Jump to content

  • Quick Navigation
Photo

Aerobic Plate Count in Finished Product

Share this

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

jwalter007

    Grade - Active

  • IFSQN Associate
  • 10 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 28 January 2015 - 04:22 PM

We have a fairly simple operation where we blend 4-6 powder raw materials to create a protein powder.

The only machine we use is a ribbon blender.

 

We combine the raw materials in a specific order and run the machine for preset times after each material is added. At completion, we acquire a representative sample & send off for lab analysis for aerobic plate count, ecoli, yeast & mold. All are consistently <10 with the exception of APC. Which vary greatly sometimes over 100.

 

My cleaning & sanitzation consists of scrubbing blender with detergent after use. Then sanitizing with steam then quat prior to use. We ATP swap test the blender before use & see very low results, often below ten or around ten RLU. All utensils & table surfaces  are sterilized with quat. floor is mopped prior. The outer layer of the bags of whey powder are peeled away before touching the blender. 

I have received large variations in test that were on batches that we run back to back. I would think these would have almost the same results. I realize that my raw material plate count will effect my final prouct plate count.

I can't think of anything that would be causing contamination. However I do not have a vast understanding of the aerobic plate count. 

 

Im asking for some direction in where/how to learn more about this so I might be able to adjust any methods that are not working.

 

I was thinking that my A/C system may be causing an issue.

thank you

 

 

 

 



Snookie

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,625 posts
  • 267 thanks
174
Excellent

  • United States
    United States
  • Gender:Female

Posted 28 January 2015 - 04:33 PM

Most food products are not sterile so am not surprised by a plate count.  Whether the plant count is acceptable, would be based on the products and the count. 


Posted Image
Live Long & Prosper

RG3

    Grade - PIFSQN

  • IFSQN Principal
  • 501 posts
  • 169 thanks
76
Excellent

  • United States
    United States
  • Gender:Male
  • Interests:"We cannot solve our problems with the same thinking we used when we created them" Albert Einstein

Posted 28 January 2015 - 05:16 PM

I agree with Snookie. APC is not something to be scared of especially if you're only around 100cfu/g with a protein and the rest of your testing is coming out clean. APC is mostly used as an indicator because it doesn't really pin point any microorganism (it grabs everything that isn't a gram positive due to the TTC dye). You're right in saying that your raw materials may have something to do with it. In fact it may solely have to do with it. If you want peace of mind you can look at all the micro specifications limits from your vendors for APC or you can even test out each ingredient for APC yourself. The heat created by friction or just the mixing of some ingredients in the ribbon blender would amplify something that is already happening on a small scale. 

 

However, if you still think it's your ribbon blender I would suggest you look at your pack seals on the axel on which the ribbon blender is actually on. I have found this is a difficult place to clean. Ask your maintenance what kind of seal you have there. Hopefully it's not made of a rope type of material (been there).



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

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

Posted 28 January 2015 - 06:41 PM

We have a fairly simple operation where we blend 4-6 powder raw materials to create a protein powder.

The only machine we use is a ribbon blender.

 

We combine the raw materials in a specific order and run the machine for preset times after each material is added. At completion, we acquire a representative sample & send off for lab analysis for aerobic plate count, ecoli, yeast & mold. All are consistently <10 with the exception of APC. Which vary greatly sometimes over 100.

 

My cleaning & sanitzation consists of scrubbing blender with detergent after use. Then sanitizing with steam then quat prior to use. We ATP swap test the blender before use & see very low results, often below ten or around ten RLU. All utensils & table surfaces  are sterilized with quat. floor is mopped prior. The outer layer of the bags of whey powder are peeled away before touching the blender. 

I have received large variations in test that were on batches that we run back to back. I would think these would have almost the same results. I realize that my raw material plate count will effect my final prouct plate count.

I can't think of anything that would be causing contamination. However I do not have a vast understanding of the aerobic plate count. 

 

Im asking for some direction in where/how to learn more about this so I might be able to adjust any methods that are not working.

 

I was thinking that my A/C system may be causing an issue.

thank you

 

Dear jwalter,

 

(1) The APC result is probably unrelated to the results for Y&M, E.coli (the APC result measures all microbial species on the product that will grow on the standard nutrient/conditions  used for the test).

(2) A low ATP result does not necessarily imply a low equipment surface APC count although yr sanitation treatment sounds quite thorough. If other data, eg inputs have low APC levels, eg <10cfu/gram, may be necessary to swab  the equipment to check its APC value.

