R8C/Tiny flash re-programming limits

Hello Newsgroup,

In the data sheets of the Renesas R8C controllers some versions are listed with lower re-programming cycles for the flash than others. Some can be re-programmed 100 times, others 1000 times. Both numbers seem kind of low. In your experience, is that just an extremely conservative spec or do they really fail when these numbers are significantly exceeded?

Regards, Joerg

formatting link

Reply to
Joerg
Loading thread data ...

This is something that used to be asked a lot on GNUH8 mailing list about H8 series devices, and after announcement from Hitachi (at the time) was down to being a WORST case guaranteed across all temperature ranges, voltage ranges and maintaining a 10 year retention period. What was also deterministic in this was the process used 0.35um etc..

Related H8 information

FAQ

copy of change notice

There should be something similar around for R8 and R16, but may take some finding.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul Carpenter

Yes, when I asked the Renesas FAE's they told me that they made experiments and gave up after 100'000 cycles.

My developement board I used to program surely got more than 10'000 cycles - no problem encountered ever.

Markus

Reply to
Markus Zingg

Hello Paul,

Thanks, that clarifies it. I had searched for that too narrowly (in connection with R8C) and came up dry.

I still find it a bit strange that one device is listed at 100 times while the next in line is listed at 100 times. But that might just be historical. And as you said in the above FAQ Renesas doesn't want to be sued. Kind of like the printing on the side of our wheel barrow: "Not for highway use".

Regards, Joerg

formatting link

Reply to
Joerg

Hello Markus,

That is good to know. Especially in experiments with large boards this is important, the kind of boards you can't throw out because duplicating them would be a lot of work. You can easily exceed 100 times on a prototype board and the worst case would be to chase a perceived coding error and not seeing that part of the flash has became corrupt. But it seems that is unlikely.

Regards, Joerg

formatting link

Reply to
Joerg

I'm using the H8 series and I've never seen one fail, though I think there's probably only one unit that's exceeded the 100 cycle limit.

Like all specs, it's probably a bit conservative, but I wouldn't plan on being able to exceed the specs.

--
Grant Edwards                   grante             Yow!  I feel like a wet
                                  at               parking meter on Darvon!
 Click to see the full signature
Reply to
Grant Edwards

You do know about cheksums, right?

--
Grant Edwards                   grante             Yow!  In Newark the
                                  at               laundromats are open 24
 Click to see the full signature
Reply to
Grant Edwards

Hello Grant,

Yes. My concern was more about bits becoming flaky or where that flakiness only happens upon ADC clock or when a heavily loaded output turns on etc. That's hard to catch.

I have done lots of EMI susceptibility and ESD tests. Often a digital system will become rather non-digital or die very gradually.

Regards, Joerg

formatting link

Reply to
Joerg

Hello Grant,

In production we certainly won't. I never go above data sheet ratings. The concern would be for the prototyping phase of large boards that have a uC on there. 100 cycles would certainly be exceeded on those. It's a pain when the 'golden unit' croaks an hour before a demo and all the dignitaries are already beginning to gather in the board room.

Regards, Joerg

formatting link

Reply to
Joerg

I asume you mean next is listed at 1000 times.

Some of that will be down to newer processes and smaller geometry (0.35um to

0.18um etc.) also down to what testing had been done. Who wants to go back to an old design and do the complicated testing of an old design at 60 degrees! Especially if you want to examine the die under a microscope or do other tests to judge the data retention period.

A bit like that and not wanting to spend time and money retesting devices at different temperatures etc..

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul Carpenter

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.