EmbeddedRelated.com
Forums
Memfault Beyond the Launch

Question about FLASH timing generator

Started by lukasz_krysiewicz December 16, 2006
Hi everyone. I read a documentation "slau049e" and in the chapter about FLASH is note

"Write and erase operations are controlled by the flash timing generator shown
in Figure 5%u22123. The flash timing generator operating frequency, f(FTG), must be
in the range from ~ 257 kHz to ~ 476 kHz (see device-specific datasheet)."

I set this timing generator using register FCTL2 and combination of value FNx. For example I select flash controler clock source (FSEL_1) to MCLK equals 8MHz. I should divide this value using FCTL2, or no? I tested this problem, and in one program I set FSEL to MCLK equals 8MHz, and don't divided this value, in secone program I set FSEL to MCLK equals 8MHz, and divided this timing generator using FNx, In result I have ferquency ~470Hz.

Whitch program is correct, first or secnond. I saw that resuts in first and secont program in FLASH memory was correct. But first program (8MHz) was faster. Now I don't know with way is correct.

Thanks for advice.

----------------------------------
smieszne, muzyka, pilka, sexy, kibice, kino, ciekawe, extreme, kabaret
http://link.interia.pl/f19d4 - najlepsze filmy w intermecie

Beginning Microcontrollers with the MSP430

On 2006-12-17, lukasz_krysiewicz wrote:
>
> Hi everyone. I read a documentation "slau049e" and in the chapter about FLASH is note
>
> "Write and erase operations are controlled by the flash timing
> generator shown in Figure 5%u22123. The flash timing generator
> operating frequency, f(FTG), must be in the range from ~ 257
> kHz to ~ 476 kHz (see device-specific datasheet)."
>
> I set this timing generator using register FCTL2 and
> combination of value FNx. For example I select flash controler
> clock source (FSEL_1) to MCLK equals 8MHz. I should divide
> this value using FCTL2, or no?

Is 8MHz between 257KHz and 476KHz?

> I tested this problem, and in one program I set FSEL to MCLK
> equals 8MHz, and don't divided this value, in secone program I
> set FSEL to MCLK equals 8MHz, and divided this timing
> generator using FNx, In result I have ferquency ~470Hz.

I presume you mean 470kHz?

Is 470Khz between 257kHz and 476kHz?

> Whitch program is correct, first or secnond.

Answer the above two questions and then you tell us.

> I saw that resuts in first and secont program in FLASH memory
> was correct. But first program (8MHz) was faster. Now I don't
> know with way is correct.

The FTG clock must be between 257kHz nad 476kHz. You have two
programs, one with an FTG clock of 8MHz and one with 470kHz.
Which do you think is correct?

--
Grant Edwards
g...@visi.com
Someone I know uses 5V with his MSP430F1121A and it works. I use 3.3V
and it works too. I cannot say that his is wrong and I am right. He
is just willing to take more risk than I do. He knows what he is
doing.

When you use FCLK without the 257 to 476 kHz range, you are taking
more risk. There will be more percentage of devices that fail to
erase/program. There will be more percentage of the devices you
successfully erased/programmed that fail later. If the benefit out
weights the risk, you are doing it "right"!

--- In m..., lukasz_krysiewicz
wrote:
> Hi everyone. I read a documentation "slau049e" and in the chapter
about FLASH is note
>
> "Write and erase operations are controlled by the flash timing
generator shown
> in Figure 5%u22123. The flash timing generator operating frequency,
f(FTG), must be
> in the range from ~ 257 kHz to ~ 476 kHz (see device-specific
datasheet)."
>
> I set this timing generator using register FCTL2 and combination of
value FNx. For example I select flash controler clock source (FSEL_1)
to MCLK equals 8MHz. I should divide this value using FCTL2, or no? I
tested this problem, and in one program I set FSEL to MCLK equals
8MHz, and don't divided this value, in secone program I set FSEL to
MCLK equals 8MHz, and divided this timing generator using FNx, In
result I have ferquency ~470Hz.
>
> Whitch program is correct, first or secnond. I saw that resuts in
first and secont program in FLASH memory was correct. But first
program (8MHz) was faster. Now I don't know with way is correct.
>
> Thanks for advice.
>
> --------------------------------
--
> smieszne, muzyka, pilka, sexy, kibice, kino, ciekawe, extreme,
kabaret
> http://link.interia.pl/f19d4 - najlepsze filmy w intermecie
>
Thanks for advice.

I think too that greater ferquency (not in range 257 to 476 kHz) equal more program/erase errors.
But when I use for example ferquency 8MHz, programing FLASH memory is many times faster. It is very big plus for aplication.
----------------------------------
smieszne, muzyka, pilka, sexy, kibice, kino, ciekawe, extreme, kabaret
http://link.interia.pl/f19d4 - najlepsze filmy w intermecie
Hey, if you just want something that runs as fast as possible,
replace the flash write command with a single NOP. That way you
will waste much less time executing code that will not performing
the work you want.
Always design within the device specifications if you don't want to
deal with customer problems.

