Jump to content
ADI Forums
BigRed

Bottle Filling Calibration

Recommended Posts

We are planning on purchasing an Xpress Volumetric Filler for bottling. Since we are in the south and its hot, won't the 750 ml calibration be out of spec due to expansion from the heat? What is the best way to make sure we aren't out of spec per the TTB regulations?

 

 

Share this post


Link to post
Share on other sites

Weigh your filled bottles, this is true for volumetric and level filling.

We keep a set of bottles off to the side that have their gram weights written them, because bottle weight can vary significantly.  Fill, weigh, adjust, repeat.  

For example, 750ml of 80 proof weighs 712 grams.  Once you move to weight, the temperature no longer matters.

If you want to push to be very accurate, you can even do a check weight midway through your run, or if you feel that the fill level isn't quite right through the run (you will notice).  But, absolutely, every time you start.

Share this post


Link to post
Share on other sites
22 minutes ago, BigRed said:

We are planning on purchasing an Xpress Volumetric Filler for bottling. Since we are in the south and its hot, won't the 750 ml calibration be out of spec due to expansion from the heat? What is the best way to make sure we aren't out of spec per the TTB regulations?

It sounds like you may be confused about how this machine operates.  The machine is not calibrated to any specific volume size, but has a timer that controls how long its fills for each time you push the fill button.  You control/adjust the fill volume by changing/adjusting this timer.  (the timer can be adjusted to the hundredth of a second)  Every time you bottle (even if you only ever bottle 750mls) you should expect to have to adjust the timer based on the product you are bottling and the temperature, etc.

Do what Silk City describes above, and use that info to adjust the timer/fill volume.

Share this post


Link to post
Share on other sites

I also like do check by weight, but for fill height shelf appearances, on a hot day I put the first bottle in the fridge and see what it looks like when it hits 70.  It shouldn't matter if the weight is right, but you do get a little bottle weight variance, so I just consider it a double check method.  You are allowed quite a bit of variance on fill volume (compared to proof) from the TTB, but you don't want the visual of a short fill.

Share this post


Link to post
Share on other sites

Careful, you have flexibility in fill variance on a bottle by bottle basis, BUT - this does NOT apply to a batch basis:

§ 19.356 Alcohol content and fill.

(a)General. At representative intervals during bottling operations, a  proprietor must examine and test bottled spirits to determine whether the alcohol content and quantity (fill) of those spirits agree with what is stated on the label or the bottle.

and:

(b)Variations in fill. Quantity (fill) must be kept as close to 100 percent fill as the equipment and bottles in use will permit. There must be approximately the same number of overfills and underfills for each lot bottled.

Share this post


Link to post
Share on other sites

Thanks for the responses. I did know that it is timer based, so I can calibrate before the run to 712 grams (if 80 proof), then check again mid run or after an hour or two. Do you guys pull random samples to very weights as part of your QC program? What do you think is an acceptable variance on fill, 1-2 ml? 

Share this post


Link to post
Share on other sites

Same section in the CFR:

In no case will the quantity contained in a bottle vary from the quantity stated on the label or bottle by more than plus or minus:

(1) 1.5 percent for bottles 1.0  liter and above;

(2) 2.0 percent for bottles 999 mL through 376 mL;

(3) 3.0 percent for bottles 375 mL through 101 mL; or

(4) 4.5 percent for bottles 100 mL and below.

So for 750ml, it is +-15ml, and for 375ml, it is +-11.25ml.  But, keep in mind: There must be approximately the same number of overfills and underfills for each lot bottled.

Share this post


Link to post
Share on other sites
1 hour ago, BigRed said:

Thanks for the responses. I did know that it is timer based, so I can calibrate before the run to 712 grams (if 80 proof), then check again mid run or after an hour or two. Do you guys pull random samples to very weights as part of your QC program? What do you think is an acceptable variance on fill, 1-2 ml? 

Silk City has answered the legal requirements here from CFR 19.356 - must be done in 'representative intervals during bottling operations'. We do one about every hour. The results need to be logged via the requirements in CFR 19.600.

Share this post


Link to post
Share on other sites

I calibrate 1 bottle by weight. Then to get a very accurate average fill as required I put about 6 cases of empty bottles plus caps on my sales. Note the tare and start filling. At the end of each filled case I check the calculated weight against the actual. If numbers are close then keep filling and checking until 6 cases filled. Can easily get to within a fraction of a gram per bottle. I use Enolmatic vacuum filler.

