Jump to content

  • Quick Navigation
Photo

Metal detector validation


  • You cannot start a new topic
  • Please log in to reply
18 replies to this topic

AJ1795

    Grade - MIFSQN

  • IFSQN Member
  • 58 posts
  • 0 thanks
0
Neutral

  • United States
    United States
  • Gender:Not Telling

Posted 15 April 2015 - 08:54 PM

We make an ice cream product that is run through a metal detector prior to packaging (we package in large containers and do not have a metal detector large enough).  My questions is how to go about validating the standards?  We obviously don't want production putting anything in the product, so right now they are using wands on the outside of the pipe segment that runs through the metal detector to set of the bypass system.  Is this enough?  Do we need to show that the detector will reject metal that is actually inside the pipe?  If so, how do we do this and how do we show that the wands simulate that situation without putting the wands inside the ice cream for every test?  Thanks in advance for your input! 



Julz

    Grade - AIFSQN

  • IFSQN Associate
  • 36 posts
  • 5 thanks
0
Neutral

  • Australia
    Australia

Posted 15 April 2015 - 11:42 PM

Hi AJ1795,

I also have a in-line pipe metal detector and use the wand test pieces as verification only.  To validate the process I have round (ball) test pieces which contain the same size metal as the wands and I add them via a ball valve (had it installed for this purpose) prior to the metal detetector.  We then place wire baskets or buckets (depending on your product being liquid or solid) to collect at both the MD reject valve and at the filler.  This way I make sure I collect the test pieces and they do not end up in finished product. This process is conducted annually.  Just a tip, check the valve type and the diameter of the valve that rejects the product ex the MD, the test pieces will need to be small enough to fit through and not get trapped in the valve.



Thanked by 3 Members:

Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 16 April 2015 - 03:45 AM

We make an ice cream product that is run through a metal detector prior to packaging (we package in large containers and do not have a metal detector large enough).  My questions is how to go about validating the standards?  We obviously don't want production putting anything in the product, so right now they are using wands on the outside of the pipe segment that runs through the metal detector to set of the bypass system.  Is this enough?  Do we need to show that the detector will reject metal that is actually inside the pipe?  If so, how do we do this and how do we show that the wands simulate that situation without putting the wands inside the ice cream for every test?  Thanks in advance for your input! 

 

Hi AJ1795,

 

I think Julz has answered this question for you. Your metal detection positioning is not ideal as I'm sure you are aware so you will also need to show that you assessed the risks of metal contamination/foreign body post metal detector and put control measures in place. Also do you filter the ice cream base prior to the freezer as this is obviously more effective in removing foreign bodies?

 

Regards,

 

Tony

 

Hi AIFSQN,

I also have a in-line pipe metal detector and use the wand test pieces as verification only.  To validate the process I have round (ball) test pieces which contain the same size metal as the wands and I add them via a ball valve (had it installed for this purpose) prior to the metal detetector.  We then place wire baskets or buckets (depending on your product being liquid or solid) to collect at both the MD reject valve and at the filler.  This way I make sure I collect the test pieces and they do not end up in finished product. This process is conducted annually.  Just a tip, check the valve type and the diameter of the valve that rejects the product ex the MD, the test pieces will need to be small enough to fit through and not get trapped in the valve.

 

 

Hi Julz,

 

I assume that you confirm you get the same result and level of detection for your wand test pieces and verification method at the same time as your validation. Also has anyone ever questioned your frequency of validation?

 

Regards,

 

Tony



Julz

    Grade - AIFSQN

  • IFSQN Associate
  • 36 posts
  • 5 thanks
0
Neutral

  • Australia
    Australia

Posted 16 April 2015 - 05:12 AM

 

Hi Julz,

 

I assume that you confirm you get the same result and level of detection for your wand test pieces and verification method at the same time as your validation. Also has anyone ever questioned your frequency of validation?

 

Regards,

 

Tony

Hi Tony,

Yes we confirm the same result and level of detection for the wand test pieces and the verification method. 

To date no one has questioned the frequency of the validation. The process is validated annually unless there is a change to the process or new product.  In the previous 12months I completed the validation twice. I am audited to SQF. Just wondering why you have asked the question about the frequency?



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 19,097 posts
  • 5313 thanks
1,281
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 16 April 2015 - 06:14 AM

Hi Tony,

Yes we confirm the same result and level of detection for the wand test pieces and the verification method. 

To date no one has questioned the frequency of the validation. The process is validated annually unless there is a change to the process or new product.  In the previous 12months I completed the validation twice. I am audited to SQF. Just wondering why you have asked the question about the frequency?

 

 

if an annual validation fails , it would appear you are faced with 1 years worth of product recall.? Sort of annual Russian Roulette.


Kind Regards,

 

Charles.C


Thanked by 1 Member:

Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 16 April 2015 - 12:44 PM

Hi Tony,

Yes we confirm the same result and level of detection for the wand test pieces and the verification method. 

To date no one has questioned the frequency of the validation. The process is validated annually unless there is a change to the process or new product.  In the previous 12months I completed the validation twice. I am audited to SQF. Just wondering why you have asked the question about the frequency?

