Re: A Question about Home Wine Making and Alcohol Content.

Hi Bill

Thank you. I know that you understand this, but the way you framed your statement leaves it wide open to be misinterpreted by others here. CJJ Berry understood it this way and_still_managed to get messed up. What folks need to understand here is that the influence of alcohol changes the reference point (if you will) and that this has to be compensated for to restore the reference point _back_ to SG 1.000 so that such a calculation _can_ be made. IOW - so we are comparing an apple to an apple and _not_ an apple to an orange. Clear as mud, right?? Maybe you can come up with some IOWs to help explain this. TIA

Frederick

Hi all

Gone for a while hunting (got a nice 7 point). Now that I am back, I find so many posts that I hardly know where to begin. I_would_like to answer a few of the posts in this thread however, in the hope that someone besides Bill will arrive at an understanding of this topic.

Everyone should understand that the numbers I am using here are_very_ generously rounded off in an effort to keep this simple. After all, the only point I am trying to make is that there is NO _fixed_ relationship between SG and PA. Sg measures _total_ solutes (both fermentable AND non-fermentable), while alcohol is calculated using _only_ the fermentable portion of that total. The exercise outlined above is just the simplest way I could think of to demonstrate this in a way that everyone here could do, even if they are "theory" challenged. ;o) (nothing better than something you can see with your own eyes)

Ben

If your theories are so compelling that you can doubt the evidence of your own eyes, there is little I can do to help you understand what is going on with this stuff. My approach to winemaking is strictly pragmatic and I use my own empirical data to make these kinds of estimates. My own understanding of theory is imperfect so I can't address the subject that way. Might I refer you to the UC Davis site where, in the second to the last para on page 10, you will find an example of the formula they use to calculate the PA numbers that are used on our hydrometers. The pages leading up to that example contain a _very_ good explanation of how they arrive at those numbers. Hopefully this will also explain why I have been saying that you can"t get more alcohol in the wine than the original PA perdicts. HTH

Frederick

PP

I have been over the material in this book a number of times with several people (I thought you were one of them). But - let me try to plod through it again. Comments interspersed:

Such formulas use an ADJUSTED final gravity and_not_the ACTUAL final gravity !!

Read this again_veeeery_carefully. G is defined as having a reference of

Not what I use, but "good enough", IF Gf is adjusted.

It _does_have this effect when Gf is _not_ adjusted.

I would view this statement as wrong in this particular context. Only _part_ of the total drop is caused by the consumption of fermentable sugars. The _rest_of the drop is due to the effect of alcohol on our end SG reading. However, (to avoid another nit) if this statement is taken to mean that the amount of drop is dependent on the amount of fermentable sugars available to the yeast, then yes I can buy that.

Wrong. PA is based _solely_ on the amount of fermentable sugars available to the yeast. Let me say once more - there is NO FIXED RELATIONSHIP BETWEEN SG AND PA !! SG is a measure of _total_solutes (not just the sugar).

Nope.

Estimate yes, rough no.

Every country and every state that has any interest at all in having a winemaking industry has government funded labs which lack for nothing by way of facilities and the best and most modern equipment. These are staffed by people who spend entire_life_ times dedicated to winemaking research. This being the case, how can you even_imagine_ such data does not already exist and has not already been incorporated into the things we use ?? (eg the PA scale on our hydrometers, etc..)

Wrong, wrong, wrong !! The final gravity_does_ need to be shifted, but the original gravity is_not_effected by the presence of alcohol so it does_not_have to be shifted. (It_already_has a reference point of

Nope - see above. In view of my previous answers here, I won't go on with this at this time. HTH and HTMS

Frederick

Ben,

I think I know partly why (although I haven't checked the exact mathematics). I seem to recall from my High School Chemistry class that water molecules and alcohol molecules fit into each other in such a way that the total volume of the mixture is less than the volumes of the two liquids unmixed. That is, if you add 20 ml ethanol to 80 ml water, you'll get noticeably less than 100 ml as a result. It's a difference of a few percent, IIRC. That means that the mixture has all of the mass of the two liquids, but less than their additive volumes. It seems to me that this would increase the mixture's density by about that same few percent, maybe accounting for the difference between the predicted density of .975 and the observed density of .990 . It's only a difference of about 1.5% that we're talking about.

