15 replies to this topic

[3esa]

Why do metal detectors do poorly detecting stainless steel wire?  We've had this issue in the past, we even had to change some of our process to prevent the use of stainless steel wire.  And what can we do about this?

 

Also, I'm most likely sure this is a programming issue, but I've noticed that after long hours of operation our metal detector will pick up a lot of false positives.  This is usually caused by one strong detection of a real contaminant, followed by many false positives.  The metal detector is a belt conveyor type, the product is powdered oregano, powdered cumin, and powdered chile.  The easiest solution I've found is to stop the line, unplug the metal detector, reconnect and restart the metal detector.  Once we do that, it works perfectly for several hours until it goes crazy again.

[jtang]

Why does metal detectors not be able to detect titanium?

Titanium is used in cutting blades.

[Charles.C]

Why do metal detectors do poorly detecting stainless steel wire?  We've had this issue in the past, we even had to change some of our process to prevent the use of stainless steel wire.  And what can we do about this?

 

 

 

Hi 3esa,

 

It might be related to magnetism.

[saucy]

I realized the same issue here when magnet separator pulled stainless wire but metal detector did not catch it.

[Snookie]

Stainless is one of the hardest to detect.  There are many things which can effect why your detector is not staying stable.  You most likely need an expert to take a look at it.  I had a similar challenge in another plant.  The detector we were using was not really designed for our product and we had a lot of "noise" (not sound) in the area.  By making some changes in the machine and in the area we were eventually able to get it more stable. 

[3esa]

Hi 3esa,

 

It might be related to magnetism.

 

yeah but why so poor detection on wire?  The metal detector works perfectly with test cards (stainless steel sphere).  And more importantly, what can be done about it?

[Charles.C]

Hi 3esa - 

 

I'm unable to detail the Physics involved here but the problem is due to the "orientation effect" as a result of the shape of contaminant.

 

These 4 files should clarify the situation and possible solutions. Preferably start at det1 and so on.This is particularly a problem in the garment industry also (needles). 

 

 det1 - mettler - MD_Improving Sensitivity.pdf   503.14KB   89 downloads

 det2 - metal detectors, X-ray - inspection systems.pdf   2.04MB   68 downloads

 det3 - Guide Metal Detection Functionality,Woolworths,2014.pdf   324.37KB   71 downloads

 det4 - metal detection - AND handbook.pdf   1.15MB   90 downloads

 

If you interested, here are approaches focused on needles which further elaborate the problem -

 

 mettler - Quality Control and Needle Managment.pdf   1.13MB   73 downloads

[DavidAR]

something i can help on for a change with some personal experience that may help you.

 

1) i would have the metal detector health checked by the manufacturer, i recently had Loma out and it cost next to nothing for a 1 hr service. which they generated a full report on whether or not the detector was working / worn out due to age ( it is a 8yr old machine )

 

2) they also could informed me that the installation settings logged for the machine were not right, which are not always the same in all environments, temperature and humidity for example and types of product to be scanned can be setup differently.

 

3) has it been moved about? has it had anything attached to it since installation? an engineer fitted the control panel too close to the aperture of the head causing false readings when it wobbled about or somebody walked passed the machine.

 

4) is it the right machine for the product you are detecting? smaller businesses and i am not saying yours is tend to not want to use multiple detectors for different product lines due to cost and will end up with a big metal detector being used for tiny products and vice versa, (this was happening where i am and rectified it, the detector worked fine but but again hard to setup and keep stable sometimes.)

 

5) presuming your packaging is not foil based? if it is you shouldn't be trying to detect stainless anyways. ( again i worked for a company that was trying to detect stainless and when they couldn't get there brand new detector to work MAGNETIZED the ball in the test piece to make it work, i don't need to say anything more as to why this was a bad thing)

 

6) is the belt clean? has it been removed and cleaned and subsequently the machine calibrated out to ignore signals from belt?

