Can the BS-1 and BS-2 actually wear out?

[jukingeo]

Hello all,

I recently read that 'supposedly' the BS-1 and BS-2 can be programmed for around 1000 times and a current program will stay in memory for about 40 years.  Now I am not too concerned about the latter part.   But the first part I find interesting as I am using one Prop-2 right now as a test bed and I have been making a lot of program writes to it...I must be over 100 already.

I was just curious about this.  I am curious to know if anyone DID encounter problems after programming a BS-1 or BS-2 over 1000 times.

So does the "1000 time" programming limit line hold water, or no?

Thanx,
Geo

[livinlowe]

I had to research this one, but your 1000 number is a little off. It's 10,000,000  ;)

Check this document out for anything else:

http://www.parallax.com/dl/docs/prod/stamps/basicstampfaq.pdf

Shawn

[JonnyMac]

The number of reprogramming cycles varies per EEPROM, but let me put your mind at ease; with all the customers I support as well as my own projects, my controllers get reprogrammed far more than the norm and I have never worn one out.

Now... there is a command in the language, WRITE, that must be used with caution.  It will write one or more addresses in the EEPROM -- a useful feature for when you want to save data that won't be lost with a power cycle.  A few people have been a bit agressive and decided to write a piece of data every second.  This works.... for a few days, at some point the write cycle limit gets hit and the EEPROM is no longer useful.

My point is that you can wear out the EEPROM, but you have to do it deliberately; regular re-programming cycles will never create a problem for you.

[JonnyMac]

I just check the specs; the EEPROM in the Prop-1 (93LC66A) is guaranteed for one million (1,000,00) write cycles and the EEPROM in the Prop-2 (24LC16B) is guaranteed for ten million (10,000,000) write cycles.  Rest easy, friends, even if you reprogrammed your Prop-1 ten times a day, every day, it would take you almost 274 years to wear out the EEPROM -- make that 2740 years for the Prop-2.

[livinlowe]

Hmmm, maybe we need to tell Parallax to update their document. But, as you said Jon, no worries!!! Unless you start doing data logging. But that isn't HALLOWEEN!!!   8)

[jukingeo]

Now... there is a command in the language, WRITE, that must be used with caution.  It will write one or more addresses in the EEPROM -- a useful feature for when you want to save data that won't be lost with a power cycle.  A few people have been a bit agressive and decided to write a piece of data every second.  This works.... for a few days, at some point the write cycle limit gets hit and the EEPROM is no longer useful.

My point is that you can wear out the EEPROM, but you have to do it deliberately; regular re-programming cycles will never create a problem for you.

Hello Jon

Ok, good to know.  The document I read DID say 1000, not 10,000,000.  Perhaps that was a printing error.  But I also have heard about the WRITE command issue and that it isn't a good idea to use it.  Would you happen to know what it is with this command that is so 'destructive' ?   For me this whole issue does seem peculiar as I am not used to memory 'wearing out'.

Geo

[JonnyMac]

WRTE changes a location in the EEPROM, just like reprogramming, so if you put the program in a loop that was constantly using the WRTE instruction you could potentially create a problem.  I've been programming the BS1 since 1994 and never had this issue come up. 

[livinlowe]

Yeah, there are different values out there. It's a pretty big number though. The reason the write command can be destructive is if, say your taking readings from a thermometer. The write command allows you to 'save' the value to the memory. If you take your readings once a second, in an hour you will have written to the eeprom 60x60=360 times. In one day it would be 8640 times you wrote to your eeprom. So, if your doing data logging like this, you could write to the eeprom so many times that it could damage it. But, just downloading programs, you will be hard pressed to download a program a million times.  :o

[jukingeo]

Yeah, there are different values out there. It's a pretty big number though. The reason the write command can be destructive is if, say your taking readings from a thermometer. The write command allows you to 'save' the value to the memory. If you take your readings once a second, in an hour you will have written to the eeprom 60x60=360 times. In one day it would be 8640 times you wrote to your eeprom. So, if your doing data logging like this, you could write to the eeprom so many times that it could damage it. But, just downloading programs, you will be hard pressed to download a program a million times.  :o

I guess in a way that could be analogous to my dealings with old power amps.  If the controls are dirty on an old amp, you can blow it to kingdom come because of the full volume spikes that occur.

I guess turning those little transistors in the EEPROM on and off so many times could have the same effect in time.

Both Jon and you did make it more understandable that if you take regular readings every few seconds and write that into memory day and night over and over again that would considerably affect the life of the internal circuitry of the memory.

Well, it is good to know that 'normal' program rewrites are not going to be too much of an issue.  I guess I brought it up mainly because I will be using one Prop-2 for research and development purposes.  But at least now I know I don't have anything to worry about.

On top of that, if Jon hasn't had a problem yet...and he has been using these devices for years, I am good to go with that.

Geo