18 replies to this topic

[Milos Vasic]

Hello all

Does anyone have some material on color sorting of grains and nuts as CCP?

How do you perform verification?

We have sieving and air cleaning (aspiration) before color sorting and I need to determine CCP and limits.

Products that we clean are sunflower seeds,pumpkin seeds, raw popcorn... which we buy from the farmers and then we remove impurities by size(sieving), weight(aspiration)  and appearance(color sorting).

[Charles.C]

Hello all

Does anyone have some material on color sorting of grains and nuts as CCP?

How do you perform verification?

We have sieving and air cleaning (aspiration) before color sorting and I need to determine CCP and limits.

Products that we clean are sunflower seeds,pumpkin seeds, raw popcorn... which we buy from the farmers and then we remove impurities by size(sieving), weight(aspiration)  and appearance(color sorting).

 

 

Dear Milos Vasic,

 

Do you have some particular reason for believing that the colour is related to food safety  for this product ?

 

Rgds / Charles.C

[Milos Vasic]

Yes, because for pumpkin and for popocorn stones and ground are differnt color from the product. Also there is special camera which uses near infrared wave lenghts which can ,to some extent, separate based on the surface characteristic stones and oher pieces of plant even if they have similar color to good product.

I say to some extent and this is the biggest problem for me with this machine, because it does not work like for example metal detector which stops machine or removes bad product from the belt. I am having trouble finding good way of verification that it is working in CCP limits. Machine is latest Sortex so there is nothing better on the market.

[Charles.C]

Dear Milos Vasic,

 

Thks for the reply.

 

As I understand, the (physical) hazard is "stones" in the raw material not being adequately removed.

 

IMO, this would rarely be risk assessed as a CCP unless you can validate the necessity, eg due to local regulations, specific consumer, requirement similar to metal fragments. Otherwise, seems easier to handle within Prerequisite functions.?

 

I guess the typical verification procedure would then be to set (hopefully validatable) target limits to measure the effectiveness of stone removal and compare to the output results.

 

Maybe other people will disagree and suggest the dimensions of a "significantly" hazardous stone ?

 

Rgds / Charles.C

[Milos Vasic]

Dear Charles C

Thank you for answering.

To tell the truth I also had difficulties establishing CCP for color sorting. First of all our legisation regrding this does not say about size of the impurities but percentage in final product. This legislation is not very strict but our customers are so they expect 100% clean product.

In our specification we use 99,5-99,9% purity statement and lab analysis confirm this, so I would like to find way to prove this fast while working.

Milos

[tanyac]

I have never heard of color as a CCP.  For the seeds we use the sieves and screens as the ccp for them when blending, is packaging our CP is sieves then ccp is the metal detector. 

[JustinD]

Hi Milos,

 

I also work in the grain industry and we use colour sorters in one of our plants. We don't have them as CCPs but instead use the previous equipment (sieves, aspiration, magnets) as the CCP. We take samples at various ponts in the process and perform basic monitoring as the product is being processed (moisture, appearance, foriegn materials, insects). If you have a point where you can collect a sample after the colour sorters you can test the grain in process with some basic equipment and then verify your in-process testing with the lab reports.

 

The critical limits will be the grain specification (or customer specification) for the finished product. If you are allowed 0.5% foreign material then that would be the critical limit and you could measure that with your testing.

 

All the best.  

[Voyager161]

Milos

 

Colour sorting (in our case optical sorting is not 100% failsafe and it would be difficult to have this as a CCP - this step should be set up as a part of your prerequisite program as a control point. We do verify the optical sorter settings by confirming rejection of a set of standard foreign body objects such as glove tip, plaster, plastic and piece of wood through our optical sorter at the start of each shift to verify the settings are correct. Agree with previous comments on the use of other equipment as CCPs.

 

regards

 

David 

[Charles.C]

Dear Justin,

 

Thks for the input.

 

I also work in the grain industry and we use colour sorters in one of our plants. We don't have them as CCPs but instead use the previous equipment (sieves, aspiration, magnets) as the CCP

So what is yr critical limit and how to validate ? Satisfactory condition of a sieve of dimension [?micron] ? Nil detection on the magnet ?
 

 

The critical limits will be the grain specification (or customer specification) for the finished product. If you are allowed 0.5% foreign material then that would be the critical limit and you could measure that with your testing

 

I deduce you are suggesting that Milos's separation step should preferably be a  QCP rather than a CCP , ie the percentage value is a "quality" CL?

 

Rgds / Charles.C

 

PS - apologies to Milos for slightly hijacking his thread

[JustinD]

Hi Charles,

 

The critical limit is determined by the grain specification. The specification will dictate what percentage of the grain can be made up of foreign material, foriegn seeds, rocks, dirt, mouldy seeds, disformed seeds etc. The purpose of all grain cleaning equipment is to reduce the contaminants or remove them all together. Each machine is built to remove a certain kind of contaminant and by sampling the grain after the machine in question and testing for the contaminant you can monitor whether the grain is meeting the specification.

