DO sampling, another thread

[paintz55]

Just wanted to start a quick new discussion on this. We are interested in sampling our beer at various points for DO, mainly cans fresh off the filler. I am looking for the best tool to do this with, and we are interested in sampling cans pre-seamer (with the lids off). This way, there will be minimal down-time for our packaging crew.

Any recommendations would be much appreciated for equipment and technique. Are there hand-held optical units or are they exclusively in-line? How long must we wait to take a sample when it comes off the filter; I know CO2 microbubbles will settle after 15 minutes but I'm not sure how this will affect DO readings.

Ideally we would like to purchase a unit that we could also use for DO readings pre- and post- filtration, etc

Thanks!

[KWLSD]

There are a number of good meters on the market that are applicable to beer.

Optical units are the later addition to the mix and are available in both inline and hand-held formats. As they are more recent, they tend to be expensive as you would most likely have to buy new.

Alternatively electrochemical systems are still the most widely used across the industry. They are readily available - we have good refurbished units on stock at significant savings on new prices.

One of the benefits of EC technology is that the meter can be set up to read a wide range with good accuracy (typically optical systems will either do low-level or high-level but not both). This means an EC unit can measure from wort aeration right through to final package.

You will hear hear it said that optical systems need less maintenance than EC; this is not strictly true, just that EC sensors tend to be more sensitive to poor housekeeping and a lack of care. For example we have clients who's EC meters need servicing only every 6 months because they take care of them.

For package sampling, again there are a range of systems from 'all-singing' units that will measure in-pack DO2, CO2, TPO, pressure, temperature, etc, etc. to straightforward piercing units that are used with an ordinary portable meter. These can be used with a simple analysis protocol to give excellent data on dissolved oxygen, TPO and can also be used to identify any filler issues.

Again the benefit of simplicity = better prices!

Regarding sampling pre-seamer, the piercers described above are quick enough that you wouldn't gain any benefit. These use an O2-free gas (nitrogen, CO2) to force the beer out of the can/bottle through the DO2 meter - with an open package this wouldn't be possible for obvious reasons

CO2 microbubbles don't tend to affect oxygen readings significantly - most systems, optical or EC, have a back-pressure control (usually a needle valve on the sample chamber outlet) to ensure the beer flow through the meter is controlled and no gas breakout occurs.

We would be happy to supply you with a quote for a meter, if that would be of interest; Knight Warner Limited is based in the UK but we have a significant number of US/Canadian clients - as well as ones in Italy, New Zealand, etc. - and have extensive industry experience (I personally worked in QA for a number of brewers for nearly 25 years - much of which I spent armed with a DO2 meter!).

Just one final note - you will see loads of cheap meters, eg. on well-known online auction sites(...) that are aimed at water analysis. Don't be tempted to 'economise' with one of these - they don't have the accuracy or resolution for very low levels as they're designed principally for measuring high levels in watercourses, fish-tanks, etc.

[Larry Doyle]

This is for the poster, paintz 55: Is sampling cans pre seamer in addition to post seamer sampling or instead of post seamer sampling?

[paintz55]

Larry, I was asking about pre-seamer measurements with a handheld unit so we, possibly the filler operator, could take a quick read and make adjustments "on the fly." We check DO and CO2 with a piercing device, but typically this is done post-packaging in the lab. We are a small staff so running up to the lab during a packaging run is an inconvenience. Sounds like a necessary evil though.

KWLSD, that is very good info, thanks very much. I am the one we cares for and maintains our lab and field testing equipment, and believe me I am obsessive about it. Hate to see that stuff laying around dirty and not working properly. As for the CO2 microbubbles, you confirmed something that I have discussed with a few other people in the last day; I thought there would be interference from CO2 when measuring DO but apparently this is not the case.

We are also interested in measuring 2 other areas: aerated wort, and O2 content inside purged, pressurized brite tanks. Do you have any info on this equipment?

Slightly off topic, could any of you share your SOP for measuring TPO in cans? Standards for agitation, results, comparing results to DO, etc

Mahalo!

[KWLSD]

Depending on the type of meter and how it's been set up, it's very easy to measure wort DO2. All you need to ensure is that the level you're measuring isn't higher than the top limit of the meter's range.

eg. If you have an EC sensor, these can be fitted with different membranes that have different porosities to O2 molecules; roughly speaking, more porous = faster response but lower maximum range.

We used to shake all small-pack samples (bottles or cans) for 3 minutes on a flat shaker. You will find there are many different protocols - time, type of shaker, etc. - but the important part is to ensure you always use the same one for all your own samples. That way you have the best opportunity to compare like with like for your own facility.

If you take cans off the line immediately after filling/capping and analyse without shaking, you should get fairly good correlation with pre-filler, ie. bright tank.

Shaking will equilibrate the dissolved and headspace oxygen, so when used in conjunction with the unshaken result gives a good guide on filler performance.

Hope that helps!

