Jump to content

  • Quick Navigation
Photo

High Microbial Counts in Cooling Water injected with Ozone

Share this

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

RMAV

    Grade - SIFSQN

  • IFSQN Senior
  • 407 posts
  • 122 thanks
44
Excellent

  • United States
    United States
  • Gender:Male
  • Location:USA - Midwest
  • Interests:QA, Micro, Sanitation;
    Meats, Juice, Condiments;
    SQF, Audit, and aviation

Posted 20 June 2013 - 12:09 AM

I have a problem:

 

Process-

Sealed, hot (180°F/82°C) bottles of various types of juice drinks enter a tunnel and sprayed with cool water through a series of pipes/manifolds and nozzles.  The water is recovered in reservoirs (separate but not sealed from each other) below and pumped back through the system to be sprayed through the nozzles again.  As the water is pumped through the system, it is filtered and injected with ozone, usually .05-.2 ppm.  Some spilled juice product is incidentally carried on the outside of some bottles and "washed" into the tunnel by the process. 

 

Microbial counts:

A 100ml vial is taken from each of the three reservoirs, of which 1ml is plated directly on APC film and coliform film respectively.  Typical counts at different times range from 300 cfu/ml - TNTC for APC, and 15-150 (sometimes TNTC!) cfu/ml for Coliform. 

 

The lab department has been hounding me about the TNTC APC's, but I am more concerned with the coliform counts.  Moreover, I have no idea what the limits should be in our application.  I also don't know how bad the counts can get...there is a limited number of data points.

 

Any experts out there:

-what do you think our limits should be for this application?

-what is the proper dosage of the ozone?

 

Thanks!



Charles.C

    Grade - FIFSQN

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

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

Posted 20 June 2013 - 03:26 AM

I have a problem:

 

Process-

Sealed, hot (180°F/82°C) bottles of various types of juice drinks enter a tunnel and sprayed with cool water through a series of pipes/manifolds and nozzles.  The water is recovered in reservoirs (separate but not sealed from each other) below and pumped back through the system to be sprayed through the nozzles again.  As the water is pumped through the system, it is filtered and injected with ozone, usually .05-.2 ppm.  Some spilled juice product is incidentally carried on the outside of some bottles and "washed" into the tunnel by the process. 

 

Microbial counts:

A 100ml vial is taken from each of the three reservoirs, of which 1ml is plated directly on APC film and coliform film respectively.  Typical counts at different times range from 300 cfu/ml - TNTC for APC, and 15-150 (sometimes TNTC!) cfu/ml for Coliform. 

 

The lab department has been hounding me about the TNTC APC's, but I am more concerned with the coliform counts.  Moreover, I have no idea what the limits should be in our application.  I also don't know how bad the counts can get...there is a limited number of data points.

 

Any experts out there:

-what do you think our limits should be for this application?

-what is the proper dosage of the ozone?

 

Thanks!

Dear RMAV,

 

Not my area but there is one obvious question -

 

Is this a new process with no previous data or you are referring to a sudden, apparentlly, loss of control ? If the latter, what were typical previous results ?

 

Rgds / Charles.C


Kind Regards,

 

Charles.C


Rodriguez-Gonzalez

    Grade - AIFSQN

  • IFSQN Associate
  • 45 posts
  • 13 thanks
7
Neutral

  • Canada
    Canada
  • Gender:Male
  • Location:Toronto
  • Interests:Food Safety Engineering, Continual Improvement, Sustainability

Posted 20 June 2013 - 04:39 AM

RMAV:

 

Based on my understanding of the ozone treatment re-circulation system I have a few observations:

 

1. The reservoirs of the re-circulation system can work as a "storage" for bacteria because the "ozoned" water gets diluted with spilled juice and then you have a media for bacterial growth. You could try replacing the water of the reservoirs more often or add some chlorine tablets to the reservoirs.

 

2. To measure the effectiveness of the ozone treatment you may want to try taking your measurements just before the cooling water gets into the reservoir. At this point you are also measuring bacteria that may be accumulating/growing in the corners of the reservoir (if is square).

