Validation of processes for production for Test Equipment

ariell3 · Jun 24, 2009

I have a customer who issued a major NC against our test equipment for not meeting the requirements of CFR 21 Sec. 820.75. We are a silicone manufacturer and use a Rheometer to measure product characteristics. These characteristics are neither toleranced, nor are they reported on the COA. The customer claims there is no protocol that outlines the acceptance criteria or a final report the details the results and conclusions.

This test equipment is calibrated by the manufacturer according to ASTM D5289 (Standard Test Method for Rubber Property - Vulcanizing Using Rotorless Cure Meter) as well as ISO 6502 (ibid.) The calibration standards used are NIST traceable and the ASTM standard addresses the correct preparation of the test samples and the precision and bias associated with the testing process. Additionally, there are other ASTM documents outlining the methodology and how the gage performance was calculated. This is a destructive test and a sample can only be tested on time.

I feel this NC is invalid and coming from TS the PPAP manual indicates that gage MSA studies are not required when standardized tests are used. I'm I correct in feeling this also applies to GMP?

Any thoughts?

Alex

Marc · Jun 24, 2009

Re: Validation of processes for production for Test Eqiupment

Can anyone help with this one?

BradM · Jun 24, 2009

Alex, I guess I'm still a bit confused at exactly what the customer's complaint is. Is it that there is no validation on the equipment, or the existing validation is insufficient?

Do you have anything in writing where you performed a qualification on the equipment?

Does your organization have a qualification requirement for all equipment, and this equipment was not qualified under that requirement?

I guess I'm trying to gauge where this is. I do think you will need something documented as a qualification on the equipment. However, not sure if the customer is looking for "thickness" :tg:

of the qualification, and not it's effectiveness.

Vicman · Jun 24, 2009

Hi, Ariell3
I think what your customer wants is something that is missing.
My understanding of a protocol is – a predefined written procedural method in the design and implementation of experiments.
Normally these protocols are written procedures in order to standardize a laboratory test method to ensure successful “replication” of the results in the same laboratory or others.
Procedures that include detail list of equipment and instruments, sometimes safety precautions and how to calculate the results and its interpretation on how to exclude some results if something undesirable happens like:
Not following calibration procedure, high ambient temperature, high humidity, low barometric pressure, turbulence flow instead of laminar flow that could affect the test results.
You may probably want to include the parameters and conditions in your final report to make the results more reliable.

Vicman

BradM · Jun 24, 2009

Vicman said:
Hi, Ariell3
I think what your customer wants is something that is missing.
My understanding of a protocol is – a predefined written procedural method in the design and implementation of experiments.
Normally these protocols are written procedures in order to standardize a laboratory test method to ensure successful “replication” of the results in the same laboratory or others.
Procedures that include detail list of equipment and instruments, sometimes safety precautions and how to calculate the results and its interpretation on how to exclude some results if something undesirable happens like:
Not following calibration procedure, high ambient temperature, high humidity, low barometric pressure, turbulence flow instead of laminar flow that could affect the test results.
You may probably want to include the parameters and conditions in your final report to make the results more reliable.

Vicman

Good writeup, Vicman. I concur with your comments, here. Hence the reason I was interested in finding out if there is any protocol, a bad protocol, etc.

If this equipment has been in use for a while, I would approach the validation differently than if it is brand new equipment. However, there needs to be some documentation on the equipment, and future scheduled requalifications and such.

Jim Wynne · Jun 24, 2009

ariell3 said:
We are a silicone manufacturer and use a Rheometer to measure product characteristics. These characteristics are neither toleranced, nor are they reported on the COA.

Why do you use the device? What do you do with the results of its use?

ariell3 · Jun 25, 2009

Thanks for the feedback! We will look into the way our test process is validated moving forward. However, in this case we just discovered the customer signed-off on the manfuacturing and testing process back in 2003. Which makes us wonder why the validation data was sufficient back then and not now?

This is a standard ASTM test for elastomeric materials and we use the data only for internal adjustments to the product. The results are recorded and not generally shared with the customer since they perform similar tests upon receipt of the material.

Thanks,

Alex

BradM · Jun 25, 2009

ariell3 said:
Thanks for the feedback! We will look into the way our test process is validated moving forward. However, in this case we just discovered the customer signed-off on the manfuacturing and testing process back in 2003. Which makes us wonder why the validation data was sufficient back then and not now?

This is a standard ASTM test for elastomeric materials and we use the data only for internal adjustments to the product. The results are recorded and not generally shared with the customer since they perform similar tests upon receipt of the material.

Thanks,

Alex

Thanks for the followup, Alex.

In this case, I am still at a loss as to exactly what the customer is having an issue with. What exactly are they saying is not appropriate; not right; unacceptable, etc.?

Juan Dude · Jun 25, 2009

ariell3 said:
I have a customer who issued a major NC against our test equipment for not meeting the requirements of CFR 21 Sec. 820.75. We are a silicone manufacturer and use a Rheometer to measure product characteristics. These characteristics are neither toleranced, nor are they reported on the COA. The customer claims there is no protocol that outlines the acceptance criteria or a final report the details the results and conclusions.

This test equipment is calibrated by the manufacturer according to ASTM D5289 (Standard Test Method for Rubber Property - Vulcanizing Using Rotorless Cure Meter) as well as ISO 6502 (ibid.) The calibration standards used are NIST traceable and the ASTM standard addresses the correct preparation of the test samples and the precision and bias associated with the testing process. Additionally, there are other ASTM documents outlining the methodology and how the gage performance was calculated. This is a destructive test and a sample can only be tested on time.

I feel this NC is invalid and coming from TS the PPAP manual indicates that gage MSA studies are not required when standardized tests are used. I'm I correct in feeling this also applies to GMP?

Any thoughts?

Alex

You mention your NC is against 820.75 which deals with Process validation. Process Validation deals with Validating your Process to be capable of constantly producing product that meets all requirements. Even though your equipment may meet all calibration requirements, this does not mean that your process is validated, which is what your NC seems to be against.

MIREGMGR · Jun 26, 2009

Per your explanation, it sounds as if your product has a significant potential process variability that your production people "know how to" control via this equipment and test method. "No tolerances" means the process is not formally controlled.

Per current FDA interpretations, your customer has no choice but to require you to at least have formally controlled processes. If your materials potentially are significant to their products' safety and effectiveness, they probably must require you to validate their processing.

As to the 2003 signoff, they have no choice as to how they proceed on a forward-going basis, so whatever happened six years ago is irrelevant.

Validation of processes for production for Test Equipment

ariell3

Marc

Fully vaccinated are you?

BradM

Vicman

BradM

Jim Wynne

ariell3

BradM

Juan Dude

MIREGMGR

Similar threads