 

7) any other electrical equipment near by that could interfere with the detector? remember the detection is not just under the metal aperture its around it (magnetic fields)  i have known some detectors to be so sensitive they react to a mobile phone in pocket or even a watch.

 

8) my 2 personal favorites. a) paperwork / clipboards near the machine, padlocks on lock boxes that are magnetic in nature can affect the machine if close to the head and b) is the right program being used? and are those settings correct for the product? some company would have a hard copy program folder with all settings for a specific product so they can be checked when something changes like a raw material is not quiet the same as it was last time.

 

9) are the test pieces the right size for the product type?

 

10) cant think of anything else of the top of my head but don't like to end on 9..

[Charles.C]

Hi David,

 

Thanks for all the suggestions.

 

As i interpret the OP, it might have been usefully restated as, for the same  "standard" MD working to the best of its capabilities, "why are stainless steel wires difficult to detect compared to  similar SS balls ? and similar/dissimilar shaped other materials.

 

As i understand, eg Pgs 2-3 of my attachment det1, the practical reason is due to Physics, notably Electromagnetics. Plus a certain amount of "Luck" from a success POV.

 

Nonetheless, the (unexplained) quantitative reason for this intriguing statement eluded me -

 

Figure 2 below shows that a piece of ferrous wire is in the most difficult orientation to detect when it is at 90° to the direction of flow, and the easiest when aligned along
the conveyor belt and the direction of travel. Non-ferrous and stainless steel wires are just the reverse. If this type of contamination is likely, the detector must be capable
of detecting it.

 

Perhaps you know ?

 

@3esa - TBH, my own response to the situation depicted in the 2nd para. of yr OP would be to immediately call in the local  representative of the machine supplier. My main experience is with frozen, glazed items and these can also suffer from false positive problems. The necessary set-up has to be a compromise IMEX but exactly where is not always a simple decision process.

[Snookie]

yeah but why so poor detection on wire?  The metal detector works perfectly with test cards (stainless steel sphere).  And more importantly, what can be done about it?

 

Most of the world thinks (and this includes more than few auditors who should know better) that if it is metal a metal detector, no matter how small, should be able to detect the metal.  The reality is far different.  When you are discussing metal detectors, you are talking about the effect of something (in this case metal) on a field and how the detector is set up, its environment etc all play a role in how well this is done.  As Charles said--physics. 

 

What can be done about it?......well that is the challenge we all fight every day.  PRP's, monitoring etc are some of the best things we can do until new technology comes along with its challenges.   

[Robert Rogers]

electrical conductivity and magnetic permeability of certain materials make them easier or more difficult to detect. Materials with both strong magnetic and conductive elements such as Ferrous are typically easier to detect. Materials with reduced magnetic and conductive elements such as SS are more difficult to detect. Spheres are utilized because they have no orientation effect meaning in any orientation (direction of travel) detection is the same. Wire contamination will exhibit an orientation effect where one orientation (direction of travel) will be easier to detect than the other. This is due to the rotation of the magnetic field (Flux) and how that field is affected in different materials.

 

Just like a screen will not capture material small enough to fit through the mesh the large the metal detector opening the less sensitive. Product conditions can also affect performance, dry non-conductive products can be run at higher sensitivities detecting smaller metal. Highly conductive products(high moisture, temperature, acidity etc.) a reduction in sensitivity and other parameters is necessary just to get the product through the system without a false rejection of non-contaminated product reducing the metal detection capability.

[DavidAR]

Hi David,

 

Thanks for all the suggestions.

 

As i interpret the OP, it might have been usefully restated as, for the same  "standard" MD working to the best of its capabilities, "why are stainless steel wires difficult to detect compared to  similar SS balls ? and similar/dissimilar shaped other materials.

 

As i understand, eg Pgs 2-3 of my attachment det1, the practical reason is due to Physics, notably Electromagnetics. Plus a certain amount of "Luck" from a success POV.