Supporting data can be seen in the chart at the end of this link,

It looks like this is another case of Reality thumbing its nose at our neat predictions.

Took me a while to figure out I had to switch to Google beta to be able to post... and now I don't seem to be able to quote the previous message to respond to explicit points. Hmm, hopefully that'll get ironed out soon.

Frederick, can you supply a better reference than "the UC Davis site" - I can't find any reference to PA articles there, nor anything resembling page 10. I'd really like to read this article as it seems to be the primary source of your PA stance.

Thx,

Pp

Hi PP

First. I really must apologize for terminating my previous answer so abruptly. Being pressed for time is not much of an excuse. Sorry.

Second. I can't seem to find my "links" list this morning. So - I have posted a separate request to the group for this link. Hopefully one of those folks will post it here (or there) for us. If not, I will have to go find it again. The reference is not a separate article on PA. It is contained in their basic winemaking book (pdf file).

It's gong to make this post *VERY* long but I will copy and paste my previous answer here so we have a point of reference: (See below the "end copy" for my further comments)

......................

PP

I have been over the material in this book a number of times with several people (I thought you were one of them). But - let me try to plod through it again. Comments interspersed:

Such formulas use an ADJUSTED final gravity and_not_the ACTUAL final gravity !!

Read this again_veeeery_carefully. G is defined as having a reference of

Not what I use, but "good enough", IF Gf is adjusted.

It _does_have this effect when Gf is _not_ adjusted.

I would view this statement as wrong in this particular context. Only _part_ of the total drop is caused by the consumption of fermentable sugars. The _rest_of the drop is due to the effect of alcohol on our end SG reading. However, (to avoid another nit) if this statement is taken to mean that the amount of drop is dependent on the amount of fermentable sugars available to the yeast, then yes I can buy that.

Wrong. PA is based _solely_ on the amount of fermentable sugars available to the yeast. Let me say once more - there is NO FIXED RELATIONSHIP BETWEEN SG AND PA !! SG is a measure of _total_solutes (not just the sugar).

Nope.

Estimate yes, rough no.

Every country and every state that has any interest at all in having a winemaking industry has government funded labs which lack for nothing by way of facilities and the best and most modern equipment. These are staffed by people who spend entire_life_ times dedicated to winemaking research. This being the case, how can you even_imagine_ such data does not already exist and has not already been incorporated into the things we use ?? (eg the PA scale on our hydrometers, etc..)

Wrong, wrong, wrong !! The final gravity_does_ need to be shifted, but the original gravity is_not_effected by the presence of alcohol so it does_not_have to be shifted. (It_already_has a reference point of

Nope - see above. In view of my previous answers here, I won't go on with this at this time. HTH and HTMS

Frederick

.................

To continue:

Please allow me to ramble on for a while. The reference you are using went out of print about a quarter of a century ago because the section on alcohol is a total mess. They confused theoretical values with "real life" values and the result is a disaster. I don't have this reference but I have discussed parts of it often enough, and I did manage to briefly get my hands on a copy. Only had about 30 minutes to review the alcohol section before I had to return it. Frankly, I laughed until I had tears. It's quite possibly the tallest mountain of bull chips in all of winemaking literature !! To understand this, one need only compare the PA numbers in their chart with the PA numbers we find on our hydrometers. Theirs are the pie in the sky maximum_theoretical_values while the PA numbers we find on our hydrometers are the realistic_practical_values that we need for_practical_applications in our every day winemaking. Simply put, in real life, sugar is converted to alcohol by a wasteful, inefficient, living organism which is FAR from perfect !! Thus, "theoretical" values are useless in a pragmatic, practical world. HTMS

Earlier we were discussing their "short" formula, but you also mentioned a different (longer) formula. Even as screwed up as Duncan was, even he recognized the fact that alcohol_does_have an effect on total drop. You will find it in that formula expressed as: "...(Go - 7)..." (eg original gravity_minus_7 gravity points). Now, it really doesn't matter if we subtract the correction from the original gravity or add the correction to the end gravity. The effect is the same in that we_decrease_the amount of total drop thus compensating for the effect of the alcohol on our end SG reading.

