8 replies to this topic

[Whitney]

Hi Can anyone tell me when completing an internal audit on 6.4, what they expect to see for section 6.4.2?

 

I how do you show you comply with the section for metal detectors : all identified measuring devices, including new equipment, shall be checked and where necessary adjusted, to a defined method traceable to a recognised national or international standard if possible.

 

Firstly, which standard are they referring to? UKAS?

 

Would your calibration contractor have a UKAS cert which shows the methods? or would the test sticks have a calibrated cert traceable to UKAS?

 

also why do they state 'if possible'?

 

thanks

 

W

[Charles.C]

Hi Can anyone tell me when completing an internal audit on 6.4, what they expect to see for section 6.4.2?

 

I how do you show you comply with the section for metal detectors : all identified measuring devices, including new equipment, shall be checked and where necessary adjusted, to a defined method traceable to a recognised national or international standard if possible.

 

 

Firstly, which standard are they referring to? UKAS?

("recognized"  is a flexible term, see the text below)

 

Would your calibration contractor have a UKAS cert which shows the methods? or would the test sticks have a calibrated cert traceable to UKAS?

(traceable to a recognized national standard, Yes)

 

also why do they state 'if possible'?

 

 

thanks

 

W

 

Hi Whitney,

 

I suspect the answer(s)  is multi-faceted, as usual.

 

An, afaik, "official" quote is –

In  general,  calibration  of  equipment  needs  to  be  traceable  to  national  measurement standards.  When  using  external  calibration  services  a  valid  certificate  should  be obtained  from  a  calibration  laboratory  that  demonstrates  competence, measurement  capability  and  traceability.

A  calibration  certificate  bearing  the  UKAS accreditation [or identity of the national standards laboratory or mark of an accreditation  body  with  which  UKAS  has  a  recognition  agreement]  for  the  relevant calibration/verification  will  be  sufficient  evidence.

 Lab24ed1.pdf   55.88KB   144 downloads

 

The official  national standards labs in UK appear to be as per this link –

http://gsi.nist.gov/...1-3/L2-9/L3-103

 

Some interpretative confusion is also possible, eg

http://www.absolutec...r-ukas-cal.html

In this context I anticipate that BRC may likely not care as to whether "exactly" national or not, eg afaik(?)  NIST is the only national standards lab in USA but is likely equally acceptable to BRC via the bracketed comment  in above quotation.

 

I am not in USA but local Calibration certificates referenced to be “traceable to NIST”  seem to be fully BRC-acceptable. And similarly for metal detector test sticks, reference thermometers etc.

 

There is a quite detailed overview here -

https://en.wikipedia...iki/Calibration

Edited by Charles.C, 23 January 2016 - 06:19 AM.
edited/expanded

[Whitney]

Hi Charles

 

The company we used to calibrate our metal detectors issued us with calibration certs with the UKAS logo on. But when spoken to they advised they had to remove the logo from the certs as they shouldn't be using them. I am trying to get information from them but I am struggling.  So would you suggest that these aren't sufficient evidence, if the logo is removed?

 

thank you

[Robert Rogers]

According to the FSPCA (Food Safety Preventive Controls Alliance), the group challenged with developing training curriculum for FSMA compliance, they reference metal detection calibration. (http://www.iit.edu/i...s_control.shtml )

 

Can you really calibrate a metal detector?

The definition of “calibration” is evolving. For a Food Safety Plan if metal is identified as a significant hazard, you must determine the parameters that will enable detecting metal in your product and package, which will also depend on belt speed, aperture, etc. This is usually conducted by the metal detector supplier. Ongoing verification by the facility to ensure that the detector continues to operate as it was set up is required.

 

Typically calibration for metal detectors would consist of ensuring internal tuning of the detector is at factory acceptable levels. This is accomplished by connecting a calibrated measuring device (oscilloscope) to the detector and observing and documenting the readings. The calibration certificate for the oscilloscope should be up to date and provided during the visit.

 

Another reading that is commonly measured from a metal detector is the "signal balance" that is again measured using the certified calibrated oscilloscope.