 

Nonetheless, the (unexplained) quantitative reason for this intriguing statement eluded me -

 

 

Perhaps you know?

 

Hi David,

 

Thanks for all the suggestions.

 

As i interpret the OP, it might have been usefully restated as, for the same  "standard" MD working to the best of its capabilities, "why are stainless steel wires difficult to detect compared to  similar SS balls ? and similar/dissimilar shaped other materials.

 

As i understand, eg Pgs 2-3 of my attachment det1, the practical reason is due to Physics, notably Electromagnetics. Plus a certain amount of "Luck" from a success POV.

 

Nonetheless, the (unexplained) quantitative reason for this intriguing statement eluded me -

 

 

Perhaps you know ?

 

 

I cant answer this scientifically however my understanding is that if you consider that the product is moving in one direction if it is lengh ways (taking longer to pass through the field) it will effect the field less than if it was horizontally through the field due to overall mass at point of detection. for example if you have a 2mm piece of metal 30 cms long its signal may b X if it goes through horizontally and spike the value quickly, if it goes thorugh lengh ways the signal peak wil be for a longer duration.

 

Something else i always think about and seems to be contradicted in alot of manufacturing companies i worked at is the weakest point is normaly defined as the centre of the aperature. alot of belts and products pass through  detector aperatures towards the bottom of deterctors.  

 

we currently pass 9kg boxes thorugh a detector which just fits and they were placing sticks under boxes on belt to save having to open cases. I soon stopped this!

 

another issue with stainless like is said in that guide, they come in many forms or grades. i have put 2.0mm stainless 304 thorugh a detector and it deteted but a 306 3.0mm did not under same conditions (personal testing no product running)

 

this is why when using foil trays or films it becomes almost impossible to detect and thus generally the HACCP will use higher ferrous sticks.

[DavidAR]

If in doubt though on any functionality of a metal detector its so cost effective to get them health checked for peace of mind its not worth faffing about with couple hundred quid or bucks and they can check all is ok etc.

[Jarb]

With the orientation effect of ferrous (magnetic) vs. non-ferrous (conductive), as I understand the detection of both materials is proportional to the "length" of the contaminant relative to the magnetic field (but of course ferrous detection is proportional to length parallel to field, and non-ferrous is proportional to the length perpendicular to the field).

 

Is the relationship to signal amplitude with the "length" of contaminant relative to the field, or is it with the "area" relative to the field?? This makes a big difference to me, because I'm trying to calculate the detectability of the angles in between 0 - 90 deg of a piece of wire. I know the best and worst cases, but want to see how our specification limit relates to different wire positions besides 0 and 90.

 

Hi 3esa - 

 

I'm unable to detail the Physics involved here but the problem is due to the "orientation effect" as a result of the shape of contaminant.

 

These 4 files should clarify the situation and possible solutions. Preferably start at det1 and so on.This is particularly a problem in the garment industry also (needles). 

 

det1 - mettler - MD_Improving Sensitivity.pdf

det2 - metal detectors, X-ray - inspection systems.pdf

det3 - Guide Metal Detection Functionality,Woolworths,2014.pdf

det4 - metal detection - AND handbook.pdf

 

If you interested, here are approaches focused on needles which further elaborate the problem -

 

mettler - Quality Control and Needle Managment.pdf

[JohnWheat]

Robert Rogers has provided (for me at least) the most accurate answer. As most of us use balanced coil detectors the main problems with stainless steel is that its nowhere as magnetic as ferrous and nowhere near as conductive as both ferrous or non-ferrous. That in basic terms is the problem with trying to detect stainless. Of course you can solve this by using a FB X-ray machine :)

[TripleM]

Good point, JohnWheat.

 

X-ray machine can give you better results on SS detection than the metal detector, but even an X-ray machine has limitation and if the thickness of the SS wire is less than 0.8mm and its length is pretty small, you will struggle to detect that on an X-ray machine..