifications require (to
be
| > in spec...).
| >
| > Dividing MCLK by 2 as you have asked will work - that means the core
will be
| > clocked at 4MHz - but
| > there's still the question of the divider circuit and the I/F circuitry
to
| > the oscillator running at 8MHz - which
| > is still out of spec.
| >
| > Regards,
| > John W.
| >
| >
| > ----- Original Message -----
| > From: mago Umandam
| > To: msp430@yahoogroups.: com
| > Sent: Thursday, February 15, 2007 4:43 PM
| > Subject: Re: [msp430] Running at 3.3V @8 MHz
| >
| >
| > Hi,
| >
| > I am using 8MHz in 3.3V VCC and so far so good. I have not read from
| > datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8
MHz can
| > be run at VCC greater than 2.8 Volts.
| >
| > mago
| >
| >
| > markcrowptm > wrote:
| > On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
| > MHz. Unfortunately I'm not in a position make any hardware changes.
| > The data sheet for the MSP430F1611 seems to indicate that the micro can
| > only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
| > issues at 4 MHz. Can I then simply program the clock control register
| > BCSCTL2 to divide MCLK by 2 and solve the problem?
| >
| > Thanks in advance.
| >
| > ---------------------------------
| > Need Mail bonding?
| > Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| > >
egroups.com
| >
| >
| > Yahoo! Groups Links
| >
| >
| >
| >
|
|
|
|
|
|
|
| Yahoo! Groups Links
|
|
|
Yahoo! Groups Links
Sometimes characterisation or specs can border the ridiculous..
As an example, in 1996 I used a PIC16LC71 in a mixed signal design where the circuit had to run
down to 3.0 Volts.
Since this was a very high volume design and the manufacturer buggered up production badly
(British Aerospace if you must know - there you go, it's long ago enough :-) - therefore faced
massive litigation, the defense tried to criticise my design by pointing out that the 16LC71 was
only rated down to 3.5 Volts.. (cheap shot, the alignment cores on RF ferrites had been cracked
with metal screwdrivers, amongst a host of others atrocities :-)
Strictly speaking the whole 16LC7xx family is rated down to 3.0 Volts...except for the LC71. (Not
that this affected my core + RF design anyway :-)
The reason ? Stupid :
The A/D converter in that family had been eventually tested down to 3.0 Volts but *never* had been
characterised in the _original_ 16LC71.
So, although this was the IDENTICAL A/D, Microchip didnt want to (bother to) rate that ADC down
to 3.0 Volts... I was told exactly this by one of the design people in Chandler (and that it was
fine at 3.0 V).
I then asked for an official statement (while on the phone) that the LC71 indeed runs down to 3.0
Volts (core), but that A/D wasn't characterised at 3.0 Volts (my design clocked the PIC at 100 kHz
btw !). I was assured I'd get one...
The next day I was Emailed an "official statement" :
"The LC71 as per datasheet is rated down to 3.5 Volts" ...
Geez, well you can't win 'em all I guess - anyhow, I was fully exonerated.
Best Regards,
Kris
-----Original Message-----
From: m... [mailto:m...] On Behalf Of Bruno Galli
Sent: Tuesday, 20 February 2007 9:41 PM
To: m...
Subject: Re: [msp430] Running at 3.3V @8 MHz
Hi folks,
not too much surprisingly even TI uses its chip beyond clock specifications.
If you look at the eZ430U development board you'll find an MSP430F16x
running at *12 MHz* at 3.6V! (It uses the 12 MHz because of the USB chip.)
I guess they had been very conservative to save on test budget.
By the way, if you want to be strictly on specs keep in mind that you have
to run above 3.6V to have enough room for regulator IC overall tolerance.
Cheers,
Bruno
-----Messaggio Originale-----
Da: "John C. Westmoreland, P.E."