[dereknobleluke]

I literally just had the pleasure of having a fella, Mr. Peter Gerard stopped by the brewery with his neat little device. His business (among others) is found at the following website: http://www.oxysense.com/

He can be reached pgerard53@cox.net.

My thought process is the following: his stuff is really expensive for a brewer that might use it one a month. but the data is invaluable. the way you collect data is non-invasive and you can get instantaneous headspace and liquid info seperately and log it over time.

BUT

What if the individual MBAA chapters were able to buy one of his machines as a group and then lend it out for a "fee" for say two weeks to a brewer interested. Its a great way to pool together resources.

From the data I just got this morning, I have some work to do and I already have a plan to to how to try and find it!

email derek@newportstorm.com if you have other questions!

[KWLSD]

The Oxysense system, and other non-invasive methods that use similar fluorescent dots can be useful for long-term trending of oxygen levels in package. We looked at them a long time ago, when I worked in QA for one of the major brewers, from which the following points came out:

- They are only useable in bottles so canned and draught beers can't be measured by the same methodology 'in pack'. To the best of my knowledge (and this may have been improved since we looked at the system) it only works with clear bottles, so ordinary brown or green production bottles can't be used. This limits the validity of the data as the packages are different;

- The sample bottles, as they contain the fluorescent dots, need to be retained so, despite being non-invasive they have no retail value;

- Using a small number of sample bottles does not necessarily give a representative view of a packaging run;

- Especially with high-speed filling lines, it is sometimes a problem to locate the sample.

Incidentally, both 'non-invasive' and 'invasive' measurement methods can provide separate headspace and dissolved data, eg. Orbisphere's 3625 TPO Package Analyzer and also give a wide range of other useful information, eg. headspace volume, fill volume, fill pressure, etc. which can be vital in either tracking the root of packaging issues or for improving the process.

Having said that, the systems are good for looking at oxygen ingress through crowns or for validating oxygen-scavenging seals, etc.

The methodology is already being applied widely in the wine industry to look at packaged oxygen levels and corks/screw caps, where it is especially useful as it provides analysis of non-carbonated products which can be difficult to measure by 'invasive' techniques...at the moment, any way

[dereknobleluke]

- They are only useable in bottles so canned and draught beers can't be measured by the same methodology 'in pack'.

NOT TRUE--you can literally drill a hole in a can lid and super glue the disk to the upper surface of a can lid. Same a with a bottle cap. we found zero O2 ingress after 1 week in a blank bottle that had been nitrogen purged.

To the best of my knowledge (and this may have been improved since we looked at the system) it only works with clear bottles, so ordinary brown or green production bottles can't be used.

I literally just got the exact same mold from O-I in clear glass. The standard 12 oz bottle, so yes you need clear glass, but it can be used again and again, using the same "dots".

- The sample bottles, as they contain the fluorescent dots, need to be retained so, despite being non-invasive they have no retail value;
[Good point. But 3 cases of the bottles is a small investment in the scope of the long term....[/U]]


- Using a small number of sample bottles does not necessarily give a representative view of a packaging run;
Yes that is a potential drawback. You must use and located the bottle and "randomness" cannot be created

- Especially with high-speed filling lines, it is sometimes a problem to locate the sample.
But GREAT for 25-100 BPM machines

Incidentally, both 'non-invasive' and 'invasive' measurement methods can provide separate headspace and dissolved data, eg. Orbisphere's 3625 TPO Package Analyzer and also give a wide range of other useful information, eg. headspace volume, fill volume, fill pressure, etc. which can be vital in either tracking the root of packaging issues or for improving the process.
This is true.

Having said that, the systems are good for looking at oxygen ingress through crowns or for validating oxygen-scavenging seals, etc.

The methodology is already being applied widely in the wine industry to look at packaged oxygen levels and corks/screw caps, where it is especially useful as it provides analysis of non-carbonated products which can be difficult to measure by 'invasive' techniques...at the moment, any way

[KWLSD]

NOT TRUE--you can literally drill a hole in a can lid and super glue the disk to the upper surface of a can lid. Same a with a bottle cap. we found zero O2 ingress after 1 week in a blank bottle that had been nitrogen purged.

Hmmm, I'm not sure how scientific this is...and how certain the data can be

I literally just got the exact same mold from O-I in clear glass. The standard 12 oz bottle, so yes you need clear glass, but it can be used again and again, using the same "dots".

But that surely defeats the object of having the same bottles as used in production, as the different glass types will allow light through in different ways and give different staling characteristics, some of which will be associated with oxygen levels, both headspace and dissolved?

I don't mean to sound negative about the system as it's been around for 10+ years now and has, I agree, some good applications for extended monitoring or where product characteristics don't lend themselves to other methods.

However - and I am sure this applies to most techniques in one way or another - the 'optical dot' has limitations that reduce its efficacy. Drilling holes in can ends and gluing in the dots is not really ideal for a simple, repeatable and robust method, as required for day-to-day QA analysis, which is what most people are looking for (certainly from our own experience feedback from our customers).