(3) The measurement accuracy of APC is typically poor, eg one data point is not very meaningful.

(4) <10 cfu/gram for E.coli and Y&M probably means "undetected" which is OK unless yr specification max. limit is a number below 10 cfu/gram.

 

What is the APC specification (limit) for yr finished product ? [value should be  (i) expressed  in cfu/gram and (ii) probably measured around 35-37degC for 48 hrs ([ii] is often unstated)].

 

I realize that my raw material plate count will effect my final product plate count

.

Yes, precisely. So what is the upper limit  of APC values for each of the various inputs ? >10 ?,  >50?,  >100 cfu/gram?

 

Rgds / Charles.C


Kind Regards,

 

Charles.C


Thanked by 1 Member:

Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

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

Posted 29 January 2015 - 03:36 AM

Dear jwalter,

 

Protein "powders" not my area but from googling a few protein concentrate finished specs, the APC max limits appear to range from 10,000 - 30,000 cfu/gram.

 

If this value is similar to your own product, it would appear a result around 100 cfu/gram is not likely to be a problem.

 

But it may depend on your specific product of course.

 

Rgds / Charles.C


Kind Regards,

 

Charles.C


jwalter007

    Grade - Active

  • IFSQN Associate
  • 10 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 02 February 2015 - 01:50 PM

Thank you for all of the replies.

 

I'd still like to research further the causes of a high apc count.

I just received results of 2 batches which were mfg immediately, back to back. The results were the first batch 4300 and second 1400. I find it odd, as I would have thought the second would be the higher level.

These do fall within spec for us, but I want to start making our test results viewable to our buyers & would like to not have an inconsistent number. It may not be achievable due to inconsistency of the raws but a better understanding on my part would help.

 

RG3 mentioned friction inside the blender may cause an elevated reading. I found that interesting.

 

Anyone have a link or source where I would find additional training in this area.

thanks



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

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

Posted 02 February 2015 - 02:50 PM

Dear jwalter,

 

I appreciate your desire to assess the significance between the numbers but the fact is that the difference between 4300 and 1400 may well be totally insignificant. 

 

If you wish to know how much significant, you need reliable data so as not be comparing apples with oranges. And you may need some statistics. :smile:

 

Nonetheless , if you had a sequence of results like 3000, 4000, 1000, 15,000 it is reasonable to assume something happened, eg 1 or more poor quality inputs, wrong input amount, bad mixing,  bad sampling procedure, faulty lab analysis, etc

 

I assume each batch has the same composition, ie same amounts of input? If otherwise, you may be comparing apples and oranges.

 

To make any kind of judgements between numbers like you mention,  you need at least 2 independent samples to be analysed for each batch and maybe data for 5-6 batches (hopefully using same lots of raw materials).

 

Just as a crude example -  if you have 3 inputs with "average" APC levels of (i)100, (ii)500, (iii) 10,000 the result after mixing  will (obviously) be (a) highly dependent on the percentage of (iii) unless it is very small  and (b) also dependent on the internal variation of the individual APC data / accuracy of the sampling procedure (and the mixing of the inputs). And, sometimes, the reliability of the  lab. results.

 

Difficult to say much more without data.

 

Anyone have a link or source where I would find additional training in this area. thanks

 

Which area do you mean ? Microbiology ? Mixing equipment ? Statistics ? The problem / solution may involve all 3 unfortunately. :smile:

 

Rgds / Charles.C


Kind Regards,

 

Charles.C


Mr. Incognito

    "Mostly Harmless"

  • IFSQN Fellow
  • 1,571 posts
  • 272 thanks
131
Excellent

  • Earth
    Earth
  • Gender:Male

Posted 02 February 2015 - 03:45 PM

I agree with Charles.  The best thing you can do at this point is trend and chart your results.  Right now you are trying to find the answer to an unknown.  Currently you don't even know if you really have a problem.

 

Your product, I suspect, has a very low aw being a dry powder product.

 

If you were consistently around 100-300 and had a run that was 1500 I could see expecting that something changed on that run. 

 

Part of your problem could be the ingredients.  Seeing as you are sending the samples out to an outside lab I expect that you are not testing those.  One ingredient with a high plate count would raise the plate count of the resulting mix.  I don't trust supplier COA's further than I can throw their building.  Having absolutely no trust in anyone in the business world has let me keep my eyes open for potential pitfalls. "Our plate count is 1500?  Send it anyways they don't test it."


