31 replies to this topic

[ArnieTheTerminator]

Hello Folks,

 

 

New to the forum here.

 

I am operating a nutraceutical lab in the UK. Our products are solid tablets and capsules.

 

We are currently talking with Loma for a IQ3+E for a metal detection unit at end of production to scan finished product in packaging. 

 

We have the specs down to non ferr/fer >1.0mm. However, we do not know what kind of rejection control we need in order to be compliant with BRC. We can only afford one of these metal detectors so we would need to get it right first time.

 

At the moment we have a quotation for stop/start on detection, however I am concerned that we may need some kind of pusher arm with a locked box? 

 

NSF consultants have told us to check with Loma about this.

 

 

[FurryCrow]

Hi,

 

This comes under BRC 4.10.3.2

 

BRC 4.10.3.2

The metal detector or X-ray equipment shall incorporate one of the following:

• an automatic rejection device, for continuous in-line systems, which shall either divert contaminated

product out of the product flow or to a secure unit accessible only to authorised personnel

• a belt stop system with an alarm where the product cannot be automatically rejected (e.g. for very

large packs)

• in-line detectors which identify the location of the contaminant to allow effective segregation of the

affected product.

 

Talk to LOMA as your consultants have suggested, they're fantastic but a bit pricey.

[Eric G]

I found this video from Fortress Technology that shows the features on their BRC compliant conveyor. 

I hope this helps. 

Eric 

[MDaleDDF]

Our machines are all Fortress.  Great machines.

[ArnieTheTerminator]

I've emailed Fortress UK to see what they can do.

 

At the moment we are getting quotes of £11k+ Ex-vat. We also got a distributor quotation for the same, which means that there's quite a bit of margin added on and some room to go down :) 

[ArnieTheTerminator]

Something that would really help is a direct contact to Kramer in switzerland.

 

We also need a combi metal detector and deduster unit, however back in 2021 we got passed to a third party distributor who was not really professional.

 

We would introduce this at a later stage at the point of exit of the tablet press

 

Deduster KD6015 Combined Unit - https://www.kraemera...s_KD6015_EN.pdf

[Charles.C]

Hello Folks,

 

 

New to the forum here.

 

I am operating a nutraceutical lab in the UK. Our products are solid tablets and capsules.

 

We are currently talking with Loma for a IQ3+E for a metal detection unit at end of production to scan finished product in packaging. 

 

We have the specs down to non ferr/fer >1.0mm. However, we do not know what kind of rejection control we need in order to be compliant with BRC. We can only afford one of these metal detectors so we would need to get it right first time.

 

At the moment we have a quotation for stop/start on detection, however I am concerned that we may need some kind of pusher arm with a locked box? 

 

NSF consultants have told us to check with Loma about this.

Hi Arnie,

 

I am intrigued as to how you decided on a spec of > 1mm ?

 

Rule of thumb ?

 

The machine capability (eg sensitivity/LOD/LOQ) depends on numerous factors of course.

 

Also see clause 4.10.1.2. The BRC Interp.Guidelines are helpful also.

 

PS - Do you use any Stainless Steel equipment ? If so you should (maybe depending on RA) include a third specification (SS is typically more difficult to detect than Ferrous/non-Ferrous.

Edited by Charles.C, 11 April 2022 - 04:58 PM.
added

[Eric G]

Arnie

 

For BRC I think you would certainly need an automatic reject device and a lockable reject bin as this is not an situation where it would be unfeasible to do so.   Kicker should work fine, but the rate should be considered to ensure the kicker can fully extend and retract before the next pack comes along.  Assuming these are bottles of tablet which are fairly light, IMO an air-blast is less hassle as you can almost ignore the rate.  Also, less of a safety concern.   Don't forget all the reject sensors that make the system essentially failsafe, even if they seem redundant.  BRC seems to call for redundancy. 

