Pressure testing bottles

[jeff.k...@gmail.com]

Hi everyone.

I have been concerned about in-bottle pasteurizing because I have 5 volumes of CO2 in my cider.  At pasteurizing temperatures, that gives pretty high pressures (7 bars or so) and I am worried about bottle bombs.

There is frustratingly little information available on actual burst pressures for different kinds of bottles so I did some testing.  

I use 330 ml Tosca "birra" bottles that are about 330 g.  I set up a testing apparatus with a high pressure handpump for air guns that can reach 200 bars.  I fill the bottles with water for safety and put them in a filled, covered bucket of water to eliminate flying shards of glass.

Bursting pressure for these "beer" bottles (there is also a thicker champagne version) is between 40-70 bars.  A lot higher than I expected.  Extrapolating table 15.3 in The New Cider Maker's Handbook, you would need an SG drop of at least 0.06 to reach those pressures.  For regular cider (I make ice cider) that would mean bottling sugar-rich apple must immediately upon pressing and letting it ferment to dryness in the bottle.  Since I doubt many people do that, it seems like the biggest challenge with bottle bombs is with flawed bottles - either from fabrication or from reused bottles that might have a small chip.

I know american beer bottles are far lighter and are a whole other issue.  

Disclaimer:  results are shared for educational purposes only.  Your bottle bombs are your own responsibility!

/Jeff  

[Casa de Oro Cider]

This is super great! Thank you for the information. Definitely following for others info. I just started bottle conditioning and am running carbonation tests so this is fascinating info

Talk to you soon.
---

Anthony López
Sidrero y Pommelier

--
--
Visit our website: http://www.ciderworkshop.com
 
You received this message because you are subscribed to the "Cider Workshop" Google Group.
By joining the Cider Workshop, you agree to abide by our principles. Please see http://www.ciderworkshop.com/resources_principles.html
---
You received this message because you are subscribed to the Google Groups "Cider Workshop" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cider-worksho...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/cider-workshop/28c1c287-0d5d-4c56-b2ca-94a5456f7ed4n%40googlegroups.com.

[Terry Chalk]

Hi Jeff... interesting stuff which happens to line up with my experience with a test that wasn't as sophisticated as yours. 

Sorry this gets a bit long winded but it might be useful for others looking at the same issue. I had also been scratching around trying to get bottle ratings and in the end I did a "destruction test" using salvaged 330ml beer bottles.I fermented sugar and water in sealed bottles to 3 volumes then pasteurised at the highest temperature I could reach (I am at 1000M so only got a bit above 90C before the water boiled). According to Andrew Lea's Carbonation Table, pressure at that point should have been >15bar or over 200psi. No bottles burst but a crown cap popped off a Peroni bottle (quite robustly!).

Further research came up with China Misa Glass Co. quoting GB4544 (a Chinese Standard) where bottles are rated at 1.2MPa - 1.8MPa (174 - 261psi) depending on the market (they appear to have Heinekin bottles shown on their web site). I imagine that these pass-fail figures are based on batch testing so it is possible for sub standard bottles to get through. The only data that I could find on the spread of bottles in a batch was a PhD dissertation by Jo Morgan Teague at Ohio State University in the 1950s. Destruction testing of batches of bottles in that study generally resulted in the worst case individual bottle failure being about 50% of the mean for a batch. Time has moved on and bottle making technology may well be much better these days, nevertheless it seems to me that it could be assumed that limiting bottle pressure to half of the quoted standard should be a good safety target (say 150psi).

Based on this, I carbonate to 2.0 -2.5 volumes and limit pasteurising temperature to 65C without any bursting bottles. Once again, using Andrew's table, bottle pressure shouldn't get much higher than 100psi. In simple terms bringing the bottles up to 65C, holding them there for a few minutes then letting them cool down  results in around 50PUs of pasteurisation.

 .. 

To view this discussion on the web, visit https://groups.google.com/d/msgid/cider-workshop/CAF%3DUv_OmhDZEPk78kNcLWxfD2-1dXEGLbQ90Q8%2Bj7fLjjt8YKA%40mail.gmail.com.

[jeff.k...@gmail.com]

Terry,

Interesting information!  Another piece of anecdotal evidence is that I have had brand new bottles burst at a rate of about 1 in 1000 when filling with a counter pressure filler.  That is at a pressure of 3 bars and is a lot scarier than my controlled test because the bottles are filled with gas when they burst.  It only took one of those to understand why the plexiglass shields are in place...  Those bursting bottles could only withstand an order of magnitude less pressure than the "average" bottle.  I can only attribute this to some flaw in manufacturing.   

Since my initial report I have also tested cheapo commercial beer bottles (again, European not American) and they took over 20 bars before bursting.  Much, much higher than I had expected.

The only conclusion I can draw is that bottle bombs are mostly caused by one of two things:

1) Bottles that have flaws either from manufacturing or from mishandling

2) Bottling unpasteurized apple must and having fermentation proceed to near-dryness

If you have 4 volumes of CO2 in your cider, pasteurizing it to 65 degrees will get you up to about 13 bar.  My tests suggest that even beer bottles will withstand this unless they have a flaw.  

Jeff

[Terry Chalk]

Hi Jeff

I think your findings with European bottles and conclusions are pretty well supported by the distribution of failures in the Jo Morgan Teague PhD thesis that I mentioned. You might find a read of it worthwhile (apart from it being written way back when we had manual typewriters and became very excited when the first +-x/ calculators became affordable).

It is available via OhioLINK. I can’t find the search reference number that I had, but I can still get to it by Googling (Stress in Glass Bottles PhD thesis Teague)

I really haven’t been able to find anything more recent than this which covers bursting pressure for bottles. Having spent some time in the glass container industry back in the day, I can’t see much difference between the manufacturing technology described by Teague and that of more recent times. As far as I know, it is still a matter of heating and shaping Silica Sand, Soda Ash, Limestone and Cullet (recycled glass).

I also understand that stress builds up slowly in glass under pressure, and is relieved if the cause is removed (e.g. the pasteurising pressure drops as the bottles cool back to normal).

The Teague study was on bottles quite similar in shape and size to those commonly used today except that they had about 20% more glass. Having said that, the Teague bottle wall thickness was quite variable from <2mm to >4mm and it could well be that the thickness and hence strength of “modern” bottles is better controlled. Interestingly the study also covered “abused” as well as new bottles where the spread of failures was similar. 

To view this discussion on the web, visit https://groups.google.com/d/msgid/cider-workshop/051ce162-37ce-4d0d-a274-b1ec7ac82d0cn%40googlegroups.com.

[AW]

"I use 330 ml Tosca "birra" bottles that are about 330 g.  I set up a testing apparatus with a high pressure handpump for air guns that can reach 200 bars.  I fill the bottles with water for safety and put them in a filled, covered bucket of water to eliminate flying shards of glass."

Hey Jeff - How are the hand pump and the bottle linked together?

[jeff.k...@gmail.com]

I use this

https://www.bostonapothecary.com/for-sale-champagne-bottle-manifold/