____________________________________________________________________________________________________

Mr. Incognito


:tardis:

Mr. Incognito is a cool frood who can travel the width and breadth of the galaxy and still know where his towel is.

Snookie

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,625 posts
  • 267 thanks
174
Excellent

  • United States
    United States
  • Gender:Female

Posted 02 February 2015 - 04:59 PM

Charles and Mr. I have given some very good advice.  Lab testing is usually not cheap but if done right can provide valuable information and you will able to determine root cause and if it is an issue or not. 


Posted Image
Live Long & Prosper

Mr. Incognito

    "Mostly Harmless"

  • IFSQN Fellow
  • 1,571 posts
  • 272 thanks
131
Excellent

  • Earth
    Earth
  • Gender:Male

Posted 02 February 2015 - 05:41 PM

One thing I learned in 6σ training was the idea of DMAIC

 

DMAIC stands for Define Measure Analyze Improve Control.  The tactic is used to help improve processes however when you learn how to apply it you see that before you can improve a process you have to be able to define and measure it so that you can analyze the issue to improve it.

 

You can look at a process that requires a lot of movement and say "We can do that better" but until you see it takes your operators 500 steps to move 3 feet in total can you say "Wow we are moving 497 steps more than we have to and that takes an extra 10 minutes out of an hour of production".

 

If you don't define, measure, and analyze any fix you try to make will really just be a shot in the dark.

 

To expound a bit:

 

APC:  

  Define:  What should our APC be?  If we can't define what it should be then we can't say we have a problem.

  Measure:  What is our APC in our current runs?  

  Analyze:  What has it been historically?   What has been our lowest?  What has been our highest?  What did we do during those runs that was different?

  Improve:  We found that has been a factor in our historical analysis so let's change this.

  Control:  Let's look back after 3 months and see if our improvement has worked.


Edited by Mr. Incognito, 02 February 2015 - 05:44 PM.

____________________________________________________________________________________________________

Mr. Incognito


:tardis:

Mr. Incognito is a cool frood who can travel the width and breadth of the galaxy and still know where his towel is.

cazyncymru

    Grade - FIFSQN

  • Banned
  • 1,604 posts
  • 341 thanks
130
Excellent

  • Earth
    Earth
  • Gender:Male

Posted 02 February 2015 - 08:13 PM

Having worked in a creamery that had 3 driers, our specs used to be 30,000 for TVC. 
You'll get possible contamination from things like conveyors, air handlers, bagging machines etc.
Caz x



cazyncymru

    Grade - FIFSQN

  • Banned
  • 1,604 posts
  • 341 thanks
130
Excellent

  • Earth
    Earth
  • Gender:Male

Posted 02 February 2015 - 08:13 PM

And don't forget, no 2 samples are the same!!


Sent from my iPhone using Tapatalk



jwalter007

    Grade - Active

  • IFSQN Associate
  • 10 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 02 February 2015 - 09:52 PM

Thanks for all the great info.

Im going to start charting my results.

 

 

Which area do you mean ? Microbiology ? Mixing equipment ? Statistics ? The problem / solution may involve all 3 unfortunately.  :smile:

 

Rgds / Charles.C

 

I would really like to learn more about microbiology. It kind of fascinates me plus advancing my knowledge in this area will help me improve my systems.

I'd like to learn how these guys move around and how they increase in numbers...sort of  a day in the life of a microorganism ...  type of info.

Hmm.... I'll google that title. Maybe there has been a documentary on that.



Snookie

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,625 posts
  • 267 thanks
174
Excellent

  • United States
    United States
  • Gender:Female

Posted 02 February 2015 - 10:18 PM

Am sorry we all missed it but better late than never..... :welcome:  to the forum. 

 

There are many interesting short courses in food microbiology.  A brief search should help you find something that meets your level of understanding.  Many are inexpensively priced. 


Posted Image
Live Long & Prosper

cazyncymru

    Grade - FIFSQN

  • Banned
  • 1,604 posts
  • 341 thanks
130
Excellent

  • Earth
    Earth
  • Gender:Male

Posted 03 February 2015 - 10:59 AM

Hi

 

I've just been looking at my Dairy Microbiology Handbook (bit of a bible really) with regards to micro on milk powders

 

A total colony count of 50,000 cfu/g is accepted as the bench mark for Extra Grade Dried Milk by the American Dairy Product Institute.

 

I should have asked, are you testing for thermodurics? Also you should be testing for Salmonella.