 

3. Because is not in direct contact with the product you can increase the ozone concentration without worrying about food safety or flavors impacts. But then you  have to keep an eye in the costs.

 

I have to research if this idea has been used in ozone treatments, but just like in chlorine treatments the ozone concentration could be applied more precisely if it could be automatically (real time) correlated to the chemistry of the water, which I imagine that in your system is changing continuously when it adds more juice.

 

I hope this helps you,


Edited by Rodriguez-Gonzalez, 20 June 2013 - 04:41 AM.

  • HARPC likes this

Thanked by 3 Members:
RMAV , sci , Simon

SUSHIL

    Grade - MIFSQN

  • IFSQN Member
  • 162 posts
  • 225 thanks
19
Good

  • India
    India

Posted 20 June 2013 - 06:38 AM

Hello Mr RMAV,

You are talking about cooling tunnel,where hot filled bottles of beverages, tomatoketchup,and likes are passed for cooling before labelling and date coding. The cooling water very fast accumulates organic matter  spilled outside the bottles after filling ,these organic matter goes on accumulating ,forms bacterial slimes on bottles conveyors ,spraying nozzles inside cooling tunnel and cannot be removed by filteration etc.

Hence ozone and chlorine can work up to certain limit .Bacterial Slime formation on conveyor and spraying nozzles cannot be removed by these  anti-microbial agents.Cooling water if kept overnight sometimes has an off-odour.Hence it should be replaced with new fresh water after cleaning of reservoirs with foam sanitizer. Also monthly whole cooling tunnel should be dismantled and thoroughly cleaned of slime and other organic matter by brushing and foam sanitizers.



Thanked by 2 Members:
RMAV , sci

SPL

    Grade - MIFSQN

  • IFSQN Member
  • 85 posts
  • 21 thanks
4
Neutral

  • United States
    United States

Posted 20 June 2013 - 01:50 PM

From my experience in the bottle water industry, your ozone concentrations are too low. From what you stated .05 -.2 ppm are target ranges from finished product ozone levels. You may want to increase your ozone concentration to .7 to 1.2 ppm. Keep in mind water temperature will also have an effect on ozone levels, if your cooling water temp is high, ozone will easily dissipate. What size filter are you using?



Thanked by 2 Members:
RMAV , sci

RMAV

    Grade - SIFSQN

  • IFSQN Senior
  • 407 posts
  • 122 thanks
44
Excellent

  • United States
    United States
  • Gender:Male
  • Location:USA - Midwest
  • Interests:QA, Micro, Sanitation;
    Meats, Juice, Condiments;
    SQF, Audit, and aviation

Posted 20 June 2013 - 09:56 PM

Dear RMAV,

 

Not my area but there is one obvious question -

 

Is this a new process with no previous data or you are referring to a sudden, apparentlly, loss of control ? If the latter, what were typical previous results ?

 

Rgds / Charles.C

 

To answer some of the questions in the thread starting with Charles C:

 

Oddly enough, there is *some* previous data but nowhere near the data needed to draw any conclusions.  The previous data suggests this is an ongoing issue, moreover, the data is mostly collected near the beginning of the use of the tunnel after cleaning so the micro counts later in the production runs (3 or 4 days later) are likely well above what we're seeing.

 

Rodriguez-Gonzales, Sushil, and SPL,

 

You're dead on - my thought is that the reservoirs are good enviornments for growth.  As water runs through the pumps and nozzles, it moves quickly due to higher pressure, but then the water is slow-moving in the reservoir.  Unfortunately, there is no way to get to the reservoir to clean it effectively.  I can open draining ports on the side, stick a foam wand in there, I can run foaming cleaner through the system, run disinfectant through the system, but I cannot scrub the surfaces in that reservoir where I need to. 

