11 replies to this topic

[Pratap Tirmare]

how we confirm  the sizes of test piece (Ferrous , Non-Ferrous And Stainless Steel) in frozen food industry??

[Charles.C]

how we confirm  the sizes of test piece (Ferrous , Non-Ferrous And Stainless Steel) in frozen food industry??

Hi Pratap Tirmare,

 

Welcome to the Forum and Thanks for yr Query !

 

The functional test-element is  normally an encased, spherical ball.

 

The test-piece should come with a "Certificate"  defining various characteristics including  a size in mm which is the diameter of the ball.

 

For example -

 

http://www.detectame...-test-stick.htm

[Simon]

I wonder if Pratap is asking about the size of the test piece?

I believe there are standards?

[Charles.C]

I wonder if Pratap is asking about the size of the test piece?

I believe there are standards?

 

You may be right, eg from my link -

 

Metal Detector Test Sticks

Available in a wide range of sizes and colours, the Test Stick is the most commonly used and easily recognisable test piece format suitable for a wide range of applications.

• Standard dimensions - 100mm x 12mm x 10mm
  (Custom sizes available on request)
• Colours - Red, Yellow, Blue, Green, Orange, Black & Clear.

So the answer could possibly be "up to [local?] supplier".

 

I guess only PT knows.

[Simon]

No I don't mean the test sticks and pieces one can buy. 

I mean does FDA or other guidelines regulate what size metal contaminants should be able to be detected.

Maybe it does not matter...perhaps I have the wrong end of the test stick

[Charles.C]

hi Simon,

 

I suppose another answer could be - "with a (calibrated) metal detector"

 

or even "a ruler".

 

It's all in the details.

[mgourley]

I think I have seen this discussion topic before. Basically, "what is the right size of test wand should we use?"

 

I don't believe there are any hard and fast rules. It all comes down to the sensitivity of your metal detector and what risk you are willing to take if metal gets in your product.

Obviously there are other factors as well, such as density of product, packaging, moisture content, etc. 

 

Attached is the study I did at a previous employer to show that the sizes of the test pieces we were using were "correct".

 

Marshall

Attached Files

•  Test Piece Validation.docx   12.55KB   258 downloads

[Simon]

Great; very useful Marshall.

 

Thanks,
Simon

[Charles.C]

Hi mgourley/Marshall,

 

Thanks for the attachment. Very interesting.

 

I liked the overall concept of the procedure. I guess this is an extract from another main document since  no  configurational settings/procedures are mentioned.

 

I assume a  “sample” refers to  the product, eg bread, with a test piece (wand) inserted so that the “active area” subsequently passes thru the central axis of the detector (assuming a symmetrical, horizontal, conveyor setup).

 

The logic for selection of (lower) critical limit appears based on choosing a value generating zero false negatives.

The Grid data, assuming a critical limit of 3mm for Fe indicates that the metal detector is also giving a (worst case) potential for approx. 50-60 % false positives (Bread/Hearth).

Is this a typical “trade-off” in the baking industry or does it indicate a need for a better metal detector.? (I recall another baking thread here where the limitation seemed to be due as much to the large sample dimension under test as maybe the intrinsic MD capabilities).

 

Might also be worth noting that the minimum size of metal which actually constitutes a significant hazard is by-passed by this approach. I  wonder how the processor (or FDA) reconcile some limits with the FDA policy that the “passed” product could be, if unlabelled, a potential hazard for, say, the sensitive consumer population ? For example in Canada/(some)EC, AFAIK the minimum hazardous “size” is defined at 2mm (perhaps they have another work-around for processors ).

[xylough]

Hi Pratap Tirmare,

 

I concur with Marshall. I do not believe there is any legislative mandate governing the sizes of test pieces to use. The only guidance is that 6 mm or greater debris is considered by the Federal (USA) government to be a choking hazard. Meeting this standard for detection and subsequently for test piece size would be the absolute minimum acceptable. Of course, now metal detectors have capabilities to detect far smaller metal debris. Two two factors in play in my own experience are the capability of the metal detector and the mandates of the most stringent standard you work to meet. For me it was always the 3rd party GMP or direct audits mandated by my large customers like Tesco and Costco. Large customers like these not only drive the test piece sizes through audit criteria, but often mandate as well that the test pieces be certified by an accredited lab and further that in your performance validations of your metal detection equipment that you "challenge"  the capabilities with the worst possible scenario conditions.

[Charles.C]

Hi xylough,

 

The only guidance is that 6 mm or greater debris is considered by the Federal (USA) government to be a choking hazard.

 

I assume the above means USFDA.

 

With all due respect, the limit referred probably does not relate to choking, eg  -

 

http://www.ifsqn.com...mit/#entry86163

 

I do agree with yr comments about Customers. From memory, i cannot reall any Fast Food chains wanting above 2.5mm for Fe but I have also never challenged their opinion, just validated conformance (or better). But not anywhere near so analytically as Marshall's procedure.

 

IMEX of frozen goods, (total) false positives can be a real headache, not just metallic ones.

[mgourley]

Charles,

 

The study was basically to see if we could go lower on our detection Critical Limits. Those limits had been defined well before i started working there, and nobody actually knew "why" those limits were set.

 

You are correct, what was posted was just the actual result data, not anything ancillary to that.

 

Marshall