  • reaction_title_1 1

Share this post


Link to post
Share on other sites
13 hours ago, Silk City Distillers said:

Weigh your filled bottles, this is true for volumetric and level filling.

We keep a set of bottles off to the side that have their gram weights written them, because bottle weight can vary significantly.  Fill, weigh, adjust, repeat.  

For example, 750ml of 80 proof weighs 712 grams.  Once you move to weight, the temperature no longer matters.

If you want to push to be very accurate, you can even do a check weight midway through your run, or if you feel that the fill level isn't quite right through the run (you will notice).  But, absolutely, every time you start.

750 mL 80 proof at 60F weighs 713.15 g      (60 F is standard temperature for USA, most of the world 20c is the standard)

713.15 g of 80 proof at 75F has a volume of 754.4 mL so they will appear overfull, which is BigRed's concern.

ie calibrate filler by weight.

Share this post


Link to post
Share on other sites

Pete,

Show your math...

------------------------

TTB Gauging Manual - Table 5 showing the weight per wine gallon at 60f 

https://www.ttb.gov/foia/Gauging_Manual_Tables/Table_5.pdf

Grams Per Liter = Wine Gallons Per Pound * Grams per Pound / Liters per Gallon

Grams Per Liter = Wine Gallons Per Pound * 453.59237 / 3.785412

80 Proof Grams per Liter = 7.92614 * 453.59237 / 3.785412

80 Proof Grams per Liter = 949.761

80 Proof Grams per 750ml = 712.321

Realistically, you probably don't have an accurate balance to the nth significant digit, so 712 grams.

 

 

Share this post


Link to post
Share on other sites

Ah ha, now I see it, so obvious when you finally see it.  You know stuff like this drives me batty, so I've been trying to get your number all morning.

Turns out, 712.321g = 713.15g (they are the same numbers)...

once you realize I'm doing the math the way the TTB expects (in air), and you are doing it OIML/internationally (in vacuum).

712.321g (in air) = 713.15g (in vacuum)

Where is @meerkat when you need him.

Share this post


Link to post
Share on other sites

Wow, thanks guys. This is really helpful. I need to get much more familiar with the gauging tables. Has anyone been audited/inspected by the TTB regarding filling? How exactly did they check? Seems like we need to be sure we choose bottles with extra room in the neck to account for the expansion.

Pete, which table did you use to determine the volume at different temperatures?

Share this post


Link to post
Share on other sites
28 minutes ago, BigRed said:

Wow, thanks guys. This is really helpful. I need to get much more familiar with the gauging tables. Has anyone been audited/inspected by the TTB regarding filling? How exactly did they check? Seems like we need to be sure we choose bottles with extra room in the neck to account for the expansion.

I would imagine they would check like this:

https://www.ttb.gov/ssd/pdf/tm516.pdf

TTB fill is assumed to be at 60F, note the fill determination procedure in the link, the TTB will correct volume for temperature before determining fill accuracy.

All decent glass manufacturers will design their bottles with the appropriate fill variation levels and headspace to account for typical temperature variations without massive fill level problems. Ask your glass manufacturer for the bottle drawings and they will typically fill levels and many times variation levels.  You don't need to do anything special when selecting glass, unless you are using glass that was manufactured for some other liquid. 

Share this post


Link to post
Share on other sites
22 hours ago, Silk City Distillers said:

Ah ha, now I see it, so obvious when you finally see it.  You know stuff like this drives me batty, so I've been trying to get your number all morning.

Turns out, 712.321g = 713.15g (they are the same numbers)...

once you realize I'm doing the math the way the TTB expects (in air), and you are doing it OIML/internationally (in vacuum).

712.321g (in air) = 713.15g (in vacuum)

Where is @meerkat when you need him.

You found the reason. Sorry but I forgot your (US) masses are measured "in air"

The "in air" inaccuracy is within TTB allowances per single bottle but maybe too large for long term accuracy.

I got my numbers from AlcoDens  by Meerkat   Just needed to click  "TTB (in air)" instead of my usual "OIML (in vacuum)"

Share this post


Link to post
Share on other sites
15 hours ago, BigRed said:

Wow, thanks guys. .......................

Pete, which table did you use to determine the volume at different temperatures?

http://www.katmarsoftware.com/alcodens.htm

  • reaction_title_1 1

Share this post


Link to post
Share on other sites
20 hours ago, BigRed said:

 Has anyone been audited/inspected by the TTB regarding filling? How exactly did they check? 