John,

PS Anyone can power an MSP430F1121 with 5 Volts...once.
--- In m..., lukasz_krysiewicz
wrote:
> Thanks for advice.
>
> I think too that greater ferquency (not in range 257 to 476 kHz)
equal more program/erase errors.
> But when I use for example ferquency 8MHz, programing FLASH memory
is many times faster. It is very big plus for aplication.
> -------------------------------
---
> smieszne, muzyka, pilka, sexy, kibice, kino, ciekawe, extreme,
kabaret
> http://link.interia.pl/f19d4 - najlepsze filmy w intermecie
>
--- In m..., "countryboy1234579"
wrote:
>
> PS Anyone can power an MSP430F1121 with 5 Volts...once.
>

Once? You can try many many times.
The chance that you get killed by 5V is small.

BTW what is the significance of that number you use?
12345679 would have been a more interesting number.
(Hint: Multiply that by 54 and chop the answer to three pieces.)
> (Hint: Multiply that by 54 and chop the answer to three pieces.)

We used to have fun with the early calculators in high school, coming up with some great ones.

Example : while looking at your calc display _upside down_ :
type in "707" -> makes LOL
type in "+"
type in "707" -> makes LOL
type in "="

Result (upside down) =>

LOL + LOL = hihi

Well, that was one of my contributions to science as a teenager :-)

Best Regards,
Kris

-----Original Message-----
From: m... [mailto:m...] On Behalf Of old_cow_yellow
Sent: Monday, 18 December 2006 1:30 PM
To: m...
Subject: [msp430] Re: Question about FLASH timing generator

--- In m..., "countryboy1234579"
wrote:
>
> PS Anyone can power an MSP430F1121 with 5 Volts...once.
>

Once? You can try many many times.
The chance that you get killed by 5V is small.

BTW what is the significance of that number you use?
12345679 would have been a more interesting number.
(Hint: Multiply that by 54 and chop the answer to three pieces.)

Yahoo! Groups Links
Oh, btw,

lol in Dutch means "fun" (as a noun).
Perhaps I should have pointed that out :-)

Best Regards,
Kris

-----Original Message-----
From: m... [mailto:m...] On Behalf Of Microbit
Sent: Monday, 18 December 2006 2:43 PM
To: m...
Subject: RE: [msp430] Re: Question about FLASH timing generator

> (Hint: Multiply that by 54 and chop the answer to three pieces.)

We used to have fun with the early calculators in high school, coming up with some great ones.

Example : while looking at your calc display _upside down_ :
type in "707" -> makes LOL
type in "+"
type in "707" -> makes LOL
type in "="

Result (upside down) =>

LOL + LOL = hihi

Well, that was one of my contributions to science as a teenager :-)

Best Regards,
Kris

-----Original Message-----
From: m... [mailto:m...] On Behalf Of old_cow_yellow
Sent: Monday, 18 December 2006 1:30 PM
To: m...
Subject: [msp430] Re: Question about FLASH timing generator

--- In m..., "countryboy1234579"
wrote:
>
> PS Anyone can power an MSP430F1121 with 5 Volts...once.
>

Once? You can try many many times.
The chance that you get killed by 5V is small.

BTW what is the significance of that number you use?
12345679 would have been a more interesting number.
(Hint: Multiply that by 54 and chop the answer to three pieces.)

Yahoo! Groups Links

Yahoo! Groups Links
Now I'm going to have to get one out and run the experiment. It's been
about four years, but I'm sure I lunched a few 1121's on a bad
regulator after replacing the MSP430Pxxx that IS specified up to 5
volts.

The significance of the 1234579 is that 1234567 was already taken in
yahoo's system. (Achem's Razor)
--- In m..., "old_cow_yellow"
wrote:
>
> --- In m..., "countryboy1234579"
> wrote:
> >
> > PS Anyone can power an MSP430F1121 with 5 Volts...once.
> > Once? You can try many many times.
> The chance that you get killed by 5V is small.
>
> BTW what is the significance of that number you use?
> 12345679 would have been a more interesting number.
> (Hint: Multiply that by 54 and chop the answer to three pieces.)
>
Now I'm going to have to go run the experiment. It's been about four
years, but I'm sure I lunched a few 1121's on a faulty regulator after
replacing an MSP430Pxxx that WAS spec'd up to 5.5 volts.

The 1234579 was selected because 1234567 was already taken in yahoo's
system. (Achem's Razor)

--- In m..., "old_cow_yellow"
wrote:
>
> --- In m..., "countryboy1234579"
> wrote:
> >
> > PS Anyone can power an MSP430F1121 with 5 Volts...once.
> > Once? You can try many many times.
> The chance that you get killed by 5V is small.
>
> BTW what is the significance of that number you use?
> 12345679 would have been a more interesting number.
> (Hint: Multiply that by 54 and chop the answer to three pieces.)
>

Memfault Beyond the Launch