 

So I am looking at managing the growth, not eliminating it.  My current thinking:

 

-ozone is not high enough as SPL pointed out, my supplier is working on that problem - if I recall from a few years ago, it needs a bigger system, I mean $$ystem!  :yikes:

 

-more frequent cleaning.  Current weekly procedure has us tearing it down as far as we can, rinsing as best we can, foaming, closing it up, running chlorinated alkaline cleaner through the system for 2 hrs, emptying, foaming with sanitizer, then leave it empty until production is ready to start.  Micro counts are hit and miss.  Perhaps an abreviated procedure each day with a more "thorough" cleaning once/week.

 

-Microbial limits: I need to know what I am shooting for, but our lab folks have no idea.  I have asked, but I fear they're going to "make something up" that is not appropriate for the application and it will be political fun challenging their result and where the limit came from - rather not go down that road if I don't have to.

 

Questions:

-Am I correct that I should worry more about the Coliform counts?

-Any idea as to what the limits for APC and Coliform should be?  (closed product, cooling tunnel is separated from exposed product by at least 30 meters, all products below 4 pH)



Charles.C

    Grade - FIFSQN

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

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

Posted 21 June 2013 - 03:52 AM

Dear RMAV,

 

I deduce the system has run for several years and that with respect to the cooling stage, no design criteria exist and similarly Process Validation. Sort of remarkable. I presume the situation has  perpetuated  since the final product has had no quality problems. Not so unusual.

 

I suppose that basically the requirements are deceptively simple, adequate cleaning and maintenance of an appropriate sanitizer level in the reservoirs.

 

It appears that one major problem is that you are unable to visibly assess yr reservoir cleaning efficiency. But  presumably it is possible to obtain within / post reservoir  data  prior to re-start so as to validate the overall cleaning, ie  under no load ? (a typical CIP-type measurement I imagine). The micro. levels  should surely be “low” and consistent, eg not comparable to yr loaded values. Obviously if this (background) data is major  unacceptable, you are probably wasting yr time in investigating actual run figures.

 

As per previous posts, if the system is simply grossly overloaded at all times, then highly variable data is probably inevitable at the current sanitizer usage levels.

 

It is possible that ozone is not the optimum choice (or yr cleaning procedure). I have no experience this chemical unfortunately. Even if no operational validation, presumably there is a design reason for this  choice (somewhere?).

 

Regarding yr quoted ozone data. I assume the ozone level you refer to is the input level. What was the level / range in the samples you withdrew for analysis ?  One would hope for some kind of correlation, ie a higher level in the very low microbial  samples, perhaps nil in others. Hopefully not “undetected” in all samples, ie system is simply out of control (this would easily explain the TNTCs :smile: ).

 

Rgds / Charles.C

 

PS - you asked about micro. limits. This needs meaningful input/output data on yr system so as to get a baseline. The ideal is presumably (hopefully) a no-load value :smile: . i experienced similar problems on quantitating BOD efficiencies of wastewater treatment systems for plant water. Significant variability is inevitable but you need a (legal in my case) maximum target (and at least the operation was mostly visible).

 

I'm only comparing specs for  hydrocoolers but IMO an equally critical figure  is the level of sanitiser in the reservoirs / output, at least for design purposes.

 

It should not be too impossible to do some static tests to see how quickly an applied ozone level to a representative reservoir sample disappears. This is analogous to breakthough type evaluation in  water treatment systems.

 

PPS - depending on their actual procedure, yr lab probably know what TNTC actually represents. 

eg > 1000 cfu/g, > 10,000 cfu/g ?


Kind Regards,

 

Charles.C


Thanked by 1 Member:
RMAV

SUSHIL

    Grade - MIFSQN

  • IFSQN Member
  • 162 posts
  • 225 thanks
19
Good

  • India
    India

Posted 21 June 2013 - 11:47 AM

Hello Mr RMAV,

Normally the cooling tunnels are designed with part of water reservoirs on the side of cooling tunnel containing mesh in slanting portion in between so that Dirt accumulation remains floating on one side of mesh and the other side contains  clear water for Input to transfer pump ,to transfer water to spray nozzles. Cooling tunnels can be opened from sides and top for thorough cleaning of conveyor and nozzles ,it also has opening on sides  near reservoirs for thorough cleaning below conveyors which contains other part of reservoir

 

Cooling tunnel is divided in three or more sections.