 

The actual tests performed on the grain will depend on what the process is trying to achieve. % of foreign material in a sample, bulk density, protien, moisture, % of disformed grain are all things that we commonly measure. For magnets we use a series of validated magnets and monitor the size and amount of particles on the magnets.

 

As far as validating the critical limit goes, here in Australia the grain specifications are regulated by national bodies and I have never had an auditor ask for more than that. Certain customers have requirements that exceed the regulated specifications but whatever the specification states, that is the critical limit.

 

Whether the step is a quality or a food safety issue is really a question of how dangerous the particles being removed by the colour sorters are. For example using the colour sorters to remove discoloured grain is a quality issue but removing mouldy grain with potential mycotoxins is definitely a safety risk. In my situation the colour sorters are primarily a quality device but there are a lot of processes before the colour sorters to get the final product as pure as possible and we are testing the grain before it gets to the colour sorters to make sure.

 

As was mentioned above colour sorters are not 100% effective and I would consider that if they are a CCP then I would consider whether further control may be required to achieve the desired result.

 

Cheers

[Charles.C]

Dear Justin

 

Interesting. Thank you.

 

Yes, there is some uncertainty (for me) whether quality / safety issues are being discussed in this thread. Or both.

 

I deduce from yr references to “specifications” that  most of the defects being discussed, eg stones, other odds-and-ends, metal fragments are not considered a (significant) incoming safety-related hazard in Australia. Perhaps mycotoxins are the only one ?

 

(Semantically I’m unsure if the term “critical limit” is reserved for CCPs or freely transferable to QCPs.)  

 

I think Milos is regarding stones as a significant physical hazard, although I'm not 100% sure. IMEX, these types of  environmental hazard are generally combined and controlled by PRP functions at reception, eg as  “extraneous matter” (added - although I guess from a haccp POV, it will depend on whether any subsequent control is present in the event of a likelihood of a validatably hazardous "material" occurring, eg metal / X-ray detectors).

 

I find it rather hard to believe that any procedures such as described here can achieve >=99.5% “purity”. Amazing.

 

Rgds / Charles.C

Edited by Charles.C, 01 November 2013 - 04:42 PM.
"added"

[Milos Vasic]

thanks all for sharing thoughts

 When I thought about post from Justine best way to look at at color sorting is that for some products it is CCP and for some quality check point. When you remove stones from product, or remove allergen (soybeans or wheat  from sunflower,popcorn  or pumpkin seeds in our case) or moldy product I see it as CCP. I agree that critical limits are local legislation or customer specification. 

 

We also have sieves and  aspiration before color sorter but in my opinion, while sieves remove bigger and smaller particles and aspiration removes lighter particles it up to color sorter to make sure that particles of same  size and weight as the product but different in color or surface get removed. So we take samples after color sorter and visually compare  it to approved sample (min.purity as specified)we also once per shift take impurities to be taken out we put  them through color sorter and it has to shot out 10 of 10. Sieves are checked visually and also final product is checked with lab sieves.  Machine doesn't have some info which could be of use to establish clearer critical limits.

[Charles.C]

Dear Milos Vasic,

 

Thks for reply. Yr process is now a little clearer to me.

 

I agree with yr comments there are analogies to other systems. In fact variations of yr question, notably for metal detectors,  occur in many other threads on this forum.

 

AFAIK, IMO, there are 3 initial, basic, questions associated with yr post, preferably interpreted within the typical chain of implementing a haccp system. The queries  / answers are inevitably debatable since many of the haccp decision procedures involve multiple factors (and opinions) together  with fundamental probability logic. Such is the nature of risk assessment.

 

(1) – Can the process step referred be justified to be pre-defined as a Prerequisite function ? If so, it will, by definition, not be a CCP.

 

(2) – If not pre-defined as a PRP, is there a significant hazard associated with the process step  for which the colour sorting activity is acting as a control measure ? The decision is based on a  (validatable) likelihood/severity risk assessment. If the answer is No then the step is not a CCP. Some experts may, now, consider it classifiable as a PRP.    If Yes, it may be a CCP (particularly in the Codex interpretation of  a CCP, eg as per the Codex tree). Some experts may, now, consider it classifiable as a PRP.

 

(3) –If a Yes answer as per (2), and you are implementing  ISO/FSSC-22000, it could also be an OPRP.

 

Personally, for “traditional” haccp, this generic analysis (Kraft) seems quite impressive IMO –

 

 haccp extraneous material.pdf   61.85KB   253 downloads

 

Rgds / Charles.C

 

PS - I would add one slight addition to the attachment, where "likelihood of occurrence" is mentioned, AFAIK,  it is typically referred to the situation at the point of consumption (assuming a finished product is discussed).