I've worked 3 places that have been audited/inspected by TTB, and they have never looked at/asked for fill level documentation or records. Which doesn't mean you shouldn't keep track, but with the error bars for proofing being 10x tighter that seemed to be a bigger concern. As long as you were diligent in tracking all the alcohol in and out and could prove you've paid all the taxes required, they didn't seem as concerned with 'gotcha'ing us on forms or technicalities. Your mileage may vary. 

Share this post


Link to post
Share on other sites
On 5/5/2017 at 8:33 AM, Silk City Distillers said:

Ah ha, now I see it, so obvious when you finally see it.  You know stuff like this drives me batty, so I've been trying to get your number all morning.

Turns out, 712.321g = 713.15g (they are the same numbers)...

once you realize I'm doing the math the way the TTB expects (in air), and you are doing it OIML/internationally (in vacuum).

712.321g (in air) = 713.15g (in vacuum)

Where is @meerkat when you need him.

Hi Sorry to drag this all back up again but Silk City can you show me the numbers used to make OIML match up with TTB. I keep running across the below numbers more.

40% @20C is 948.05kg/m3 

750mL* .94805g/mL= 711.04

also have to admit that I am a bit confused about the logic behind something being measured in a vacuum weighing more then in air. Thanks in advance for any help

Share this post


Link to post
Share on other sites

I was confused for a while about measuring mass in a vacuum. My thought was that if you had a beaker of alcohol and tried to weigh it in a vacuum it would instantly boil off. 

The reason to measure in a vacuum is because it is more accurate ( very slightly)

The difference is because of the buoyancy of air, that buoyancy is not always the same (extremely slight variation)

Buoyancy in water is a bit more obvious. If you weigh a 1 liter block of aluminium you get about 2.7Kg

Weigh it again when it is in fresh fresh water and it will only weigh 1.7 Kg because of the buoyancy of the water.

Weigh it in sea water and it will be less again, or in the Dead Sea and it will be even less.

Air creates buoyancy, that is how hot air balloons float.

Why does the rest of the world do our alcohol calculations based on "in a vacuum" when in practice we weigh in air?? Beats me.

(And a bit more confusing for some is the MASS of that block is 2.7 Kg everywhere, even in no gravity orbit!!)

ps. if you need to weigh a container of alcohol in a vacuum it needs a well sealed lid

Share this post


Link to post
Share on other sites

Thanks PeteB, I have done a bit of reading on it and am starting to get it. I am still interested in the actual math behind the 713.15 grams. Cause As I said when I looked up the OIML chart I kept reading the 948.05kg/m3 number.

Share this post


Link to post
Share on other sites

ReadeHud, your math is correct.  There are two reasons why the mass you have calculated for 750 ml is different from the values calculated by PeteB and Silk City earlier.

The first reason is that the values calculated by Pete and Silk were for 750 ml at 60°F but you are working at 20°C.  20°C is a bit hotter than 60°F and the spirit expands and for the same volume you have less mass.

The second reason is that although 80 Proof is equal to 40 Vol% at 60°F it is not the same as 40 Vol% at 20°C.  Alcohol and water have different rates of thermal expansion and as the spirit is warmed from 60°F to 20°C the alcohol portion expands (very slightly) more than the water.  This changes the volumetric ratio between the alcohol and the water.

80 Proof is equal to 40.07 ABV at 20°C, so when you buy equal volumes of spirit at 40 ABV from US and European suppliers you actually get a bit more alcohol from your US supplier.

A large part of the confusion between weights and masses in air and in vacuum is our loose use of the terms weight and mass.  We use the terms weight and mass interchangeably, but they are really two entirely different physical quantities.  Weight is actually a force, and is related to mass by Newton's second law ( F = m x a ). The most common way to determine mass is to actually measure the weight (i.e. force of gravitational attraction to the earth measured on a balance or scale) and then infer the mass from the second law.  Of course we don't actually do the math every time and the "a" term is built into the calibration of the scale and we simply read out the result as a mass in pounds or kilograms.

The mass of an object is not affected by the presence of surrounding air, water or other fluid.  Nor is it affected by the force of gravity.  But weight is obviously affected by both.

Using gravity to measure weight and inferring the mass is not the only way to measure mass.  When astronauts spend extended periods in space it is very important for them to know how their mass is changing (for health reasons) but because they are weightless in space a normal scale will not work.  They measure their "inertial mass", which is actually the same as the "gravitational mass".

If all this is not sufficiently confusing, try using the Canadian alcohol tables which measure ABV using the "in vacuum" density value, and then use the "in air" density value to determine the volume of the spirit.

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×