IC test Specification - absolute maximum ratings

[Joe G \(Home\)]

Hi All,

I have an IC ... say something like a RS232 level converter.

In the specification of the "Limiting Values" one of the input pins (IN) is
specified as -27v to +40v.

There is no test conditions specified with this limiting value.

QUESTIONS

Q1
Under normal operating conditions (normal Vdd supply range = 12v with inputs
output terminated correctly)
Would you expect the IC to FAIL if IN= 35v DC is applied?

Would you expect the device to not operate or shut down..... but recover if
powered off and IN = normal 12v

Q2
How are IC's tested to meet their maximum ratings?

Q3
Can users of the silicon prove IC's meet their limiting values or absolute
maximum ratings?


Thanks in advance.

Regards
JG

 

[Jon]

Joe,

The absolute maximum ratings are usually interpreted to mean "Maximum
stress, beyond which the device may be degraded". The device is not
necessarily guaranteed to operate at anything beyond the recommended
temperature, supply or input values. If these ratings are exceed, you
might expect (in addition to catstrophic failure), degraded
performance, such is increased offsets, gain out of spec, output
voltage out of spec, decreased reliability, etc.
Regards,
Jon

 

[Rene Tschaggelar]

Hi All,

I have an IC ... say something like a RS232 level converter.

In the specification of the "Limiting Values" one of the input pins (IN) is
specified as -27v to +40v.

There is no test conditions specified with this limiting value.

QUESTIONS

Q1
Under normal operating conditions (normal Vdd supply range = 12v with inputs
output terminated correctly)
Would you expect the IC to FAIL if IN= 35v DC is applied?

Would you expect the device to not operate or shut down..... but recover if
powered off and IN = normal 12v

Q2
How are IC's tested to meet their maximum ratings?

Q3
Can users of the silicon prove IC's meet their limiting values or absolute
maximum ratings?

You must be physicist. None else sane would even try
to come close to those values specified under absolute
maximum ratings.
These are the people considering absolute maximum
ratings as hints that can easily be exceeded by 10 to
50% under all conditions.
And all the others that stick to the "normal supply
range" are considered fluffy, weak, and such ?

Rene

 

[Jim Thompson]

You must be physicist. None else sane would even try
to come close to those values specified under absolute
maximum ratings.
These are the people considering absolute maximum
ratings as hints that can easily be exceeded by 10 to
50% under all conditions.
And all the others that stick to the "normal supply
range" are considered fluffy, weak, and such ?

Rene

Actually a LOT of ABSOLUTE MAXIMUM ratings are pure BS. Put there to
avoid testing at manufacture.

...Jim Thompson

 

[Ian Stirling]

Hi All,

I have an IC ... say something like a RS232 level converter.

In the specification of the "Limiting Values" one of the input pins (IN) is
specified as -27v to +40v.

There is no test conditions specified with this limiting value.

QUESTIONS

Q1
Under normal operating conditions (normal Vdd supply range = 12v with inputs
output terminated correctly)
Would you expect the IC to FAIL if IN= 35v DC is applied?

Would you expect the device to not operate or shut down..... but recover if
powered off and IN = normal 12v

I would expect its operation to be undefined - it may overheat and
need to cool down.
Unless specific operating limits are made, I'd assume input of 40V
was safe, for the short term in any condition.

Q2
How are IC's tested to meet their maximum ratings?

They often are not - it's a matter of design.

Q3
Can users of the silicon prove IC's meet their limiting values or absolute
maximum ratings?

Well, you can test each one at its maximum voltage, to see if it explodes.
In some designs, this may cause irreversable long-term damage that may not
be immediately apparent.

 

[Rene Tschaggelar]

Actually a LOT of ABSOLUTE MAXIMUM ratings are pure BS. Put there to
avoid testing at manufacture.

But I assume they are on the safe side ? If a customer
has failing parts far inside the safety margins,
there possibly is room for a liability case.

Rene