 

I think the 1.0mm spec seems reasonable for packaged product.  Depending on the aperture, if relatively small then I think 0.8mm shouldn't be out of the question.   If it was a system running pills right from the press, then I would expect down to about 0.5mmFE.  

 

To address your price concerns, it doesn't seem unreasonable to me and usually you get what you pay for.  My experience is a rock solid system that will last forever if not completely neglected. Not to mention, easy to use, maintain and always supported.

 

Eric

[Charles.C]

Arnie

 

For BRC I think you would certainly need an automatic reject device and a lockable reject bin as this is not an situation where it would be unfeasible to do so.   Kicker should work fine, but the rate should be considered to ensure the kicker can fully extend and retract before the next pack comes along.  Assuming these are bottles of tablet which are fairly light, IMO an air-blast is less hassle as you can almost ignore the rate.  Also, less of a safety concern.   Don't forget all the reject sensors that make the system essentially failsafe, even if they seem redundant.  BRC seems to call for redundancy. 

 

I think the 1.0mm spec seems reasonable for packaged product.  Depending on the aperture, if relatively small then I think 0.8mm shouldn't be out of the question.   If it was a system running pills right from the press, then I would expect down to about 0.5mmFE.  

 

To address your price concerns, it doesn't seem unreasonable to me and usually you get what you pay for.  My experience is a rock solid system that will last forever if not completely neglected. Not to mention, easy to use, maintain and always supported.

 

Eric

Hi Eric,

 

Thks for comments.

Excellent Sales Pitch also !

[kingstudruler1]

There is also some parameters in (calibration) 6.4.1 - regarding "prevention of adjustment ......"   Most models have away lock / password protect  these settings .    Sensitivity, reject delay, etc

 

Ive seen facilities that have speed controls for a conveyor belt not locked.   This will cause issues with reject timing -, depending on reject device used.   

[ArnieTheTerminator]

Hi Arnie,

 

I am intrigued as to how you decided on a spec of > 1mm ?

 

Rule of thumb ?

 

The machine capability (eg sensitivity/LOD/LOQ) depends on numerous factors of course.

 

Also see clause 4.10.1.2. The BRC Interp.Guidelines are helpful also.

 

PS - Do you use any Stainless Steel equipment ? If so you should (maybe depending on RA) include a third specification (SS is typically more difficult to detect than Ferrous/non-Ferrous.

 

We use stainless steel definitely. I will mention this. 

Loma sales visited the premises and came up with that measurement. I guess it was based on the aperture size. If we want under <1.0mm we would probably do this with a pharma tablet detector straight out of the tablet press.  

 

Because we have different sized product packaging (stand up pouches and bottles) we have a larger aperture.

Edited by ArnieTheTerminator, 12 April 2022 - 09:19 AM.

[ArnieTheTerminator]

Arnie

 

For BRC I think you would certainly need an automatic reject device and a lockable reject bin as this is not an situation where it would be unfeasible to do so.   Kicker should work fine, but the rate should be considered to ensure the kicker can fully extend and retract before the next pack comes along.  Assuming these are bottles of tablet which are fairly light, IMO an air-blast is less hassle as you can almost ignore the rate.  Also, less of a safety concern.   Don't forget all the reject sensors that make the system essentially failsafe, even if they seem redundant.  BRC seems to call for redundancy. 

 

I think the 1.0mm spec seems reasonable for packaged product.  Depending on the aperture, if relatively small then I think 0.8mm shouldn't be out of the question.   If it was a system running pills right from the press, then I would expect down to about 0.5mmFE.  

 

To address your price concerns, it doesn't seem unreasonable to me and usually you get what you pay for.  My experience is a rock solid system that will last forever if not completely neglected. Not to mention, easy to use, maintain and always supported.

 

Eric

 

Hey Eric, 

 

Thanks for the information and knowledge. 

 

Loma has pushed back about a non alarm/belt stop system, as they said this is BRC compliant and because we are not running a continuous production line it wouldn't make sense. I forgot to mention that our production line is not continuous. Products will be manually placed onto the detector so it will always have an operator. 

