Is Wafer Probe (On Wafer Test) a Laboratory? Semiconductor Industry

[dlsquality]

To My Semiconductor Friends:

Does a wafer probe (or on wafer test) operation qualify as a laboratory as defined in ISO/TS 16949?

 

[Jen Kirley]

Re: Is Wafer Probe a Laboratory?

To My Semiconductor Friends:

Does a wafer probe (or on wafer test) operation qualify as a laboratory as defined in ISO/TS 16949?

Welcome to The Cove!

I have seen such inspection/reliability analysis activities listed in alignment with 7.6.3.1 although most often I expect people to consider it a calibration control element because of the note commenting about ISO/IEC 17025.

 

[vanputten]

Hello Dlsquality:

I reviewed the defintion for "laboratory" which is 3.1.4 in T 16949. Then I reviewed the definition for laboratory in the Semiconductor Supplement, which is on page 11 of 27. clause 3.1.5S.

After that, I re-read 7.6.3, 7.6.3S, and 7.6.3.1S.

I have some ideas but would like to hear your thoughts first.

7.6.3 Internal Laboratory applies to test, inspection, or calibration services.

To Jennifer: By the way, IEC/ISO 17025 is applicable to all organizations performing tests and/or calibrations. The scope of IEC/ISO 17025 is greater than just calibration. These tests and/or calibrations can be a component of inspection and product certification.

Regards,

Dirk

 

[dlsquality]

Based on the semiconductor supplement definition of laboratory (i.e. laboratories are those facilities that perform reliability, qualification or durability testing to the requirements of a customer documented specification), it doesn't appear that the wafer probe area would fall under that definition. Right?

 

[qualityboi]

It's not a lab. Reliability testing is usually done in QA. Probe is more of a final electrical test at wafer level. I don't think its the intent of the standard to call any process that qualifies or tests product a lab nor does either standard state what you do and don't call a lab, it only states the requirements for a lab.

Here is an example of what the quality manual states for AMI semiconductor who is registered to the 16949 and it passes just fine...

9.6.1 Laboratory Requirements

The responsibility for performing test and calibration to meet the requirements of AEC-Q100 is under the control of our internal qualified laboratory as well as under commercial independent qualified laboratories or external best in the market in the given country.
9.6.1.1 Internal Laboratory

All laboratories in AMIS are engineering laboratories without any certification rights and can be used only for verification of the product performance.

Laboratory scopes and flow are available for laboratories that perform repetitive activities that need safety regulations, related to validation processes and customer feedback.
9.6.1.2 External Laboratory

For calibration of test and measurement equipment, accredited laboratories (if possible commercially independent) are used. Calibration can be done either by calibration labs (external calibrations) or per AMIS #5500103, Product Realization and Process System Requirements.

 

[Hershal]

Your operation may potentially be considered a lab.....not a hard and fast truth however.....

What are you testing for and under what criteria? If you are checking functionality, then an argument can be made that the operation is a lab, whether at the end of the production line or in a separate room.

If a lab, then consider whether to seek accreditation under ANS/ISO/IEC 17025.

Just some thoughts.

Hershal

 

[vanputten]

Hello All:

I agree with what everyone has posted. I think if someone wanted to play hardball, they could make a case that wafer sort is a lab based on the defintions and the wording of TS 16949 and the Semiconductor Supplement. I hope that the intent of these requirements is as is stated in this thread; wafer sort is NOT performed in a "lab."

We don't consider electrical test during wafer fabrication, wafer sort, or IC electrical test (final test) a "Laboratory" or lab function. We consider these to be inprocess / final inspections. We do consider failure analysis and product qualification to be performed in a "lab" and apply the 7.6.3 requirements to these areas.

It is my expereince that other semiconductor manufactueres do the same thing.

I think Dlsquality has asked an excellent question particularily for the semiconductor industry.

Thank you for asking the quesiton!

Regards,

Dirk

 

[vanputten]

Forgot to add....

Qualityboi stated "nor does either standard state what you do and don't call a lab,"

I think 3.1.4 of TS and 3.1.5S of the Semiconductor Supplement do state what is done in a lab. The definitions do not state what is a lab (room, area, facility, etc.) but they do state what is done in a lab. Based on the defintions, a lab is any location where these activites are performed, imho.

Thankfully the Semiconductor Supplement seems to reduce the scope of activities in a lab to "reliability, qualification, and durability testing" which matches what is discussed in this thread.

Regards,

Dirk

 

[qualityboi]

Well I think you know where I am coming from, which is reducing the scope to what is reasonable. We don't call the Probe and Test processes a lab at our semi-con facility, nor do we apply those standards. I don't think anyone else in the industry does either (two I know of do not and are certified), but who knows you could get a 3rd party auditor that thinks anywhere qualification, reliability and testing intersects the process should be called a lab. Good topic!