BRC 2.8.1 - Metal Detector Critical Limits

I'm hoping someone can help with BRC 2.8.1. This clause discusses the establishment of critical limits for CCP's. In my case the CCP is metal dectection with critical limits of 1.5 Fe, 2.0 Nfe, and 2.5 SS, which we monitor on an ongoing basis with test pieces of the same size as the critical limits. Our BRC auditor has challenged us to provide rational on the establishment of the critical limits. Is anyone aware of any guidance or studies that have been performed on selecting critical limits for metal detectors?

Do you have any customer standard that states what you are required to meet? That is generally the easiest evidence of why you selected your limits.

In most cases it is either going to be what your customer wants or the limit that your detector can find. We used a combination of the two for our BRC audit in March, "this is the best our detectors can do and it is below the limit expectations of our customers".

Not many regulatory bodies or governments are going to commit to a limit because it would pull them into the middle of lawsuits!

Thanks for the response. I'm glad to hear it's not as difficult as I interrupted the clause to be.

I use the FDA guidance as a help for the validation. For adults, they indicate <7mm to be safe from a review of incidents. It's not law in the UK but this has been used successfully by me in several factories to validate piece size. The FDA are a respected organisation wherever you go so an auditor couldn't really complain with you using this:

http://www.fda.gov/I...l/ucm074554.htm

Just realised I was having a blonde baldy moment and as you're in the US, an auditor definitely shouldn't mind you using this!!!

GMO,

Thanks for the link.

I actually dont think is what they referring to. We were presented with a similiar question awhile back and were then directed to the following AIB Enhancement:


e1.15.a Validation studies are conduced and documented to demonstrate that the foreign material control device used is at its maximum sensitivity for the product being produced.

In other words they want logic behind the metal detector test pieces we were using ,not so much the standard or laws surrounding the CCP.


We acheived this by ordering a range of test piece; types and sizes. We simply tested each piece type and size on each unique product and made record of it as a study. As a company we then decided what would be the "best" test spheres based upon the makeup of the product being tested and our findings.

Hopefully this is helpful.

I actually dont think is what they referring to.
e1.15.a Validation studies are conduced and documented to demonstrate that the foreign material control device used is at its maximum sensitivity for the product being produced.

In other words they want logic behind the metal detector test pieces we were using ,not so much the standard or laws surrounding the CCP.

I can understand how you have set up your metal detectors, however, BRC requires that critical limits are established taking into account industry codes of practice and legislation. With regards to setting up and sensitivity of metal detectors "best practice" should be applied.

Regards,

Tony

Dear All,

"Best practice" is such a lovely phrase.

Additionally, as discussed in other threads on this forum, another quite reasonable viewpoint IMO is to determine what is the “critical” size of any metallic etc contaminant. Existing US regulatory documents offer a value much larger than the detector sensitivity numbers being discussed here. Although this latter method perhaps rather diminishes the pride in one’s (expensive) metal detector, referential validation for any of these approaches can be found.

Normally, IMEX, auditors are very non-aggressive over such minutiae, I think the group involved were just looking for trouble, not enough nonconformances (or aperitifs ) perhaps.

Rgds / Charles.C

As part of our validation process we run tests with smaller than specified test pieces just tp porve we can't consistently detect at less than our current criteria which are dictated by our customers with the exception of KFC who accepted a LoV as we porved we couldn't achieve the levels the KFC standard required

BRC requires that critical limits are established taking into account industry codes of practice and legislation.

Im sorry, I guess I'm a little confused then. Would that just be FDA rules and regulations in regards to this then for the United States? Im looking for what I could print to include in my BRC book under "2.8" for a metal detector as our CCP. Any help would be of great value.

Amber

Im sorry, I guess I'm a little confused then. Would that just be FDA rules and regulations in regards to this then for the United States? Im looking for what I could print to include in my BRC book under "2.8" for a metal detector as our CCP. Any help would be of great value.