I suppose (if he hadn't been using theoretical numbers) we could modify his short formula to incorporate this as follows:

Either: Gs - (Gf +7)/7.4 = Ae (eg end alcohol) Or: (Gs-7) - Gf/7.4 = Ae ( ditto )

I really hate discussing this reference because I always have to convince people that authors are people, people make mistakes, and some of these mistakes find their way into print - so - you CAN'T believe everything you read !! If you base your work on a reference where the author got it wrong, your own work will turn out to_also_be wrong !! It once appeared in writing that the earth was flat. I don't use that as a reference any more.

We can go on discussing this stuff this way but I suggest that you put that book aside and find yourself some other reference. ANY other reference !!

HTH and HTMS

Frederick

PS - Please never be offended by anything I say. I really_am_only trying to be helpful here.

frederick ploegman wrote: ..... because the section on alcohol is a total mess.

Frederick,

You're absolutely right. I still have a copy, and it's a remarkable source of mis-information in some areas, like alcohol. Even as a self taught winemaker 30 years ago, I knew there was something quite wrong with their either words or my works. It took me a while to realize it was them, but 'til that point, they had my attention. Having learned a bit since then, the book lies unused. Other sources do it better, like Lum's, or Jack's, or Ben's sites.

Mike MTM, Cokesbury, New Jersey, USA

Hi Frederick et al.:

I've finally put together my reply - it's long as well, so I've decided to just summarize where we were in the discussion and do it the post in

Here goes:

Let me briefly summarize the issue as I understand it. The current round of PA discussion was trigerred by Jack's post, in which he described a method of determining PA by:

Frederick then put this into question by claiming that no such table exists and people who tried to produce anything along these lines failed, the reason basically being that alcohol produced during fermentation shifts the reference point below sg 1.000. Thus, any specific gravity drop below 1.000 is caused by alcohol and as such should be disregarded because it contributes nothing toward the amount of alcohol in the wine.

I then argued that while it is certainly true that alcohol affects the final sg reading, this in itself doesn't preclude one from designing a formula/table that works exactly as Jack described it. The idea is that instead of basing the formula only on the amount of sugar in the must, the formula should be based on the total gravity drop during the fermentation. I offered Duncan & Acton's formula (the simple version) PA = (Gs-Gf)/7.4, where G stands for gravity, or sg*1000, as an example.

At this point, the discussion went basically off track, most likely because of the negative connotations that some of us in this thread have with D&A's work on PA. So what I thought I'd like to do in this post is to stop talking about things on theoretical level and do what Frederick originally asked for - provide a formula/table that works as Jack described.

For the sake of this discussion, I'm adopting the following assumptions:

For simplicity, I will work with the sg range of 1.060-1.120, as this should more than cover the wine alcohol range. According to my hydrometer (regular triple scale model), the PA levels for this range (start, middle, end) are: sg PA

I want to end with a formula that uses both initial and starting sg and that predicts the same (with a reasonable deviation) PA as hydrometer in the sg range 1.060-1.120. The starting point will be D&A's formula, but instead of the constant divider 7.4, I will first use an unknown value X: PA = (Gs-Gf)/X or PA = 1000(SGs-SGf)/X

To determine X, I will make a third assumption, based on experience:

This gives us (using only the gravity version here): PA = (Gs-Gf)/8.33 as the formula we were looking for. The only remaining question is to see how its predictions compare with the hydrometer values. This depends on the average final gravity for dry wines, but we can again assume as a benchmark here the value of 0.990: sg PA hydrometer PA formula

So the results are within +/- 0.4% of the hydrometer value in the PA range 8-16%. For the more typical wine PA range of 10-14%, the maximum difference is only +/- 0.2%. That's about as close as two different formulas will get in practice, but if we wanted to get really fancy, we could make the divider X depend on initial sg as D&A's (complex) formula does. I won't go into that, but it should be fairly obvious that that could be done to reduce the difference between the 2 approaches even further.

To recap, we've ended up with a simple formula PA = (Gs-Gf)/ 8.33, which:

- needs only 2 sg readings - initial and final,

- works equally well for dry, off-dry, or sweet wines,

- by subtracting the 2 gravity readings, automatically accounts for the effect of soluble solids other than sugar on the calculation, and

- is within +/- 0.2% of the PA hydrometer values in the typical wine PA range 10-14%.