 

Caz x



John Moreton

    Grade - AIFSQN

  • IFSQN Associate
  • 37 posts
  • 7 thanks
3
Neutral

  • United Kingdom
    United Kingdom

Posted 11 May 2015 - 08:06 AM

this may be a bit late, however I use the British, Health Protection Agency Guidelines for assessing the Microbiological Safety of Ready-to-eat Foods Placed on the Market as my guide to Micro limits. It includes the tables for all relevant Micro limits and risks

 

usually the american and European limits are pretty much the same.



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

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

Posted 11 May 2015 - 09:15 AM

this may be a bit late, however I use the British, Health Protection Agency Guidelines for assessing the Microbiological Safety of Ready-to-eat Foods Placed on the Market as my guide to Micro limits. It includes the tables for all relevant Micro limits and risks

 

usually the american and European limits are pretty much the same.

 

Hi John Moreton,

 

Thks for input.

 

The link seems to not work but I'm guessing you refer to this popular item -

 

www.hpa.org.uk/webc/HPAwebFile/HPAweb_C/1259151921557

 

May i ask which food category in the list you use for yr "protein powders"  (I couldn't see an obvious analog) ?
 

 

usually the american and European limits are pretty much the same.

 

Not my product area but seems a bit dubious inasmuch as the methodologies are significantly diffferent, eg -

 

HPA (UK Std?) –  30degC/48hrs (actual conditions unstated (!) in 2009 ed. I assumed unchanged from prev. PHLS version)

USA (BAM) –        35degC/48hrs

EC (ISO 4833) –   30degC/72hrs

 

Globally microbiologists seem to delight in diversity, unfortunately. IMEX of various disputes, such variations can make real differences.

 

Of course it may be that, in present case, everyone is copying from each other ? :smile:


Kind Regards,

 

Charles.C


John Moreton

    Grade - AIFSQN

  • IFSQN Associate
  • 37 posts
  • 7 thanks
3
Neutral

  • United Kingdom
    United Kingdom

Posted 11 May 2015 - 09:35 AM

Hi Charles

 

I look at our product a little different form how "normal" food seen.

 

As we only deal with powders, they all have a Water Activity below 0.6 (They vary but all the ones i test are below 0.6) therefor there is no option for the growth of micro ( That's not to say there is none present...it just cant grow.) so as long as the micro tests we do are below the limits for all of the category's I accept that as good.

 

There is no specific category really for food supplements listed...we tend to be the forgotten industry so i have to assume the closest area. so number 4 "bakery and confectionery products with out dairy cream, powdered food"

 

We have no cooking times or added chemical to the production process, simply blend and press in to tablet or capsule or final blended powder.

 

the EU guidelines for testing methods are in CFU/25g so as long as my results come back <10 we are good.

 

The only Micro test we carry out are :

 

TVC

Entrobateriaceae

Ecoli

Bacilius Cereus

Clostridia

Yeast and Mold

Salmonella

 

Hope that makes sense =)



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

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

Posted 11 May 2015 - 09:40 AM

the EU guidelines for testing methods are in CFU/25g so as long as my results come back <10 we are good

 

 

For APC (= TVC) ?, this would be unique IMEX. Normally cfu/gram unless this is some rare document.

 

(perhaps 25g sample blended water, 0.1g on plate > no colonies, implies <10cfu/g) ?? Not impossible i guess.

 

On the other hand, bakery/HPA seems to be APC ("satisfactory") max 10,000cfu/g. Looks equally probable. :smile:


Kind Regards,

 

Charles.C


John Moreton

    Grade - AIFSQN

  • IFSQN Associate
  • 37 posts
  • 7 thanks
3
Neutral

  • United Kingdom
    United Kingdom

Posted 11 May 2015 - 09:57 AM

Sorry my mistake...per g for TVC, Entro and E-Coli  :oops2:



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 20,542 posts
  • 5662 thanks
1,544
Excellent

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

Posted 11 May 2015 - 10:10 AM

Sorry my mistake...per g for TVC, Entro and E-Coli  :oops2:

 

 

No problem. Micro. units are messy.

 

So  TVC spec. is presumably max 10cfu/gram. Similar to the APC data referred by OP but limit seems even tighter. The items i compared to prev. in thread were max. in the thousands !. Maybe a different product type. Or somebody's dreaming. :smile:

 

And Y&M same spec ?


Kind Regards,

 

Charles.C




Share this


Also tagged with one or more of these keywords: total plate count, aerobic plate count, atp

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users