Amber

Yes, the FDA rules and regulations (SECTION 555.425 -Foods - Adulteration Involving Hard or Sharp Foreign Objects) is a good starting point but they are often superceded by customer expectations (Not many people will accept a 7 mm piece as outlined by the FDA)

I would suggest referencing FDA and customer requirements as your tools for deciding where to set you metal detection. That seemed to satisfy everyone that has audited my company (BRC, ISO, AIB, customers).

TS

(Not many people will accept a 7 mm piece as outlined by the FDA)

Presumably the National Asssociation of Dentists are a fairly strong supporter

@QAMB, All the refs to the USFDA related directives are on this forum but probably not in the same place unfortunately. There are 3-4 standard links including, from memory, one which effectively says you can take yr pick with respect to the max 7mm thingy and the detection limit philosophy.

Rgds / Charles.C

Ok-thank you, I will take all of this into account.

AM

The size/diameter of test pieces (Fe, Non-Fe or SUS)is depending on your detector capability. However, you can refer to FDA regulaiton at minimum but for customer' point of view, i think it is not OK.

For my experience, i will set the critical limitd as follow:
- Metal detector is functional (both detecting and rejecting) when we pass the test pieces (Fe, NonFe or SUS) with specific sensitivity and phrase (depending on metal detector capacity)
- Metal detector can detect the foreign matters (Fe, NonFE or SUS) at defined size/diameter.

Most of factories define only the detecting mechanism and forget for rejection mechanism. Can you imagine if it detect but not reject and are the products safe?

NY

Most of factories define only the detecting mechanism and forget for rejection mechanism. Can you imagine if it detect but not reject and are the products safe?

NY

Hi Narongchai

Could you please elaborate on this comment as I have to disagree. Most factories will check that the metal detector rejects the test piece so I am not quite sure what you are saying?

Regards,

Tony

"Most of factories define only the detecting mechanism and forget for rejection mechanism. Can you imagine if it detect but not reject and are the products safe?"




IMEX.. at my company if our metal detector detect metal piece in our product, the machine line will stop and we can separate the product and after that we must find metal piece in products ussually using sieveter. And disposition for the products is rework .. so i think with that system we are sure that our product safe... and i think same system applied by other company.

I also have a problem keeping this clause happy:

We do not have metal detectors but we do use a fixed magnet underneath a sieve used during packing of powdered food additives. The sieve and the magnet are each CCPs.
We do not have any specification on the magnet - its strength, its performance capability, whether it loses power over time... (I have asked Engineering to chase this from the supplier). I think we can say we have verified its use - it picks up metal 'filings' that get in with the (milled) raw materials but what would be the best way to validate it is fit for purpose in terms of particle size?

On a related note, I suppose the same question applies to the sieve. Also, we use filters for fluids as CCPs. The mesh size for each is based on the particle size (for solids) and viscosity (for fluids) so they go through at a reasonable rate but any suggestions for substantiating they are suitable? (They do seem to do what is required in terms of taking out foreign matter).

"We do not have metal detectors but we do use a fixed magnet underneath a sieve used during packing of powdered food additives. The sieve and the magnet are each CCPs.
We do not have any specification on the magnet - its strength, its performance capability, whether it loses power over time... (I have asked Engineering to chase this from the supplier). I think we can say we have verified its use - it picks up metal 'filings' that get in with the (milled) raw materials but what would be the best way to validate it is fit for purpose in terms of particle size?"

Youch! I hate magnets because of the above and also because if overloaded they have a habit of building up then dumping the contaminants back into the mix. Personally I don't think a magnet can ever be a suitable CCP as a result.

