Device data retention times

classic Classic list List threaded Threaded
6 messages Options
len
Reply | Threaded
Open this post in threaded view
|

Device data retention times

len
Hi All,

 

            I'm constantly doing projects with microcontrollers and
programmable logic devices.  I've been doing this long enough that I've got
devices that have "lost their memory."  Things that I designed and built
decades ago no longer function.  Over time flash will lose its mind, I guess
just like we do.

 

At some point in time that will happen to our lovely little radios.  From an
operator stand point we can re-flash the firmware as long as the boot code
is functional.  Does Elecraft have a provision in the radios that will allow
an update of the bootloader to keep it fresh, so to speak?  That is without
sending it back to the factory?

 

I know a lot of people don't keep stuff forever, but I pretty much do.   In
old radios we replace caps, and a few other parts that age.  With
programmable chips you need and ISP, unless there is another provision.  

 

73

 

Len

 

           

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Device data retention times

Lynn W. Taylor, WB6UUT
Last time I saw numbers on this, it was for the configuration memory on
a TAPR TNC-1.

The documentation said 100,000 writes, but it only counted if you
actually flipped a bit -- and it was per-bit, not per word or across the
entire device.

That means re-flashing with the same firmware shouldn't count at all,
since you're writing the same bits into the same words in storage.

I'm sure current parts are better.

73 -- Lynn

On 11/9/2016 12:07 PM, [hidden email] wrote:

> Hi All,
>
>  
>
>              I'm constantly doing projects with microcontrollers and
> programmable logic devices.  I've been doing this long enough that I've got
> devices that have "lost their memory."  Things that I designed and built
> decades ago no longer function.  Over time flash will lose its mind, I guess
> just like we do.
>
>  
>
> At some point in time that will happen to our lovely little radios.  From an
> operator stand point we can re-flash the firmware as long as the boot code
> is functional.  Does Elecraft have a provision in the radios that will allow
> an update of the bootloader to keep it fresh, so to speak?  That is without
> sending it back to the factory?
>
>  
>
> I know a lot of people don't keep stuff forever, but I pretty much do.   In
> old radios we replace caps, and a few other parts that age.  With
> programmable chips you need and ISP, unless there is another provision.
>
>  
>
> 73
>
>  
>
> Len
>
>  
>
>              
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Device data retention times

Jessie Oberreuter-2

      I believe Len is talking about the durability of the "most recent"
flash, not the number of times the device may be re-flashed.

      For rigs like the K1 and K2, the firmware is literally "burned" into
the chips by applying enough voltage to open internal fuses.  The data is
permanent.  For most re-programmable memory, however, the "chemistry" is
more volatile and the data can degrade over time.  This isn't a problem if
you can reprogram the chips, but the radio may well outlive the hardware
and software needed to perform such a reprogramming.

- kb7psg

On Wed, 9 Nov 2016, Lynn W. Taylor, WB6UUT wrote:

> Last time I saw numbers on this, it was for the configuration memory on a
> TAPR TNC-1.
>
> The documentation said 100,000 writes, but it only counted if you actually
> flipped a bit -- and it was per-bit, not per word or across the entire
> device.
>
> That means re-flashing with the same firmware shouldn't count at all, since
> you're writing the same bits into the same words in storage.
>
> I'm sure current parts are better.
>
> 73 -- Lynn
>
> On 11/9/2016 12:07 PM, [hidden email] wrote:
>> Hi All,
>>
>>
>>              I'm constantly doing projects with microcontrollers and
>> programmable logic devices.  I've been doing this long enough that I've got
>> devices that have "lost their memory."  Things that I designed and built
>> decades ago no longer function.  Over time flash will lose its mind, I
>> guess
>> just like we do.
>>
>>
>> At some point in time that will happen to our lovely little radios.  From
>> an
>> operator stand point we can re-flash the firmware as long as the boot code
>> is functional.  Does Elecraft have a provision in the radios that will
>> allow
>> an update of the bootloader to keep it fresh, so to speak?  That is without
>> sending it back to the factory?
>>
>>
>> I know a lot of people don't keep stuff forever, but I pretty much do.   In
>> old radios we replace caps, and a few other parts that age.  With
>> programmable chips you need and ISP, unless there is another provision.
>>
>>
>> 73
>>
>>
>> Len
>>
>>
>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Device data retention times

n6hz
In reply to this post by len
Hi Len,

That certainly used to be a problem with older windowed EPROMS.  I know that it's possible for FLASH cells to flip from cosmic radiation, and cells can go bad from repeated writing, but the bootroms we employ are loaded into write protected memory and are usually never re-written.  

There is typically no way to re-flash the bootrom without a suitable programmer (hardware).   In-field write protection of the bootrom is critical for a robustness.  

Kindly,

Paul
Reply | Threaded
Open this post in threaded view
|

Re: Device data retention times

Nr4c
Don't forget there is a small battery in your K3. How long have you had the radio????

Sent from my iPhone
...nr4c. bill


> On Nov 9, 2016, at 7:11 PM, n6hz <[hidden email]> wrote:
>
> Hi Len,
>
> That certainly used to be a problem with older windowed EPROMS.  I know that
> it's possible for FLASH cells to flip from cosmic radiation, and cells can
> go bad from repeated writing, but the bootroms we employ are loaded into
> write protected memory and are usually never re-written.  
>
> There is typically no way to re-flash the bootrom without a suitable
> programmer (hardware).   In-field write protection of the bootrom is
> critical for a robustness.  
>
> Kindly,
>
> Paul
>
>
>
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/Device-data-retention-times-tp7623895p7623901.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Device data retention times

n6hz
The battery is for the real time clock and has nothing to do with the flash on the MCU's.

-Paul