[Eric G]

Arnie

 

After reviewing the newest standards again, I think you will likely be okay with the belt stop alarm option.  However, you may additionally have to supply reassurance your procedures will all but eliminate the risk of operatives putting the product back into good product flow as there is higher risk of this with a belt stop alarm.

 

IMO the standards seem to prefer an automatic rejection with a locking bin to segregate contaminated product. Though, it does indeed state that a belt stop alarm is also acceptable. 

 

Eric 

[ChristinaK]

Where I am we have a LOMA IQ3+ and an IQ4, each with different reject methods, in our production rooms. We're also BRC-certified and have not gotten any NC's on our setup. 

 

IQ3--Bulk Packaging Room: Our product is cut and then travels along a conveyor through the aperture before being packed in 2-to-55-gallon bulk containers and sealed. If Fe/NFe/SS are detected, the belt stops. The Operators are trained to search the section for the metal contaminant and then reject all the pieces into a discard bin. To continue, they must push a button to start the conveyor again. QA checks the function hourly with test wands placed on the belt. Fe is 3.5mm, NFe is 4.5mm, and SS is 5.0mm in size. We also have a blue bandage "test wand" (metal detectable bandage attached to business card and laminated).

 

IQ4--Retail Packaging Room: Our product is cut, then packaged into 8oz-1-gallon jars, then passes through the aperture before being filled with salad oil and sealed.  If Fe/NFe/SS are detected by the aperture, a reject arm pushes it into a marked "bin." We use glass containers in this room, so the "bin" is a stainless steel attachment that's level with the conveyor to prevent container breakages. Operators are trained to pass the item through the aperture again, or alert QA if there are more than 5 rejects during a production batch. QA checks the function hourly with test wands placed inside a product container. Fe is 3.5mm, NFe is 4.5mm, and SS is 5.5mm in size. We also have a blue bandage "test wand" (metal detectable bandage attached to business card and laminated). 

 

For both metal detectors, I perform an annual PVS test and record the data. Our past auditors seem satisfied with that, plus the QA logs and hazard analysis which details why the metal detectors are located at those production points for each production room and why they have those rejection mechanisms.

 

Hope this helps you in deciding what to choose.

[Charles.C]

Where I am we have a LOMA IQ3+ and an IQ4, each with different reject methods, in our production rooms. We're also BRC-certified and have not gotten any NC's on our setup. 

 

IQ3--Bulk Packaging Room: Our product is cut and then travels along a conveyor through the aperture before being packed in 2-to-55-gallon bulk containers and sealed. If Fe/NFe/SS are detected, the belt stops. The Operators are trained to search the section for the metal contaminant and then reject all the pieces into a discard bin. To continue, they must push a button to start the conveyor again. QA checks the function hourly with test wands placed on the belt. Fe is 3.5mm, NFe is 4.5mm, and SS is 5.0mm in size. We also have a blue bandage "test wand" (metal detectable bandage attached to business card and laminated).

 

IQ4--Retail Packaging Room: Our product is cut, then packaged into 8oz-1-gallon jars, then passes through the aperture before being filled with salad oil and sealed.  If Fe/NFe/SS are detected by the aperture, a reject arm pushes it into a marked "bin." We use glass containers in this room, so the "bin" is a stainless steel attachment that's level with the conveyor to prevent container breakages. Operators are trained to pass the item through the aperture again, or alert QA if there are more than 5 rejects during a production batch. QA checks the function hourly with test wands placed inside a product container. Fe is 3.5mm, NFe is 4.5mm, and SS is 5.5mm in size. We also have a blue bandage "test wand" (metal detectable bandage attached to business card and laminated). 

 

For both metal detectors, I perform an annual PVS test and record the data. Our past auditors seem satisfied with that, plus the QA logs and hazard analysis which details why the metal detectors are located at those production points for each production room and why they have those rejection mechanisms.

 

Hope this helps you in deciding what to choose.