A:
Data invio: venerd16 febbraio 2007 20.12
Oggetto: Re: [msp430] Running at 3.3V @8 MHz
| The design team on a project I worked on a couple of years ago went round
and round on this. Talked to TI several times, etc. This could be part
specific maybe; we were using the '1611 - since it had to pass FDA - it was
determined that we had to have Vcc be 3.6V.
|
| Sent via BlackBerry from T-Mobile
|
| -----Original Message-----
| From: Hugh Molesworth
| Date: Fri, 16 Feb 2007 10:45:49
| To:m...
| Subject: Re: [msp430] Running at 3.3V @8 MHz
|
| We asked TI about this and they stated that the frequency restriction only
| applies to the cpu core and not the crystal and associated divider
circuits.
| Therefore use of 8MHz is within spec provided that the core and
peripherals are
| clocked - as suggested - at 4MHz.
|
| Hugh
|
| --- "John C. Westmoreland, P.E." gineering.com> wrote:
|
| > The graph on the '1611 is clear regarding running at 8MHz - the supply
| > voltage needs to
| > be 3.6VDC. If you want to run out of spec - that's up to you - but if
your
| > project requires
| > running in spec - then your Vcc needs to be 3.6VDC.
| >
| > Having said that - when hunting for speed paths - I've run the '1611
(rev. B)
| > at 10MHz and 12MHz, but
| > my Vcc was always 3.6VDC. The part will work with different
| > voltages/frequencies - but it all depends
| > if you want to be in spec - or if the project specifications require (to
be
| > in spec...).
| >
| > Dividing MCLK by 2 as you have asked will work - that means the core
will be
| > clocked at 4MHz - but
| > there's still the question of the divider circuit and the I/F circuitry
to
| > the oscillator running at 8MHz - which
| > is still out of spec.
| >
| > Regards,
| > John W.
| >
| >
| > ----- Original Message -----
| > From: mago Umandam
| > To: msp430@yahoogroups.: com
| > Sent: Thursday, February 15, 2007 4:43 PM
| > Subject: Re: [msp430] Running at 3.3V @8 MHz
| >
| >
| > Hi,
| >
| > I am using 8MHz in 3.3V VCC and so far so good. I have not read from
| > datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says that 8
MHz can
| > be run at VCC greater than 2.8 Volts.
| >
| > mago
| >
| >
| > markcrowptm > wrote:
| > On my project, VCC can run from 1.8V to 3.3V and my oscillator is 8
| > MHz. Unfortunately I'm not in a position make any hardware changes.
| > The data sheet for the MSP430F1611 seems to indicate that the micro can
| > only run at 8 MHz if VCC is 3.6V. It indicates I can run with my VCC
| > issues at 4 MHz. Can I then simply program the clock control register
| > BCSCTL2 to divide MCLK by 2 and solve the problem?
| >
| > Thanks in advance.
| >
| > ---------------------------------
| > Need Mail bonding?
| > Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| >
| > >
egroups.com
| >
| >
| > Yahoo! Groups Links
| >
| >
| >
| >
|
|
|
|
|
|
|
| Yahoo! Groups Links
|
|
|
Yahoo! Groups Links
Reply by ●February 21, 20072007-02-21
Hi Bruno,
The processor isn't necessarily running at 12 MHz... The clock and
peripherals can run faster than the core. The specs say 10 MHz at 3V,
and the designer may have extrapolated for 12 MHz at 3.6V.
Most likely, the designer used the 12 MHz clock from the TUSB3410 to
save the cost of another crystal, and divided the clock before it
went to the core and/or peripherals.
Mike.
Kaelin Consulting
m...@kaelinconsulting.com
--- In m..., "Bruno Galli" wrote:
>
> Hi folks,
>
> not too much surprisingly even TI uses its chip beyond clock
specifications.
> If you look at the eZ430U development board you'll find an
MSP430F16x
> running at *12 MHz* at 3.6V! (It uses the 12 MHz because of the USB
chip.)
>
> I guess they had been very conservative to save on test budget.
>
> By the way, if you want to be strictly on specs keep in mind that
you have
> to run above 3.6V to have enough room for regulator IC overall
tolerance.
>
> Cheers,
> Bruno
>
> -----Messaggio Originale-----
> Da: "John C. Westmoreland, P.E."