I hope this clearly shows that there is no methodological problem per se in using final gravity readings for determining the PA content of wine, one only has to know how any given formula/table/scale should be correctly used. And that brings me to the last point, really an aside. My hydrometer (as I said, a typical triple scale hydrometer, made in Japan) actually came with an instruction sheet in the tube, and this is what the instructions say:

"In order to determine the alcohol content of a wine, you will need to make two readings, one before fermentation commences and another after fermentation stops. Subtract the final figure from the first reading and you will have the alcohol content percentage by volume. For example: 13% - 1% = 12%."

So here we have a standard winemaking tool, that Frederick assumes to be authoritative and based on lots of practical data and research, with instructions that clearly contradict the assumption that the hydrometer PA scale is only based on the initial sg reading. No wonder this topic is confusing to so many winemakers!

Hope this all makes sense,

Pp

Fredrick, pp, et.al.,

Nice discussion...I've enjoyed it all. But, it seems we are just back to where it started. The potential alcohol markings on a hydrometer, pp's formula and a simplistic alcohol formula based only on the starting gravity (assumes fermentation to dryness at 1.000 sg) all give alcohol percentages essentially the same and they are only estimates;

Theoretical starting Spec. Grav. 1.060 1.090 1.120 potential alcohol from hydrometer 8.0 12.0 16.0 pp's formula 8.4 12.0 15.6 simplistic calculation using only OG 7.9 11.8 15.8

So, I guess it's use whatever you want unless you can actually measure the alcohol with an ebulliometer or some other such instrument.

Bill Frazier Olathe, Kansas USA

Bill:

The hydrometer readings and the simplistic OG calculation are probably the same, those numbers look like what's actually on the hydrometer. My formula was "calibrated" to the hydrometer values, so it makes sense that in all 3 cases we'd get basically the same predictions. By necessity, any formula will be only an estimate of the actual alcohol level in the wine, so the best one can hope for is for the prediction to be within a reasonable margin of error from the actual value.

In terms of what formula to use - I'd agree that it's up to everybody's choice and what we believe is a good estimation of the PA value. The main point is to use the formula/scale correctly, i.e., to avoid artificially increasing or decreasing the PA by using final gravity with tables designed to work only with initial gravity and vice versa. Personally, I use Margalit's formula PA = B * 0.57 because I trust his work in general, it's the most recently published reference, and it seems to be best supported by references. This gives slightly higher values than the hydrometer readings.

As for measuring the alcohol by ebulliometer - I've never tried that but from what I've heard, it's very hard to get a better precision than

Pp

William Frazier wrote:

percentages

measure the

Bill Frazier from the flat lands of Kansas wrote (in part):

OK, how does an Ebulliometer work? Is it reasonably possible to build one?

Dick

Fotunately I had my mug in my left hand when I read this. So when the Mead spat out of my mouth, it went into the mug and not into the keyboard. I've cleaned a keyboard of beer and wine, but Mead would be a sticky mess.

A pox upon your house - may your next batch taste somewhere between Steel Reserve Malt Liquor and Chateau Mad dog and may your next lover be a New York Yankee fan who drinks T.J. Swan.

Dick

Dick - Go to

Bill Frazier Olathe, Kansas USA....11F this morning

Bill, It was -7 at my humble abode this morning (and that doesn't include the wind chill factor). Darlene Wisconsin

Too cold for me Darlene. However, my wife says winter is now over since the days start to get longer. She's such an optimist. I'll tell her about the balmy temperature a few miles north. Even if cold we lucked out on the big storm. Here in KC bone dry with a south wind. I've got a golf date on Monday so looking forward to better conditions.

Bill Frazier Olathe, Kansas USA

Bill, Your wife is right, the days are going to start to get longer. Golf, huh...so you all missed the snow and the cold. Guess that front went between us, because we didn't get the snow either, but the cold wave is here. They say, by Monday, we'll be in the balmy upper 20's - heat wave for sure. The only thing keeping us warm is the Green Bay Packers win over the Vikings. Happy holidays. Darlene Wisconsin (just north of Milwaukee)