Maker Pro
Maker Pro

Expected life of 89c2051

A

Archer

Jan 1, 1970
0
Hi everybody

I have made a system using 89c2051 and it senses a port bit whether it
is high three times at intervals of 15 minutes and performs other
duties. This checking goes on day in and out.

My question is what is the reasonable life of the chip if this goes on
like this without any respite.

I have considered the question of using the idle provision but am
unable to use that effectively.

Hope some of you will have an answer for me.
 
T

TheM

Jan 1, 1970
0
Archer said:
Hi everybody

I have made a system using 89c2051 and it senses a port bit whether it
is high three times at intervals of 15 minutes and performs other
duties. This checking goes on day in and out.

My question is what is the reasonable life of the chip if this goes on
like this without any respite.

I have considered the question of using the idle provision but am
unable to use that effectively.

Hope some of you will have an answer for me.

I would think several decades at least, probably a lot more. Logically first chips to die would be
hot-running semiconductors with tiniest process (65um in modern cpus). A modern cpu would
probably be toasted in 20-50 years. I think Jim could say more on that matter.

M
 
M

Martin Griffith

Jan 1, 1970
0
Hi everybody

I have made a system using 89c2051 and it senses a port bit whether it
is high three times at intervals of 15 minutes and performs other
duties. This checking goes on day in and out.

My question is what is the reasonable life of the chip if this goes on
like this without any respite.

I have considered the question of using the idle provision but am
unable to use that effectively.

Hope some of you will have an answer for me.
see
http://www.atmel.com/dyn/resources/prod_documents/89C2051.pdf

I dont know how long the flash will retain it's memory, possibly
decades.
What will probably fail will be any electrolytics involved with the
circuit


martin
 
S

Spehro Pefhany

Jan 1, 1970
0
Hi everybody

I have made a system using 89c2051 and it senses a port bit whether it
is high three times at intervals of 15 minutes and performs other
duties. This checking goes on day in and out.

My question is what is the reasonable life of the chip if this goes on
like this without any respite.

I have considered the question of using the idle provision but am
unable to use that effectively.

Hope some of you will have an answer for me.

You need to consider Ta, and reliability will also probably be
affected if you're drawing/sourcing a bunch of current from port pins:

http://www.atmel.com/dyn/resources/prod_documents/89C2051.pdf



Best regards,
Spehro Pefhany
 
M

mpm

Jan 1, 1970
0
This checking goes on day in and out.

My question is what is the reasonable life of the chip if this goes on
like this without any respite.

Like they say.. "No moving parts!"

-mpm
 
J

Joerg

Jan 1, 1970
0
Martin said:
see
http://www.atmel.com/dyn/resources/prod_documents/89C2051.pdf

I dont know how long the flash will retain it's memory, possibly
decades.
What will probably fail will be any electrolytics involved with the
circuit

The failure process could be speeded up a bit by using tantalums. SCNR.

But yes, the chip itself will probably live almost forever unless it is
flying about in outer space or otherwise exposed to radiation. There is
stuff here in the house that contains a uC and has run for more than two
decades almost non-stop, only briefly interrupted by moves. If the flash
is a concern it could be re-flashed every five years or so.

If it's mission critical Sandy could consider using 3-5 uCs with
majority decision and inconsistency alert, similar to how it is done on
board of an aircraft. Or at least some kind of code redundancy in flash
plus regular CRC.
 
A

Archer

Jan 1, 1970
0
You need to consider Ta, and reliability will also probably be
affected if you're drawing/sourcing a bunch of current from port pins:

Many thanks to all of you who were kind to respond to my query and
clear my worry.

And the link to the Atmel paper was a great help.

Thanks again
 
M

mpm

Jan 1, 1970
0
Also, I know it wasn't exactly your question, but the 89C2051 has (or
is in the process of being?) replaced by the 89S2051. Which in this
case is a good thing because it gives you lower voltage options and in-
system program capability.

So in that sense, the C2051 is already "dead".
 
Top