Searching \ for 'Interfacing PICs to large (8 megabit) FLASH memori' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/memory.htm?key=flash
Search entire site for: 'Interfacing PICs to large (8 megabit) FLASH memori'.

Truncated match.
PICList Thread
'Interfacing PICs to large (8 megabit) FLASH memori'
1999\05\27@214553 by tewart Greenhill

flavicon
picon face
Hi Folks,

I'm new to the PIC list and am back into PIC development mode after an absence
of a few years.

I'm working on a data logging application using a PIC16F84. I need a
low-power design with a large memory. These loggers may be recording for
several years under battery power before being interrogated.

Currently, I'm looking at a couple of 8 megabit serial flash memory
devices: the NexFlash NX25F080A and the ATMEL AT45DB081. Both operate at
2.7 volts, have a 4-wire serial interface and dual SRAM sector buffers
on-chip. This is important because it allows sectors to be assembled
on-chip before being programmed into FLASH. This style of memory requires
an entire sector to be written at once, and there is not enough RAM in a
PIC to hold the data.

The ATMEL part looks a little more flexible, since both sector buffers can
be programmed directly to FLASH. The NexFlash part has separate PROGRAM and
SRAM buffers; although data can be copied between the two, only the PROGRAM
buffer can be written to FLASH. However, the decider for me is the power
consumption. At 2 uA standby, the ATMEL part requires twice the power of
the NexFlash part (1 uA).

Does anyone have any experience (positive or negative) with either of these
parts? Are there any other devices that I should consider using?

Thanks,
- Stewart Greenhill

1999\05\27@215849 by Peter Grey

picon face
At 09:45 PM 27/05/99 -0400, you wrote:
I am using 24LC256 from Microchip. I use 8 of them and they have a 1uA
standby current. I am using them in a data logging application to give me 2
megabits. However, I have access to power and am using a 16C74 as I have a
lot of other functions to take care of (LCD, 4x4 keypad, RTC, RS232 link,
alarms, radio link)

good luck


Peter Grey

>Hi Folks,
>
>I'm new to the PIC list and am back into PIC development mode after an
absence
{Quote hidden}

1999\05\27@221604 by Jim Paul

picon face
Stewart,

I have samples of the 5 volt version of the 2 and 4 Megabit parts by ATMEL.
I looked at several parts before deciding on these particular parts and this
particular manufacturer.   I chose these mostly because they would be easy
to interface (They are serial devices),  because I could get them in the
package and voltage I wanted, they had the capacity I needed, past
experience
with ATMEL parts, and last but not least, cost.   Even in quantities of 1 or
2
pieces, they are only about 3 or 4 dollars each.  This  is acceptable to me.
The power consumption in my application is only of moderate concern, so I
can't give much advice in that area.   Also, with these parts you can write
to
the buffer until it id full, and then send a command to write this buffer
full of
data to main FLASH memory.  After issuance of this write command, the
operation is invisible to the user.  In the meantime, you can write to the
second buffer while the write of the first buffer is going on in the
background.
This allows you to continue datalogging without missing any data.  Once this
second buffer gets full, you can write it to main FLASH, and go back to
using
the first buffer again.  Another point of interest to me is that the page
size of
these memories is 264 bytes.  These 8 extra bytes can be used to store
checksum data or other information you may want to include.  I think from a
cost/performance/feature point of view, these parts are hard to beat.  But,
only
you can decide if they do or do not meet the specifications you have set
forth
as being important in your application.

I hope this helps you.  If nothing else, I hope to have given you some
points to
think about.   If you need any more info on these parts, let me know and if
I can,
I'll help in whatever way I can.


Regards and good luck,


Jim




{Original Message removed}

1999\05\27@224327 by Antonio L Benci

flavicon
picon face
Here's a suggestion. Now i don't know what the operating current for
these devices is BUT how about powering up the flash memory from a PORT
pin. The 16F84 can supply up to 20mA/pin. When not being accessed the
port pin is set LOW, no power is consumed. When the 16F84 is placed into
sleep mode the previous pin states are retained... The only current
being used is the sleep current for the 16F84...

I have used this method for the common garden variety serial eeprom's
with no problems...

Stewart Greenhill wrote:
{Quote hidden}

--
******************************************************
* Antonio (Nino) Benci                               *
* Electronic Services Manager                        *
* Monash University - Dept of Physics                *
* Wellington Rd, Clayton. 3168                       *
* Victoria, Australia.                               *
* TEL    - 61 3 9905 3649, FAX - 61 3 9905 3637      *
* Mobile - 0414 764 763 (private and ah only)        *
* EMAIL  - spam_OUTnino.benciTakeThisOuTspamsci.monash.edu.au (work)       *
*        - .....fleatechKILLspamspam@spam@excite.com (private)             *
* WWW    - http://www.physics.monash.edu.au/                *
******************************************************

1999\05\27@225920 by Antonio L Benci

flavicon
picon face
Here's a suggestion. Now i don't know what the operating current for
these devices is BUT how about powering up the flash memory from a PORT
pin. The 16F84 can supply up to 20mA/pin. When not being accessed the
port pin is set LOW, no power is consumed. When the 16F84 is placed into
sleep mode the previous pin states are retained... The only current
being used is the sleep current for the 16F84...

I have used this method for the common garden variety serial eeprom's
with no problems...

