Al,
have you already test the reliability of the flash reprogrammation in
function of the flash timing generator frequency? Can it be reliably
push faster than the 476kHz.
 
Paul,
Have you test this when you have developed your USB fet?
 
Regards,
Fred
 

-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.9.1 - Release Date: 4/1/2005
 

No. I tend to run it around 400khz. In my testing it was OK at 476, but 
higher frequencies  required higher voltage levels, and lower 
temperatures. Although Nothing was out of spec regarding flash, my 
testing was quite hard, and I determined that for my systems I would run 
flash at no less than 3V, no more than 400khz. The fastest I tried 
running on a non-clock pushed system was around 800kHz. It didn't like 
that one bit. the fastest I had success at was around 615khz, but this 
proved to error prone. I figured that minor increases in speed were not 
really worth the risk. In very long term testing I saw some failures at 
470khz (8MHz/17) at higher temps, 2.7V. Not many but enough to stay my 
enthusiasm. treatyed gently the life is excellent. I have stuff in the 
field running 1 second updates of 1 word at a time, in a two segment 
rollover scheme (ie the segment is erased every 512 seconds) that has 
been re-writing its memory for 4 years now. The system does this for 8 
hours every day. so to date it has erased/re-written flash around 86,000 
times so far. The machines flag unrecoperable errors by storing the 
address of the failed word in dataflash segment B. So far none has 
reported 'half full". ie none of the 500 systems out there has yet 
logged 32 unrecoverable flash write failures. this has sttunned me, 
quite honestly. I expected more failures than this.

Al

Frederic Beaulieu wrote:

>Al,
>have you already test the reliability of the flash reprogrammation in
>function of the flash timing generator frequency? Can it be reliably
>push faster than the 476kHz.
> 
>Paul,
>Have you test this when you have developed your USB fet?
> 
>Regards,
>Fred
> 
>
>  
>

Fred,

> Al,
> have you already test the reliability of the flash reprogrammation in
> function of the flash timing generator frequency? Can it be reliably
> push faster than the 476kHz.

You can program it faster with provisos.  TI calibrate the DCO to about
470kHz when programming the flash.

> Have you test this when you have developed your
USB fet?

I have programmed some devices faster, but TI just say that device
reliability is reduced if programmed outside specification.

-- Paul.