Page 1 of 1

Long term reliability test

Posted: 06 October 2008, 18:02 PM
by Don_Kirby
After getting distracted for the past few months, I've returned to a ZBasic project that has been running on the bench for quite a while now. As a testament to the ZBasic platform, and the underlying Atmel hardware, the project has been running for a little over 5000 hours (214 days to be precise) continuously. This number is only as accurate as the ZX-24n clock source, though, so this isn't a scientifically accurate test, but it does go to show how well these devices can hold up.

One of the reasons for this test was to verify that the circular buffer used to keep track of the total running time would perform it's job of wear leveling the EEPROM properly. I have yet to see a single error. Mathematically, the EEPROM should last over 300,000 hours.

-Don

Re: Long term reliability test

Posted: 06 October 2008, 19:11 PM
by dkinzer
Don_Kirby wrote:Mathematically, the EEPROM should last over 300,000 hours.
But that's only 34 years. What are you going to do then? ;)

Posted: 06 October 2008, 20:54 PM
by stevech
I have a ZX24 that runs 24/7 - monitors garage door IR sensors for cars coming in. Turns on green-yellow-red superbright LEDs to tell the drivers (2 car garage) when the rear end has cleared the sensors. Ignores most non-car things going through the IR beams.

It logs to EEPROM the date/time each time a car enters/leaves. Uses a circular buffer with an even-wear scheme as mentioned. Sends logged events on serial port to PC.

Has run for a bit over a year now.

Re: Long term reliability test

Posted: 07 October 2008, 14:57 PM
by Don_Kirby
dkinzer wrote:But that's only 34 years. What are you going to do then? ;)
Retire.