 

​Other checks conducted during the calibration service should include confirmation of working auxiliary functions such as, reject operation, air fail monitoring systems, reject confirmation systems, beacons, horns, or other devices associated with the system as well as identifying detection capabilities utilizing certified test samples.

 

This is typically completed by the original manufacture as they have they best understanding as to what is defined as within factory acceptable limits.

[Charles.C]

Hi Whitney,

 

The meaning of  "calibration"  has been much argued over here. Just like “validation” which it “may” overlap. The conclusion has so far been, I think, "no consensus" , ie it’s up-to-you unless the standard is definitive which it usually isn’t (although the auditor may be !). 

 

BRC7 - Calibration - A set of operations that establish, under specified conditions, the relationship between values of quantities indicated by a measuring instrument or measuring system, or values represented by a material measure or reference material, and the corresponding values realised by standards.

 

So IMO, BRC "calibration" implies, crudely, activities carried out to enable/implement/evaluate  a  comparison .

 

You queried "if possible". i doubt that there is an Int.std method for tuning a MD ?

 

“Where necessary” IMO implies - as required for satisfactory monitoring of the (identified) CCP/legality. This is case-specific, eg “balances” may be checked every day/shift/month etc, the “hazard”/risk  is that the results are “not reliable/fail legality”.

 

IMO, from BRC's POV, the relevance of yr "Certificate of Calibration" will depend on the motivation / contents /results Plus the scope of  what it actually covers, eg - is the Certificate referred  intended to cover  6.4.3 as well ? If so there must be traceability/uncertainty comments included etc.( for sec. 4.10.3 also).

[Tony-C]

Hi Whitney,

 

6.4.2 Interpretation Guidelines
Calibration checks
The site needs to establish the method by which the precision and accuracy of equipment is verified.
This must include:
• a predefined check frequency, based on a risk assessment (e.g. historical reliability, nature of use, manufacturer’s recommendations)
• who is authorised to complete the checks (i.e. trained staff)
• the method to be used (which must, where possible, be traceable to a recognised standard, e.g. use of a master calibration thermometer that has a certified test certificate traceable to a national standard).

 

So your methods of checking should be based on industry code of practice/manufacturer's recommendations and your test pieces should have a certificate traceable to a national/international recognised standard. So in this case the laboratory issuing the certificate is accredited by UKAS for the test in question.

 

Kind regards,

 

Tony

 

[Charles.C]

Hi Whitney,

 

You might consider posting yr MD calibration certificate in this thread to enable comment.

I suspect the specific difficulties referred may relate to the second URL in post 2.

 

@Tony - thks for the guidelines. Some useful titbits, some intriguing ones. I wonder why BRC believe that manufacturer's recommendations are equivalent to a risk assessment.

[Tony-C]

@Tony - thks for the guidelines. Some useful titbits, some intriguing ones. I wonder why BRC believe that manufacturer's recommendations are equivalent to a risk assessment.

 

Maybe because the manufacturers of a metal detector know more about their capabilities than anyone else?

Regards,

 

Tony

[Charles.C]

Maybe because the manufacturers of a metal detector know more about their capabilities than anyone else?

Regards,

 

Tony

 

Hi Tony,

 

Unfortunately, IMEX, what manufacturer's know about their equipment, and what they "say" it can do, can be rather different  species.

Plus (sometimes) their knowledge of the product's intrinsic risk "status" and the surrounding process / environment is limited. Caveat emptor.

 

Maybe the Guideline is intended to imply a potential necessity to blend all 3 of the quoted elements. Acceptable responses  to the frequency RA request appear in practice to be relatively pragmatic/concise, eg based on results.  i would agree with that. "Nature of use" is a particularly convenient term  for RA purposes.

 

@Robert Rogers - As you say, the meaning of "calibration" is open to interpretation, and scope. BRC conveniently use the phrase "set of operations" which is sort of infinitely flexible. Their Code definition is textually more "process" oriented than your own  IMO. Just a question of emphasis perhaps.