Hi Christina,

 

Thanks for the detailed description.

 

What is the Product being cut ? I deduce the pieces for IQ3 are "large". (Hence maybe the slightly diminished sensitivity).

 

I can understand the atypical location of MD prior to packaging in IQ3 but I daresay this complicates the Validation (non-uniform sized target ?) step.

Again, I can understand the practical limitation but the monitoring/verification procedure for IQ3 (ie "placed on the belt") might be questioned with respect to the Standard.

 

Not sure what PVS test is (=Validation?)?

[ArnieTheTerminator]

This is the reply from Loma

 

Unless we test your products, we are unable to give a sensitivity guarantee.

 

The metalised film will have a significant effect and, in our experience, a 350mm wide x 200mm high head is likely to achieve a 3.5mm to 4.0mm stainless steel sensitivity.

 

We would be happy to test a few products if you would like us to verify this.

 

Regards,

 

I suppose, if we put our stand up pouches and bottles laying flat, we can reduce the aperture head height from 200mm to 150mm, but this limits our future proofing for products later on. 

Edited by ArnieTheTerminator, 14 April 2022 - 12:01 PM.

[ChristinaK]

Hi Christina,

 

Thanks for the detailed description.

 

What is the Product being cut ? I deduce the pieces for IQ3 are "large". (Hence maybe the slightly diminished sensitivity).

 

I can understand the atypical location of MD prior to packaging in IQ3 but I daresay this complicates the Validation (non-uniform sized target ?) step.

Again, I can understand the practical limitation but the monitoring/verification procedure for IQ3 (ie "placed on the belt") might be questioned with respect to the Standard.

 

Not sure what PVS test is (=Validation?)?

 

Hi Charles,

 

The production room with the IQ3 handles cucumber slices & spears, which exit the metal detector belt and are packed into bulk containers with brine to cure into refrigerated fresh-pack pickles. So for that product, it's not really possible to insert the test wands into the slices or spears, but the wands are placed with the product running on the belt to test them (simulating what kind of metal contamination we would likely have).  Historically, any metal contamination would be from the blades used for slicing. Any metal contaminants present in the received cucumbers would sink to the bottom of our wash tank and/or fall through the lift conveyor out of the wash tank. And if they did somehow pass beyond that, they'd either be seen by the inspection conveyor operator or cause a jam or other breakage in our slicing equipment. We have 3 operators along the line at the slicing, IQ3 stop area, and filling area who also visually inspect. The IQ3 we use has a short & wide aperture, about 6" high from the belt, and our sliced products reach maybe 2" in height max. 

 

We're not a very large or "advanced" production facility with very much automation, so the majority of our flow steps/processes have at least 1 dedicated operator for the shift.

 

Oh, yes, PVS for the IQ4 is a built-in validation program that can be set up to accommodate the size of your test pieces & the product's detection program (sensitivity, gain, limits, etc.). I believe it can also be programmed to prompt an operator to regularly test the system during production as well, although we don't use the prompt feature at this time. LOMA advertises it as a way to prove validation to auditors.

[ChristinaK]

This is the reply from Loma

 

 

I suppose, if we put our stand up pouches and bottles laying flat, we can reduce the aperture head height from 200mm to 150mm, but this limits our future proofing for products later on. 

 

Oh yes, the metalized film does make detection more difficult, especially for products that are closer to the max allowed for the aperture. 

 

If you are able to have LOMA test some of your products, I definitely suggest doing so.

Have you looked into using x-ray detection? I believe it's popular for products with uniform size and shape. Not possible for our products here, so I've never looked into it much. 