 

PPS, Qu2, Qu4 of the tree in attachment, are very slightly modified from the Codex tree. Qu2 in the latter might be (slightly) more definitive for yr current situation.

[Milos Vasic]

To conclude and I would like opinion from people who have experience with color sorters.

First to say again I work in factory which cleans grain and nut products popcorn, pumpkin seed, sunflower for roasting, sunflower for de-hulling....

My idea for our procedure is:

-sampling of raw material when it arrives to factory

-checking for all impurities 

-dividing impurities into Critical - for example stones, metal, moldy, cross-contaminant allergen and Aesthetic impurities for example husk on popcorn, shell on de-shelled sunflower...

-Checking that impurities are removed either by screens or Color sorting, if not than trying with new settings for color sorter until it is taken out if not possible than material cannot be cleaned and has to be returned to supplier 

-controlling that machine shoots and extracts  min 10 pieces of critical impurities on start of the shift

-samples taken 2 times per shift for verification of good cleaning if samples have critical impurities in percentage higher than in the specification that material made since last good check has to be marked as potentially unsafe and re cleaned after . And operator needs to reset the machine or to reduce capacity in order to achieve good cleaning. All the product made while setting the machine until 2 good consecutive result has to be marked as potentially unsafe and re cleaned.

Thanks everybody for helping  

[Charles.C]

Dear Milos Vasic,

 

I am unclear if you are still seeking an answer to yr OP regarding CCP or not ?

 

Yr word "critical" seems to include a mixture of "quality" defects, both safety / non-safety related ?

Therefore yr use of the words "verification", "unsafe" now seems more related to "quality" than safety ?

 

If you are still interested in people's opinions regarding safety / haccp status of yr sorting step, I suggest you post/attach  yr hazard analysis.

And yr critical limits if a CCP.

And  yr validation for the critical limits.

 

Rgds / Charles.C

[APPAJI]

Dear Milos,

I agree to hat you say. Yes The mouldy product or aflatoxin are great concern. these are chemical hazards which have to be removed. Now Sortex works on UV light and understanding the colour change under UV as these grains with aflatoxin will be shiny / fluoresce which are detected and eliminated by Sortex hence a typical CCP. Now limit for CCP is the operational limit and not the acceptable hazard limit. Hence the Critical limit could be the flow rate at which the equipment is able to detect and eliminate all the effected grains. higher the rate the chances of failure will be there, this needs to be validated based on flow rate.  

 

"NEW METHODS FOR RAPID DETECTION OF AFLATOXIN
C. W. HESSELTINE and 0. SHOTWELL
Northern Regional Research Laboratory, Peoria, Illinois, 61604, USA
ABSTRACT
The rapid detection of mycotoxin contamination in agricultural commodities is an area of research that requires more attention. The many companies buying hundreds of bushels of cereal grains daily, need to inspect their pur- chases immediately and to make a decision in minutes. Arising from this need to detect aflatoxin, we have developed two methods for rapid identification of aflatoxin in corn. The first is based upon a glowing greenish-gold fluorescence produced under ultraviolet light (365 nm) by corn kernels that contain afla- toxin. The fluorescent material is not aflatoxin hut a compound associated with it. Fluorescence depends upon the interaction between enzymes in corn and a compound produced by members of the Aspergillusfiavus series. Heat- sterilized corn inoculated with A. parasilicus does not fluoresce, although aflatoxin is formed. Fluorescence is seen in broken kernels but it is not visible in intact corn kernels until they are broken. Laboratory studies show that this method of detection is applicable for other cereals, including wheat, rice, oats and barley, but is apparently not effective for soybeans and peanuts. The second is a modification of the chromatographic mini-column method devised for corn."

 

Hope this is a help. Kindly give feed back

Regards

Appaji

[Charles.C]

Dear Appaji,

 

I'm a little confused.

 

The equipment referred by OP is stated as using infra-red, not ultra-violet.

And I don't think the OP had any particular focus on aflatoxin as his primary objective.

 

Or perhaps you are suggesting a new application/equipment from the same manufacturer. ?

 

Rgds / Charles.C

[monkeyman]

Hi

 

For Aflatoxin you have to follow a statistically representive sampling plan and then test a homoginous sample. Check local legislation and comply with this, in Europe we have 401/2006 which was amended in 2010.

 

On optical sorters my experience is that they are maybe 90 to 95% efficient which is not really good for a CCP but can be acceptable for a quality improvement method.

 

I worked in a peanut factory quite a while back and we used a shaking fluidised bed to remove stones from the product, we had test stones painted with a metallised material and if we lost them they could be picked up by metal detectors. However it depends on the density of the stones you are finding, pumice would be a problem!

[JessicaBeak]

Binsort have decades of years of development in sorting equipment industry, the market has come up with very intelligent sorting equipment with nice appearance, good capacity and handsome sorting accuracy.  https://www.binsort.cn/