 

Hi Julz,

 

The SQF Code requires validation, it doesn't specify a frequency so in theory you are compliant.

 

SQF Code Edition 7.2 11.7.6 Detection of Foreign Objects:
11.7.6.2 Metal detectors or other physical contaminant detection technologies shall be routinely monitored, validated and verified for operational effectiveness.

SQF Code Edition 7.2 Guidance 11.7.6 Detection of Foreign Objects:
What do I have to do?
Metal detectors, x-ray, color sorters (if used for defects or foreign material) and all other detection devices must be validated to ensure that they can effectively detect a foreign object within the packaged product that is passed through the device. The passing of wands through the device to ensure that it is working is verification. An example of a means for validation of a metal detector could be the placing of a piece of metal within the package of product (product would be marked to ensure it does not enter market). All types of packaging and sizes of product that are passed through the device must be validated as well as all new packaging or size of product.
Auditing Guidance
Evidence may include:
• Physical contaminant detection technology is validated;

 

More demanding customers would expect validation more frequently and I would for my own peace of mind.

 

I thought you might be interested in this extract from a retailer COP regarding sensitivity of test pieces:

 

Attached File  COP for Metal Detection Image.jpg   325.76KB   31 downloads

 

Regards,

 

Tony



Thanked by 2 Members:

AJ1795

    Grade - MIFSQN

  • IFSQN Member
  • 58 posts
  • 0 thanks
0
Neutral

  • United States
    United States
  • Gender:Not Telling

Posted 16 April 2015 - 02:17 PM

Hi AJ1795,

 

I think Julz has answered this question for you. Your metal detection positioning is not ideal as I'm sure you are aware so you will also need to show that you assessed the risks of metal contamination/foreign body post metal detector and put control measures in place. Also do you filter the ice cream base prior to the freezer as this is obviously more effective in removing foreign bodies?

 

Regards,

 

Tony

 

We do not currently filter, but that is a great idea.  I will look into this immediately.  Thanks!



Robert Rogers

    Grade - AIFSQN

  • Corporate Sponsor
  • 32 posts
  • 24 thanks
4
Neutral

  • United States
    United States
  • Gender:Male

Posted 24 April 2015 - 11:03 AM

The weakest point of sensitivity for a metal detector is at the absolute center of the opening. In pipe line applications it can be challenging to place the test samples in the weakest point. One way, as mentioned, is to actually insert the test sample into the product flow ensuring both the reject and accept sides are able to collect the test sample. With this method there can be concerns with sample retrieval, as well as, cross contamination and sanitary concerns. 

 

Placing the test samples outside of the pipe places the sample in a much more sensitive area, therefore if this method is used it must be validated in comparison to the center detector (weakest sensitivity). Typically a smaller sample would be utilized for testing in the stronger detection point outside of the pipe. 

 

Validation would be a series of tests (you determine the number of passes or samples) documenting sensitivity capability, detection levels and reject operation when tested in the center area. Then identifying a smaller size test sample that registers and equivalent detection level when passed out side of the pipe.

 

Validation should occur annually at a minimum and after any significant change.

 

You would then continue to verify proper operation by ensuring the detection levels are continuously met by passing the samples out side of the pipe. The frequency determined by your ability to control or rework potentially non-conforming product in the event of a verification failure. 

 

Both validation and verification should be conducted with product and conditions as they are during normal operations. 



Thanked by 1 Member:

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 19,097 posts
  • 5313 thanks
1,281
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 24 April 2015 - 11:11 AM

Hi Robert,

 

Many thanks for the above.

 

It seems to me that this particular situation has somewhat "challenged" the customary interpretations of Validation and Verification.


Kind Regards,

 

Charles.C


Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 24 April 2015 - 04:19 PM

Hi Julz,

 

The SQF Code requires validation, it doesn't specify a frequency so in theory you are compliant.

 

SQF Code Edition 7.2 11.7.6 Detection of Foreign Objects:
11.7.6.2 Metal detectors or other physical contaminant detection technologies shall be routinely monitored, validated and verified for operational effectiveness.

SQF Code Edition 7.2 Guidance 11.7.6 Detection of Foreign Objects:
What do I have to do?
Metal detectors, x-ray, color sorters (if used for defects or foreign material) and all other detection devices must be validated to ensure that they can effectively detect a foreign object within the packaged product that is passed through the device. The passing of wands through the device to ensure that it is working is verification. An example of a means for validation of a metal detector could be the placing of a piece of metal within the package of product (product would be marked to ensure it does not enter market). All types of packaging and sizes of product that are passed through the device must be validated as well as all new packaging or size of product.
Auditing Guidance
Evidence may include:
• Physical contaminant detection technology is validated;

 

More demanding customers would expect validation more frequently and I would for my own peace of mind.

 

I thought you might be interested in this extract from a retailer COP regarding sensitivity of test pieces:

 

attachicon.gifCOP for Metal Detection Image.jpg

 

Regards,

 

Tony

 

 

The weakest point of sensitivity for a metal detector is at the absolute center of the opening. In pipe line applications it can be challenging to place the test samples in the weakest point. One way, as mentioned, is to actually insert the test sample into the product flow ensuring both the reject and accept sides are able to collect the test sample. With this method there can be concerns with sample retrieval, as well as, cross contamination and sanitary concerns. 