> A:
> Data invio: venerd16 febbraio 2007 20.12
> Oggetto: Re: [msp430] Running at 3.3V @8 MHz
> | The design team on a project I worked on a couple of years ago
went round
> and round on this. Talked to TI several times, etc. This could be
part
> specific maybe; we were using the '1611 - since it had to pass FDA -
it was
> determined that we had to have Vcc be 3.6V.
> |
> | Sent via BlackBerry from T-Mobile
> |
> | -----Original Message-----
> | From: Hugh Molesworth
> | Date: Fri, 16 Feb 2007 10:45:49
> | To:m...
> | Subject: Re: [msp430] Running at 3.3V @8 MHz
> |
> | We asked TI about this and they stated that the frequency
restriction only
> | applies to the cpu core and not the crystal and associated
divider
> circuits.
> | Therefore use of 8MHz is within spec provided that the core and
> peripherals are
> | clocked - as suggested - at 4MHz.
> |
> | Hugh
> |
> | --- "John C. Westmoreland, P.E." > gineering.com> wrote:
> |
> | > The graph on the '1611 is clear regarding running at 8MHz - the
supply
> | > voltage needs to
> | > be 3.6VDC. If you want to run out of spec - that's up to you -
but if
> your
> | > project requires
> | > running in spec - then your Vcc needs to be 3.6VDC.
> | >
> | > Having said that - when hunting for speed paths - I've run
the '1611
> (rev. B)
> | > at 10MHz and 12MHz, but
> | > my Vcc was always 3.6VDC. The part will work with different
> | > voltages/frequencies - but it all depends
> | > if you want to be in spec - or if the project specifications
require (to
> be
> | > in spec...).
> | >
> | > Dividing MCLK by 2 as you have asked will work - that means the
core
> will be
> | > clocked at 4MHz - but
> | > there's still the question of the divider circuit and the I/F
circuitry
> to
> | > the oscillator running at 8MHz - which
> | > is still out of spec.
> | >
> | > Regards,
> | > John W.
> | >
> | >
> | > ----- Original Message -----
> | > From: mago Umandam
> | > To: msp430@yahoogroups.: com
> | > Sent: Thursday, February 15, 2007 4:43 PM
> | > Subject: Re: [msp430] Running at 3.3V @8 MHz
> | >
> | >
> | > Hi,
> | >
> | > I am using 8MHz in 3.3V VCC and so far so good. I have not read
from
> | > datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says
that 8
> MHz can
> | > be run at VCC greater than 2.8 Volts.
> | >
> | > mago
> | >
> | >
> | > markcrowptm > wrote:
> | > On my project, VCC can run from 1.8V to 3.3V and my oscillator
is 8
> | > MHz. Unfortunately I'm not in a position make any hardware
changes.
> | > The data sheet for the MSP430F1611 seems to indicate that the
micro can
> | > only run at 8 MHz if VCC is 3.6V. It indicates I can run with
my VCC
> | > issues at 4 MHz. Can I then simply program the clock control
register
> | > BCSCTL2 to divide MCLK by 2 and solve the problem?
> | >
> | > Thanks in advance.
> | >
> | > ---------------------------------
> | > Need Mail bonding?
> | > Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers
users.
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | > >
> egroups.com
> | >
> | >
> | > Yahoo! Groups Links
> | >
> | >
> | >
> | >
> |
> |
> |
> |
> |
> |
> |
> | Yahoo! Groups Links
> |
> |
> |
>
The processor isn't necessarily running at 12 MHz... The clock and
peripherals can run faster than the core. The specs say 10 MHz at 3V,
and the designer may have extrapolated for 12 MHz at 3.6V.
Most likely, the designer used the 12 MHz clock from the TUSB3410 to
save the cost of another crystal, and divided the clock before it
went to the core and/or peripherals.
Mike.
Kaelin Consulting
m...@kaelinconsulting.com
--- In m..., "Bruno Galli" wrote:
>
> Hi folks,
>
> not too much surprisingly even TI uses its chip beyond clock
specifications.
