Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conductor)

[QE]

Dear Coves.

can any one help me to point to the tool which can help me to predict the reliability of an Inductor (semi-conductor).

can weibull plot help ?

- QE.

 

[brahmaiah]

Re: Reliability Prediction

Dear Coves.

can any one help me to point to the tool which can help me to predict the reliability of an Inductor (semi-conductor).

can weibull plot help ?

- QE.

Reliability is not a measurable characteristic.Therfore it can be designated as a Special Characteristic.All specicial characteristics are achieved by controlling the inputs and the manufacturing p[rocess.So you have to follow following procedure:
a) Document a process
b)Validate by trials and controlling inputs
c)Monitor the manufacturing process strictly to conform to the validated process
d)Maintain a log of the process details.

There is no other shortcut available for your problem.
V.J.Brahmaiah

 

[Miner]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

QE,

Are you a manufacturer of or a purchaser of inductors? Are you trying to perform accelerated life testing to predict actual life, or trying to model the reliability of an assembly using the reliability of the inductor?

Please provide enough information for us to help you.

 

[QE]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

Miner

We are purchasing this inductor from the commodity market. I am intrested to predict failure rate which can help understand the liklihood of failure of this inductors and assist in a recall decession.

The inductor is used in a life saving device and it failed once and caused the device to fail.

QE.

QE,

Are you a manufacturer of or a purchaser of inductors? Are you trying to perform accelerated life testing to predict actual life, or trying to model the reliability of an assembly using the reliability of the inductor?

Please provide enough information for us to help you.

 

[Miner]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

Ideally, you get this direct from the supplier, but low cost suppliers may not test or have a clue.

Here are some resources that may help you.

• NASA paper on inductor reliability, including sources for inductor reliability data
• Reliability tests for inductors
• Inductor derating guidelines Derating reduces the stresses on the inductor to increase the reliability lifetime.

A point of caution. External source of inductor reliability typically provide an average failure rate. As with any distribution, this is a point estimate and actual failure rates may vary dramatically. This may be the result of supplier to supplier variation, lot to lot variation, etc.

 

[QE]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

Miner, appreciate your quick response.

1. can you please suggest if the paper is a published journal paper
2. CAN YOU ALSO SHARE THE SOURCE FOR THE TEST DATA SHEETS (PDF)

QE.

Ideally, you get this direct from the supplier, but low cost suppliers may not test or have a clue.

Here are some resources that may help you.

• NASA paper on inductor reliability, including sources for inductor reliability data
• Reliability tests for inductors
• Inductor derating guidelines Derating reduces the stresses on the inductor to increase the reliability lifetime.

A point of caution. External source of inductor reliability typically provide an average failure rate. As with any distribution, this is a point estimate and actual failure rates may vary dramatically. This may be the result of supplier to supplier variation, lot to lot variation, etc.

 

[Miner]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

I found the paper through this website for NASA's Electronic Parts and Packaging Program.

The reliability tests were from a supplier called Sun Spark Technology. They are meant to illustrate what one supplier does for reliability testing, not to endorse this supplier or the tests.

 

[Bev D]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

And as always generalized and theoretical models typically model what happens when nothing goes wrong at the manufacturer or the assembler, etc.

Real life is often messy.

A parallel path is to understand the exact causal mechanism of your failure and determine how prevalent that mechanism is in your inductor, assembly and/or use conditions. Remember that sometimes subcomponents fail when the causal factor lies in the assembly or usage. Stress - Strength.

Knowledge of the actual details combined with the theoretical models may help you understand if this specific failure mode is at a higher level than what can be tolerated.

Remember statistics can only help interpret what has happened and help interpret what will most likely happen if nothing changes. But physics always trumps statistics...

 

[w_grunfeld]

Re: Reliability Prediction - Predicting the Reliability of an Inductor (Semi-conducto

QE,
You got some good advice, however you asked the wrong question. I don't undersatnd why are you concerned with predicting the failure rate of inductors in a generic way, your concern should be your specifc part's probability of failure in practical terms. First off, most if not all part manufacurers data sheet specifically includes a disclaimer that they are NOT INTENDED FOR LIFE SUPPORTING APPLICATIONS.
What you should do is contact several manufacturers of the part and ask for their FIT data , based on actual life testing, not prediction. Also note that published data is by industry standard given at 60% confidence level, so ask them if they are willing to share the number for 90 0r 95 % confidence. Second you should look to use premium parts , such as parts that are screened per MIL-STD-883, or even better MIL SPEC PARTS, not ommercial common stock, you'll have to pay a premium price of course.Thirdly, follow the advice you received on design factors that you can control : derating, thermal management, HALT,etc.
Last but not least you should ESS screen (HASS ) your final product. All of thee are practical measures to increase reliability.