Stewart Greenhill wrote:
{Quote hidden}

Nino
--
******************************************************
* Antonio (Nino) Benci                               *
* Electronic Services Manager                        *
* Monash University - Dept of Physics                *
* Wellington Rd, Clayton. 3168                       *
* Victoria, Australia.                               *
* TEL    - 61 3 9905 3649, FAX - 61 3 9905 3637      *
* Mobile - 0414 764 763 (private and ah only)        *
* EMAIL  - nino.bencispamKILLspamsci.monash.edu.au (work)       *
*        - .....fleatechKILLspamspam.....excite.com (private)             *
* WWW    - http://www.physics.monash.edu.au/                *
******************************************************

1999\05\27@232658 by Mike Keitz

picon face
On Fri, 28 May 1999 12:42:18 +1000 Antonio L Benci
<EraseMENino.Bencispam_OUTspamTakeThisOuTSCI.MONASH.EDU.AU> writes:
>Here's a suggestion. Now i don't know what the operating current for
>these devices is BUT how about powering up the flash memory from a
>PORT
>pin. The 16F84 can supply up to 20mA/pin. When not being accessed the
>port pin is set LOW, no power is consumed.

Overall it's not a bad idea.  For an application that is mostly or
exclusively reading it would be ideal.  But when the power is cut off,
the data in the SRAM part of the flash chip will be lost too.  So if the
PIC doesn't have enough SRAM to hold up to a whole block of data before
moving it to flash, there is a problem.

It would be possible to write less than a full block at a time by reading
a block from flash into the SRAM, adding a few bytes of new data, then
storing it back to the same block of flash before turning the power off.
But the extra write cycles will probably use more power than just leaving
the flash chip in standby and holding the partial block in the SRAM.

___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]

1999\05\28@010905 by Antonio L Benci

flavicon
picon face
Ok, the SRAM will be lost BUT for do you really want to keep logged data
in SRAM for an extended period. It's safer to just store is on
completeion of one session...

Mike Keitz wrote:
{Quote hidden}

To reduce the write cycles the software would only write once a complete
block is ready... This would mean that it may be some time for a
complete block to be ready but once it is so then store and turn off...

>
> ___________________________________________________________________
> You don't need to buy Internet access to use free Internet e-mail.
> Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
> or call Juno at (800) 654-JUNO [654-5866]

--
******************************************************
* Antonio (Nino) Benci                               *
* Electronic Services Manager                        *
* Monash University - Dept of Physics                *
* Wellington Rd, Clayton. 3168                       *
* Victoria, Australia.                               *
* TEL    - 61 3 9905 3649, FAX - 61 3 9905 3637      *
* Mobile - 0414 764 763 (private and ah only)        *
* EMAIL  - @spam@nino.benciKILLspamspamsci.monash.edu.au (work)       *
*        - KILLspamfleatechKILLspamspamexcite.com (private)             *
* WWW    - http://www.physics.monash.edu.au/                *
******************************************************

1999\05\28@081547 by Andy Kunz

flavicon
face
>Does anyone have any experience (positive or negative) with either of these
>parts? Are there any other devices that I should consider using?

You can write to the Atmel's RAM buffer at any speed you desire.  I use it
to log data myself (but then, I generate a 256-byte page (no, I don't fill
it completely for addressing reasons) in 16 mS, then flush to flash.  If
you  do your algorithm smartly, you can be writing to one buffer while
burning the other.

I recommend the AT part because I've used it and I'm happy.

Andy

==================================================================
  Montana Design Tech Support - http://www.montanadesign.com
==================================================================


'Interfacing PICs to large (8 megabit) FLASH memori'
1999\06\07@230558 by tewart Greenhill
flavicon
picon face
Hi Folks,

Many thanks to those who replied to my PIC/FLASH question.

Both Jim Paul <RemoveMEjamespTakeThisOuTspamINTERTEX.NET> and Andy Kunz
<spamBeGonesupportspamBeGonespamMONTANADESIGN.COM> said they were happy with the Atmel FLASH part.

When I tried to get hold of some NexFlash parts in Australia, neither of
the distributors had even heard of the company (NexFlash).  The best answer
I got from them was to try an outlet in Taiwan. I can source NexFlash parts
from the US, but they're not cheap. On the other hand, the Atmel parts are
available as a stock item from a local electronics distributor at ~$12
($US7) for one-off quantities. Admittedly, its not quite as good as Jim's
price ($3-4 for the 5v parts), but its certainly a better starting point.

Antonio L Benci <TakeThisOuTNino.BenciEraseMEspamspam_OUTSCI.MONASH.EDU.AU> pointed out that its
possible to power these parts from a port pin and switch them off when not
required, thereby reducing the power consumption. I've found this to be a
good technique in the past, and will probably use it to save power on the
A/D converters and the analog system. As Mike Keitz <RemoveMEmkeitzspamTakeThisOuTJUNO.COM>
pointed out, powering down the FLASH parts causes the SRAM contents to be
lost. In my particular case this is a problem, since the devices are
sampling fairly slowly and may take hours (possibly days) to collect enough
data to fill a FLASH sector. Another issue is that if the FLASH is powered
down, the I/O lines (serial in, serial out, clock) can't be used to talk to
other devices (without exceeding the maximum rating of Vcc+0.5V for any
device pin). This may or may not be an issue, depending on how many port
pins I eventually need.

Peter Grey <martechEraseMEspam.....OZEMAIL.COM.AU> said that he uses 8 x 24LC256s to get 2
megabits storage for his logger. This is another possibility that I
considered. These parts allow bytes to be individually programmed (no
sector requirements) and so could be powered up/down and programmed as
required. If larger density serial flash memories were not available, I
might have chosen this option.

Once again, thanks everyone for your input. On balance, the ATMEL part
looks a little more attractive (especially with my difficulties obtaining
NexFlash supplies). I'll have to look carefully at my power budget once I
have a complete design.

Cheers,
- Stewart

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...