[Charles.C]

Hi Charles,

 

The production room with the IQ3 handles cucumber slices & spears, which exit the metal detector belt and are packed into bulk containers with brine to cure into refrigerated fresh-pack pickles. So for that product, it's not really possible to insert the test wands into the slices or spears, but the wands are placed with the product running on the belt to test them (simulating what kind of metal contamination we would likely have).  Historically, any metal contamination would be from the blades used for slicing. Any metal contaminants present in the received cucumbers would sink to the bottom of our wash tank and/or fall through the lift conveyor out of the wash tank. And if they did somehow pass beyond that, they'd either be seen by the inspection conveyor operator or cause a jam or other breakage in our slicing equipment. We have 3 operators along the line at the slicing, IQ3 stop area, and filling area who also visually inspect. The IQ3 we use has a short & wide aperture, about 6" high from the belt, and our sliced products reach maybe 2" in height max. 

 

We're not a very large or "advanced" production facility with very much automation, so the majority of our flow steps/processes have at least 1 dedicated operator for the shift.

 

Oh, yes, PVS for the IQ4 is a built-in validation program that can be set up to accommodate the size of your test pieces & the product's detection program (sensitivity, gain, limits, etc.). I believe it can also be programmed to prompt an operator to regularly test the system during production as well, although we don't use the prompt feature at this time. LOMA advertises it as a way to prove validation to auditors.

Hi Christina,

 

Thks for the enlargement. Interesting process. I guess the set-up qualifies the MD as an "inline" detector from BRC's POV which entitles you to a little more flexibility.

 

PS - I'm rather surprised that you don't achieve better sensitivity (ie lower mm) for Fe but maybe a result of effective encapsulation of the wand by product.

[ArnieTheTerminator]

So,

 

We had some test results back from LOMA, they tested two types of packaging that we have. One had an aluminium lid liner inside of the lid which they didn't bother testing, but this passed their 1.2mm test result for NON-FE/FE with the lid removed. (Plastic PET bottle)

 

However, and rather more disappointingly, our stand up pouches (PET/AL) that had a metalised film welded results which were not great. They came back as the following results

 

Quotation: IQ3+E metal detection machine - new 

1.2 - non-fe

1.2mm - fe

2-3mm - stainless steel

 

Results: 

"We have tested your pouches and the results were not great due to the large surface area of the pouches and the thickness of the met film.

 

We had to increase the aperture height from the 200mm quoted to 250mm to reduce the amount of signal from the pouches.

 

For the largest met film pouch we were only able to reliably detect 4.0mm Ferrous, 4.5mm non-Ferrous and 7.0mm stainless steel.

 

The only way to achieve lower sensitivities is to consider an x-ray inspection system which would likely achieve 1.2mm to 1.5mm for all metals.

 

Please let me know how you would like to proceed."

 

 

Does this mean that the metalised film is the issue, or just the large surface area of the doy packs. To be honest this is quite disappointing because of doypacks are not huge in surface area, they don't even go above 1kg in weight for powder material. 

Edited by ArnieTheTerminator, 20 May 2022 - 02:04 PM.

[ArnieTheTerminator]

Redacted version of the test report - with names removed attached

 

I noticed it was cheekily tested on an IQ4 which is much more expensive than our IQ3+E quote

 

 

Edited by ArnieTheTerminator, 20 May 2022 - 02:16 PM.

[ArnieTheTerminator]

Here's the report 

Attached Files

•  Redacted Version JS - 2022 Product Test report.pdf   3.35MB   34 downloads

[Charles.C]

So,

 

We had some test results back from LOMA, they tested two types of packaging that we have. One had an aluminium lid liner inside of the lid which they didn't bother testing, but this passed their 1.2mm test result for NON-FE/FE with the lid removed. (Plastic PET bottle)

 

However, and rather more disappointingly, our stand up pouches (PET/AL) that had a metalised film welded results which were not great. They came back as the following results

 

Quotation: IQ3+E metal detection machine - new 

1.2 - non-fe

1.2mm - fe

2-3mm - stainless steel

 

Results: 

"We have tested your pouches and the results were not great due to the large surface area of the pouches and the thickness of the met film.

 

We had to increase the aperture height from the 200mm quoted to 250mm to reduce the amount of signal from the pouches.

