Thanks for the reply Charles.
I should clarify that we have not changed our standard pipeline metal detector. It has been fixed and continues to be used as normal. The testing we did of the finished product was using a temporary metal detector as a way to try and get some metal detection completed on the product.
Unfortunately re-running the affected stock through our pipeline MD is not possible. Our product is hot filled so this would require rip and tip of finished product, heating to >80oC, then passing through MD and into VFFS. This would breakdown all structure of the product rendering it useless to customers.
I agree that test piece size/MD critical limit is hard to define, and certainly here in NZ there is no legislation defined other than food must be safe for consumption. Our target market is not vulnerable populations.
I guess I am asking that if we as a business were prepared to accept a change in MD critical limit as a one off for this batch, how would we go about documenting that for this specific instance.
Thks. I see. The lack of re-run option is unfortunate. I am slightly curious as to yr current , documented, corrective action for segregated goods - write-off?
In most haccp plans there is somewhere, eg system verification, a requirement that a significant process change requires a revalidation of haccp plan. A change in MD critical limit (CL) would normally IMO fall into that.
I suggest to issue an appropriate, retrospective, revalidated, haccp plan (with the changed, validated critical limits) pertinent to cover the times of temporary change in MD.
The logical revalidation is to interpret the critical limit as the LOD of the detector as per yr (I assume) available test results.(I don't know what yr current method of haccp CL validation is)
I suppose an auditor may complain about (a) the relatively low sensitivity of backup MD and (b) the belated late issuance of relevant documents but perhaps you can balance those negatives against yr own estimated risk of (i) there actually being some contamination and (ii) the cost of a write-off.
As long as there is no regulatory MD/hazardous metal/best practice critical limit, the choice (within reasonable bounds, eg FDA) for (a) is IMO defendably based on the instrument only. There are, I think, published haccp opinions to support this logic. With the slight exception of the 5mm SS value, I think most auditors will be anyway satisfied with yr secondary CLs (my experience is only belt conveyor systems, i assume pipeline MD similar).
PS - customer specs which match yr upper MD limits wud also probably help but IMEX customers rarely quantitate such details except for the Big Boys..
PPS - a couple quite readable theads on MD / CL here (there are many more) -
(esp post 7)