 

Placing the test samples outside of the pipe places the sample in a much more sensitive area, therefore if this method is used it must be validated in comparison to the center detector (weakest sensitivity). Typically a smaller sample would be utilized for testing in the stronger detection point outside of the pipe. 

 

Validation would be a series of tests (you determine the number of passes or samples) documenting sensitivity capability, detection levels and reject operation when tested in the center area. Then identifying a smaller size test sample that registers and equivalent detection level when passed out side of the pipe.

 

Validation should occur annually at a minimum and after any significant change.

 

You would then continue to verify proper operation by ensuring the detection levels are continuously met by passing the samples out side of the pipe. The frequency determined by your ability to control or rework potentially non-conforming product in the event of a verification failure. 

 

Both validation and verification should be conducted with product and conditions as they are during normal operations. 

 

Appreciate you confirming what I'd already posted.

 

Regards,

 

Tony



bibi

    Grade - MIFSQN

  • IFSQN Member
  • 119 posts
  • 4 thanks
0
Neutral

Posted 01 June 2015 - 08:09 PM

Thanks all of you for your comments.

AJ1795 just wondering with your ice cream did you do the filtration? how?

How do you investigate the rejects from your pipe line.

 

Bibi



Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 02 June 2015 - 03:44 AM

Hi Bibi,

 

I suggested filtration to AJ1795, so perhaps I can help answer your question. Like metal detection it is best practice to place the filter as close to the end of the process/production line as possible. This would normally be somewhere between the buffer tanks and the freezer. Care should be taken to ensure the filter is of a design that can be cleaned adequately, normally via CIP.

 

You should examine rejects to find why the product was rejected and try to identify the source of any objects found. If you can identify the source then you can take action to prevent a recurrence and reduce the chances of a product with metal contamination entering the market.

 

Regards,

 

Tony



AJ1795

    Grade - MIFSQN

  • IFSQN Member
  • 58 posts
  • 0 thanks
0
Neutral

  • United States
    United States
  • Gender:Not Telling

Posted 02 June 2015 - 01:55 PM

Hi Bibi,

 

I had engineering install a filter at the point where the base mix enters the machine.  I would have liked it closer to the output end, but the way that our machine is set up, this was the only reasonable place to put it (we have an attachment that adds in the inclusions prior to extrusion, so obviously we could not filter after that point). 

 

For metal detection, we have the reject pipeline connected to a waste bucket and any rejected materials would be manually examined for the source of the reject.  As Tony says, this is then used to find the cause and hopefully then prevent it from recurring!  



Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 02 June 2015 - 05:59 PM

Hi Bibi,

 

I had engineering install a filter at the point where the base mix enters the machine.  I would have liked it closer to the output end, but the way that our machine is set up, this was the only reasonable place to put it (we have an attachment that adds in the inclusions prior to extrusion, so obviously we could not filter after that point). 

 

For metal detection, we have the reject pipeline connected to a waste bucket and any rejected materials would be manually examined for the source of the reject.  As Tony says, this is then used to find the cause and hopefully then prevent it from recurring!  

 

Great that you have installed the filter. How close is the metal detector and is it serving a purpose now or could you trade it in for an end of line metal detector? That is how I design production lines, filter before the freezer then metal detection at the end of the line.

 

Regards,

 

Tony



AJ1795

    Grade - MIFSQN

  • IFSQN Member
  • 58 posts
  • 0 thanks
0
Neutral

  • United States
    United States
  • Gender:Not Telling

Posted 02 June 2015 - 06:01 PM

Hi Tony,

 

The metal detector is at the end of the line =)



Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 02 June 2015 - 06:29 PM

Hi Tony,

 

The metal detector is at the end of the line =)

 

Perfect :thumbup:



bibi

    Grade - MIFSQN

  • IFSQN Member
  • 119 posts
  • 4 thanks
0
Neutral

Posted 02 June 2015 - 07:49 PM

Thank you for your reply :thumbup:

Interesting to know what other processing are doing.

May I ask what kind of filter you are using(specification?)

 

Bibi 



Tony-C

    Grade - FIFSQN

  • IFSQN Fellow
  • 3,397 posts
  • 1030 thanks
291
Excellent

  • United Kingdom
    United Kingdom
  • Gender:Male
  • Location:Koh Samui
  • Interests:My main interests are sports particularly football, pool, scuba diving, skiing and ten pin bowling.

Posted 03 June 2015 - 04:33 AM

One of these:

 

Attached File  Inline Dairy Filter.jpg   79.65KB   0 downloads

 

Stainless steel, can CIP. May be able to use 1mm without any detriment to the base if not 2mm should be fine.

 

Regards,

 

Tony



Thanked by 1 Member:

bibi

    Grade - MIFSQN

  • IFSQN Member
  • 119 posts
  • 4 thanks
0
Neutral

Posted 03 June 2015 - 07:55 PM

Many thanks

 

Bibi






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users