 

For the largest met film pouch we were only able to reliably detect 4.0mm Ferrous, 4.5mm non-Ferrous and 7.0mm stainless steel.

 

The only way to achieve lower sensitivities is to consider an x-ray inspection system which would likely achieve 1.2mm to 1.5mm for all metals.

 

Please let me know how you would like to proceed."

 

 

Does this mean that the metalised film is the issue, or just the large surface area of the doy packs. To be honest this is quite disappointing because of doypacks are not huge in surface area, they don't even go above 1kg in weight for powder material. 

Metalised film is logically going to be (negatively) different to plastic. Just look at the numbers.

 

It is unclear to me how Loma define "sensitivity" in respect to an acceptable observed signal (eg see the red "xs" in attachment).

 

It is unclear to me where the test wands were positioned with respect to the samples.

 

From Loma's website -
 

 

So, what is the answer to “Can metal be detected in metalized packaging?”. Generally speaking, metalized packaging is not always compatible with conventional metal detectors and so it is unlikely to meet retailer requirements. The complexity is that metalized packaging creates a large product signal which can force the metal detector to operate at inferior sensitivity levels. This reduces inspection performance and typically makes the metal detector an ineffective solution for detecting metal contaminants.
 

The good news is there is an alternative to using a metal detector for inspecting foods that are in packaging that is metalized. The solution is an X-ray system! Not only are X-ray systems superior at metal detection through metallized packaging, they also have the ability to detect other contaminants like bone, stone and glass, plus check the integrity of the product being inspected.

https://www.loma.com...lized-packaging

 

I recall  discussions on the capabilty of MDs for Al trays started several years ago on this Forum. A Japanese brand then claimed to achieve surprisingly good sensitivities (IIRC ca. 1mm/Fe) although comments from users of other brands were not very optimistic.

[ArnieTheTerminator]

Metalised film is logically going to be (negatively) different to plastic. Just look at the numbers.

 

It is unclear to me how Loma define "sensitivity" in respect to an acceptable observed signal (eg see the red "xs" in attachment).

 

It is unclear to me where the test wands were positioned with respect to the samples.

 

From Loma's website -
 

 

I recall  discussions on the capabilty of MDs for Al trays started several years ago on this Forum. A Japanese brand then claimed to achieve surprisingly good sensitivities (IIRC ca. 1mm/Fe) although comments from users of other brands were not very optimistic.

Since they've not indicated where the test wands were placed I am non the wiser. 

I guess since there is no specific criteria for MD in BRC or otherwise then really it comes down to justifying it to an auditor? 

We can move to non metalised pouches which isn't too much of an issue as all of our competitors use non-met stand up pouches anyway. 

In terms of affecting product, in not quite sure what affect it has, potentially freshness?

[Charles.C]

Since they've not indicated where the test wands were placed I am non the wiser. 

I guess since there is no specific criteria for MD in BRC or otherwise then really it comes down to justifying it to an auditor? 

We can move to non metalised pouches which isn't too much of an issue as all of our competitors use non-met stand up pouches anyway. 

In terms of affecting product, in not quite sure what affect it has, potentially freshness?

Hi Arnie,

 

the BRC clause says it all -

 

4.10.1.2 - The type, location and sensitivity of the detection and/or removal method shall be specified as part of the
site’s documented system. Industry best practice shall be applied with regard to the nature of the ingredient,
material, product and/or the packed product. The location of the equipment or any other factors influencing
the sensitivity of the equipment shall be validated and justified.

 

 

The BRC I.Guidelines somewhat amplify/interpret the above.

 

"Industry best practice" is IMEX flexible but often vaguely correlated to typical customer expectations although the latter often tend to be somewhat "ambitious".  Are you aware that Loma et al publish typical MD sensitivities for (afaik) non-metallized packaging / typical, practical, apertures and product types (eg wet/dry) ? The Tables are available in various threads on this Forum.

 

Actually I'm a little surprised that for yr specific case Loma didn't mention  X-Ray (eg Post 18). Probably more expensive though.