First section contains hot water through heat exchanger  in reservoir for cleaning of spilled product on outside of bottles and second reservoir contains normal water and third reservoir is supplied with cold water through heat exchanger for final cooling of bottles.

 

 You can contact your cooling tunnel machine designer for modification for better cleaning and sanitizing.

 

Regarding coliforms Please test you raw water before it enters in cooling tunnel and  the Source of raw water for coliforms and other pathogens.

Most coliform bacteria do not cause disease. However, strains of E. coli, particularly the strain 0157:H7, can cause serious illness. Cooling tunnel water should be free of E.coli and other pathogens

 

Since the PET /GLASS bottles are hot filled products and  with plastic caps /crown /roll on screw caps with high tamper evident and Seal Integrity, there is less chance of product contamination with cooling water. There fore  Ingress of cooling tunnel water will be due to loss of seal integrity of caps. This problem of seal integrity is  more in lug capped bottles rather than screw cap bottles.

Maintaining a clean and properly sanitized cooling tunnel is a key part of establishing the commercially stablility hot filled process.

Your cooling water cannot spill over to filling machine as filling machines are enclosed on all sides with plexi-glass.and cooling water tunnel is some distance away from filling machine..

On line quality checks are carried out for hot filling temperature of product, cap seal integrity,Torque test of caps etc.

 

Also check for vacuum in bottles of finished products, ph and commercial sterility, fermentation test and other physical,chemical and  microbiological test are carried out.

 

Also have your cooling tunnel water and raw water from source periodically  tested from accredited outside labs

Traditional methods for testing Coliforms

Traditionally the agar plate count method using VRBA (Violet Red Bile Agar) and the MPN methods are being used.  The plate count method takes 24 hours to perform

and can use 1.0 ml of 1:10 dilution and as a result has a sensitivity of <10 cfu/gram.

When higher sensitivity is required the MPN (Most Probable Number) method can be used.  Multiple tubes (typically 3) are inoculated for each dilution.  The method involves the inoculation of at least 3 decimal dilutions (a total of 9 tubes) of LST (Lauryl Tyrptose Broth).  Any LST tube that shows growth and gas production is transferred to BGLB (Brilliant green lactose bile), the MPN is calculated based upon the positive BGLB tubes.  It takes 3-5 days to complete.

These coliforms can be further tested for presence of E.coli  which should be absent.

 

Presently there is no data or specifications for cooling tunnel water  .since spoilage of cooling Water depends on spillage of  product outside bottles and  production activity (no of shifts run/no of bottles packed in filling machine per minute).cooling tunnel is good environment for microbes to grow fast hence the only solution is proper maintainence and cleaning and sanitizing  and to control product spillage in filling machine and bacteria in source of water and replacing of cooling tunnel water

as required and keeping the cooling tunnel dry after production and cleaning is over and during weekly off. RAW WATER TO COOLING TUNNEL TO BE ADDED AT THE START OF SHIFT.

Hence you have to establish your own data/specifications for cooling water and it should not contain any pathogens.

 

Raw Milk contains lots of bacteria per ml and some coliforms ,but it is pasteurised.

 

Attached Files



Thanked by 1 Member:
RMAV

SPL

    Grade - MIFSQN

  • IFSQN Member
  • 85 posts
  • 21 thanks
4
Neutral

  • United States
    United States

Posted 21 June 2013 - 12:35 PM

You could supplement your system with a GE osmonics (will be a different company by now) for $800 ~ $1000

 

Does your system use a hyperozonator?

 

If your process is not 24/7, I would look into cleaning per day it should reduce the load in your water.

 

At what rate iswater in the reservoirs replace, or is it a static fill?

 

Do you have an data on your incoming water supply and is it treated?



Thanked by 1 Member:
RMAV


Share this


Also tagged with one or more of these keywords: water, ozone, microbial, apc, coliform

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users