Back on the FDA thing I posted (aaaages ago) I don't see why this can't be used as part of your validation. Validation basically means "can it work?" ie "in theory is it capable?" Now metal detection is there to prevent choking or damage to teeth or soft tissue. I think some research conducted by the FDA on what sizes of contaminants cause injury is a very good reference for validation, however, it is only a reference. I would also consider your system and what sizes of contaminant are likely and what they are composed of. I know the UK approach is very much to use literature, codes of practice and legislation to back up validation. I agree that 7mm seems big, however, it doesn't stop you going smaller than that. The next part of your validation would be to ensure your machine is capable of detecting the critical limit you have decided upon. You could do some trial work to prove that or get some equipment manufacturer guidance. This is all different to monitoring (which would be your half hourly or hourly challenge tests for example) and verification (which could include complaint monitoring etc.)

Dear D-D,

On a related note, I suppose the same question applies to the sieve. Also, we use filters for fluids as CCPs. The mesh size for each is based on the particle size (for solids) and viscosity (for fluids) so they go through at a reasonable rate but any suggestions for substantiating they are suitable? (They do seem to do what is required in terms of taking out foreign matter).

This is very much related to actual operational data. The logical comment is to try a smaller one following the larger one and see what happens. Zero catch is good news, especially if you now hv very little flow through.

and regarding 7mm. I agree not many customers suggest this. I recall the logic to support was that FDA hv (successfully?) attacked product on this basis in USA. However, I think one can see dozens of ("optional") product recalls for much, much smaller particle size contaminations in the published lists.
I hv noticed (1/16) in. for Ferrous CL mentioned several times recently. This is pushing metal detectors a bit maybe.

Rgds / Charles.C

Dear D-D,



This is very much related to actual operational data. The logical comment is to try a smaller one following the larger one and see what happens. Zero catch is good news, especially if you now hv very little flow through.

and regarding 7mm. I agree not many customers suggest this. I recall the logic to support was that FDA hv (successfully?) attacked product on this basis in USA. However, I think one can see dozens of ("optional") product recalls for much, much smaller particle size contaminations in the published lists.
I hv noticed (1/16) in. for Ferrous CL mentioned several times recently. This is pushing metal detectors a bit maybe.

Rgds / Charles.C

It all depends on what you are detecting on how far you are pushing. Smaller "Point of us" apperatures are capable of detecting very small pieces. We have in line packaging detectors that can go down to .8 F, .8 NF and 1 SS (mm for all).

We unfortunately can't set them for that or we would reject just about everything because our supplies only go down to 1.5 across the board and some not even that far.

The verbiage can be particularly hard to understand when it comes to the BRC auditing scheme. The critical limit for metal detection should state as followed "All product passes through an operating metal detector" or something similar to this. Your standards used should also be referenced, usually as your verification.

The verbiage can be particularly hard to understand when it comes to the BRC auditing scheme. The critical limit for metal detection should state as followed "All product passes through an operating metal detector" or something similar to this. Your standards used should also be referenced, usually as your verification.

 

Above is an antique thread although I agree the topic itself remains a classic example of confusion and misinterpretation.

 

Maybe try this more recent "generic" discussion (albeit still illustrating the recurring difficulties) -

 

https://www.ifsqn.co...on/#entry133860

 

Here is CFIA's similar but more implementable version of your Critical Limit -

 

MD Critical Limit.PNG   29.6KB   3 downloads

This may help...

 

QAD 609 Procedure for the USDA´s Quality Assessment Division Procedure available at: 
https://www.ams.usda...l Detection.pdf

This may help...

 

QAD 609 Procedure for the USDA´s Quality Assessment Division Procedure available at: 
https://www.ams.usda...l Detection.pdf

 

It's an interesting document. Thanks for linking.

 

2 comments -

 

(1) Useful demonstration of the operational HACCP limitations associated with defining the absolute size of a HACCP "hazardous metal" rather than focussing on the equipment. Nonetheless the (FDA?) concept of adulteration although unmentioned will presumably still supercede the (sometimes avoidable) increased lolerances implied in the text ?.

 

(2) The sampling/monitoring procedure is IMO logistically illogical.