Alcohol Tester

I presume you meant "accurate enough". Answer: Refractometer

Dick

Wrong answer! Try an "Ebulliometer"

Charlie

Dick Adams wrote:

If the truth be known, I believe most of them use no more than the average winemaker uses. A hydrometer or refractometer to determine sugar content and a table to determine alcohol. This is especially true on the east coast. Big wineries, especially on the west coast, will send samples off to laboratories. Some will use an Ebulliometer, though I am told it is not really that much more accurate than the tables and the labs use much more accurate (and expensive) equipment now days.

Ray

Wrong answer! Try a frat party.

:-)

I thought that only gave a percentage of sugar from grape juice. Did I miss something? (as usual?!)

LOL!!! If the balcony falls off the house while the party is in progress, you know it's right! ;-)

Yes, it give percentage of sugar and yes you missed something. From the percentage of sugar that the yeast consume you can determine how much alcohol they generate. There are tables for this that will give it within a percent or two. Some people claim they are closer than that but I am not convinced, especially if you only use the table. For most amateurs this is good enough. There are instruments that will measure alcohol directly after fermentation. These are usually beyond most amateurs. You can also send samples to a lab and they will test it for you. But then why. All you need is a ball park number to balance your wine. If you would like some calculation methods with detailed descriptions, you can read "The Unified Theory of Gravity" by Dr. Ray Calvert, Wine Maker Magazine, April-May, 2004, or let me have an email address and I will send you a preprint.

Ray

Nope!!

Wineries use ebu's!! "Big" wineries use the $20,000 enzymatic gee-whiz-a-ma-trons the labs do. When a winemaker puts 13.8% on thier label they make "damn" sure it is not really 14.01% !! Else the Fed's come over with a BIG citation.

Charlie

The other method not discussed here is to use a proof and tralles hydrometer in a properly treated sample. It's less expensive and as accurate as an ebulliometer, but very tedious.

Joe

pcw wrote:

The precision of the numbers is only an illusion. AFAIK American laws allow for 1.5% v/v error in alcohol estimation, so something that's labelled 13.8% can very easily really be 14.5% or more. This is precisely because the exact number is so hard to measure without very expensive equipment - not to mention that it might vary from barrel to barrel, so it will be an average number in any case.

Pp

You missed my point! If a winery "labels" a product at 13.8% (i.e. submits a COLA and pays excise tax accordingly)...and the wine is actually >14.00% ...even 14.01% ... they're in BIG TROUBLE!! If they were to do this repeatedly, they could, and probably WOULD loose their permit! Every winemaker I know uses an Ebu (or a Lab) to "confirm" what they put on thier label.

As I recall, the original poster asked about "wineries".

Charlie

pp wrote:

I didn't miss your point and I don't think your info is correct. Yes, there is a higher tax for 14% and above, but this is based on the label info, and the law has a pretty big leeway of 1.5% alc. So I don't see how the wineries could get into trouble if they miss the "real" value (whatever it is) by 0.2%.

Ebulliometers are imprecise because they are influenced by external conditions - do some research on this. Even if they were more precise, how realistic would it be to expect that wineries could adequately measure their alcohol content to 0.01%?

I am not saying wineries don't use ebulliometers. They're fine precisely because the law has that big margin of error built into it.

Finally, how would you address my point that the alcohol level is average by definition because different barrels/batches of the same wine can differ substantially in their alcohol level, would you want to do a precise test on every single batch... and print different labels?

All this is to say there is a good reason for the 1.5% leeway. I'm sure it would have been tighter if it was practical to narrow it down.

Pp

Excuse me!! You DID miss my point!! And "what" info did I give that is incorrect? If you think the ATF's "leeway" of 1.5% will allow you to "label" a wine at 13.8% that they can pull off the shelf and "measure" at 14.01% you are sadly mistaken!! As I said, if you repeatedly do that, they WILL take away your permit! There is ZERO tolerance for "crossing" tax brackets!!