> If you look at the eZ430U development board you'll find an
MSP430F16x
> running at *12 MHz* at 3.6V! (It uses the 12 MHz because of the USB
chip.)
>
> I guess they had been very conservative to save on test budget.
>
> By the way, if you want to be strictly on specs keep in mind that
you have
> to run above 3.6V to have enough room for regulator IC overall
tolerance.
>
> Cheers,
> Bruno
>
> -----Messaggio Originale-----
> Da: "John C. Westmoreland, P.E."
> A:
> Data invio: venerd16 febbraio 2007 20.12
> Oggetto: Re: [msp430] Running at 3.3V @8 MHz
> | The design team on a project I worked on a couple of years ago
went round
> and round on this. Talked to TI several times, etc. This could be
part
> specific maybe; we were using the '1611 - since it had to pass FDA -
it was
> determined that we had to have Vcc be 3.6V.
> |
> | Sent via BlackBerry from T-Mobile
> |
> | -----Original Message-----
> | From: Hugh Molesworth
> | Date: Fri, 16 Feb 2007 10:45:49
> | To:m...
> | Subject: Re: [msp430] Running at 3.3V @8 MHz
> |
> | We asked TI about this and they stated that the frequency
restriction only
> | applies to the cpu core and not the crystal and associated
divider
> circuits.
> | Therefore use of 8MHz is within spec provided that the core and
> peripherals are
> | clocked - as suggested - at 4MHz.
> |
> | Hugh
> |
> | --- "John C. Westmoreland, P.E." > gineering.com> wrote:
> |
> | > The graph on the '1611 is clear regarding running at 8MHz - the
supply
> | > voltage needs to
> | > be 3.6VDC. If you want to run out of spec - that's up to you -
but if
> your
> | > project requires
> | > running in spec - then your Vcc needs to be 3.6VDC.
> | >
> | > Having said that - when hunting for speed paths - I've run
the '1611
> (rev. B)
> | > at 10MHz and 12MHz, but
> | > my Vcc was always 3.6VDC. The part will work with different
> | > voltages/frequencies - but it all depends
> | > if you want to be in spec - or if the project specifications
require (to
> be
> | > in spec...).
> | >
> | > Dividing MCLK by 2 as you have asked will work - that means the
core
> will be
> | > clocked at 4MHz - but
> | > there's still the question of the divider circuit and the I/F
circuitry
> to
> | > the oscillator running at 8MHz - which
> | > is still out of spec.
> | >
> | > Regards,
> | > John W.
> | >
> | >
> | > ----- Original Message -----
> | > From: mago Umandam
> | > To: msp430@yahoogroups.: com
> | > Sent: Thursday, February 15, 2007 4:43 PM
> | > Subject: Re: [msp430] Running at 3.3V @8 MHz
> | >
> | >
> | > Hi,
> | >
> | > I am using 8MHz in 3.3V VCC and so far so good. I have not read
from
> | > datasheet that 8 MHz requires 3.6V VCC. Datasheet page 24 says
that 8
> MHz can
> | > be run at VCC greater than 2.8 Volts.
> | >
> | > mago
> | >
> | >
> | > markcrowptm > wrote:
> | > On my project, VCC can run from 1.8V to 3.3V and my oscillator
is 8
> | > MHz. Unfortunately I'm not in a position make any hardware
changes.
> | > The data sheet for the MSP430F1611 seems to indicate that the
micro can
> | > only run at 8 MHz if VCC is 3.6V. It indicates I can run with
my VCC
> | > issues at 4 MHz. Can I then simply program the clock control
register
> | > BCSCTL2 to divide MCLK by 2 and solve the problem?
> | >
> | > Thanks in advance.
> | >
> | > ---------------------------------
> | > Need Mail bonding?
> | > Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers
users.
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | >
> | > >
> egroups.com
> | >
> | >
> | > Yahoo! Groups Links
> | >
> | >
> | >
> | >
> |
> |
> |
> |
> |
> |
> |
> | Yahoo! Groups Links
> |
> |
> |
>