Your argument about differences between barrels or batches is meaningless in a commercial winery. We generally blend or "cuvee" from multiple barrels and check the alcohol on the blend for the COLA. And we, and everyone else that I know, use an ebulliometer because we believe it to have the best accuracy for the buck!! When I calibrate my ebu to atmospheric pressure with water, I can get repeated measurements on a given sample within +/- 0.2% over the course of an hour or so. In these days of longer hang time we're seeing more and more wines in the high 13's. If I got multiple readings of 13.8 or

13.9 with the ebu, I would send a sample to ETS labs before committing to the number I put on my label, and pay the "right" tax.

Hence, my original response to Maurice!

Charlie

Hi Joe,

Thanks for bringing this up!! When I recieved my lab training I read about such procedures, but we only "practiced" ebulliometry. Zoecklein mentions a few methods involving distillation (into a Kjeldahl flask?) of the sample but I don't see "proof and tralles" mentioned. I assume it is a distillation followed by hydrometer? This would be a special, close range hydrometer?? My problem with hydrometers for "accurate" readings is one of "interpretation" of the meniscus ... often no more accurate than a whole division on the scale. With a mercury thermometer (as on my ebulliometer) I can interpolate down to a half a division easily! ... which is 0.05% alcohol. Another difference is the "efficiency" of a refluxer (ebulliometer) vs a distillation coil. I use ice water in my refluxer and I believe it returns nearly 100% of the EthOH to the sample. I don't think a distillation coil is as efficient. Anyone?

Anyway, any info you have on the method you bring up would be interesting!!

Thanks,

Charlie PCW

Hi Charlie,

Both hydrometry and ebulliometry have an uncertainty in the range of

0.3% when measuring alcohol by volume as I recall.

You are right on in your post, a Proof and Tralles hydrometer is calibrated for measuring proof and ABV (Tralles). You can also get narrow range Tralless, they make the most sense for wine. I have seen

10 to 15% ABV narrow range hydrometers.

The biggest problem using this method is getting a good seal between the boiling chamber and the condenser, efficiency doesn't matter since you distill most of the sample over. It's not like your refluxer in that sense, but if the seal is bad you can lose the alcohol and get false low readings.

(By the way, this should NEVER be used to distill for consumption - at least in the US if anyone is thinking about that. It's not only illegal it's stupid. Liquor is cheap, impisonment and loss of the house and everything in it is pretty expensive. It's illegal for more than tax reasons, there are byproducts of distillation that are poisonous present using this procedure, it is NOT for consumption. I'm not going to discuss that further, it's just the reality of the situation.)

You can get a good hydrometer from Kessler and a whole boiling chamber/ graham condenser setup is usually around $100 US if you are ok with all that in glass, it can break pretty easily. Out the door it's probably more like $200 US if the stands, standard volume flask or graduate and burner need purchased too.

You can use the residue for a disolved solids test if desired too, so it's a 'multi-tasker'.

Anyway, all you are really trying to do is remove the dissolved solids in this method by leaving them behind. The main interferences are sugar, (just like your ebulliometer) and a high acetic acid level.

You start with a standard volume of say 200 ml at 60F, boil most of it over and restore the volume to 200 ml at 60 F with distilled water and measure with the special hydrometer.

As to reading the meniscus on the hydrometer, I agree most people can go wrong when it comes to reading them, it's almost an art.

Unless specifically told otherwise by the manufacturer, the correct procedure (an abbreviated form of what NIST does follows):

• ensure the hydrometer is very clean and at the same temperature as the fluid under test, (usually 60 F is prefered)

• measure the temperature of the fluid under test

• spin the hydrometer as it is lowered into the fluid under test.

• ensure no bubbles adhere to the hydrometer

• once it stabilizes, lower the eye below the fluid level and look straight across, raise the eye level until the ellipse formed by the fluid boundary just disappears, you are now at the level of the fluid boundary.

• read the hyrdometer as it passes through this level and apply temperature corrections if necessary.

(NIST has a PDF on hydrometry available on the web, it does a better job of explaining this.)

I have a set of DuJardin Salleron hyrometers that specify reading the top of the meniscus; it's odd but they are precision instruments so I do what they say with those.

Zoecklien et al and Margalit both have pretty detailed procedures in recent publications as you mentioned. There is a company in New Zealand that has a pretty good description on the web too, it might be Monash or Monarch Scientific, something like that.

HTH.

Joe

pcw wrote:

Zoecklein

flask?)

assume