abv calculations, hydros vs lab

[barleyfreak]

Just sent 3 beers for lab testing. The results all show the abv less (sometimes much less) than my calcs based on pre/post gravity drop/hydrometer readings.

Beer abv lab/in house

IPA -- 4.6%/6.6%
Double IPA -- 7.4%/8.3%
Porter -- 5.3%/5.9%

I figured that the results were pretty likely to be different than I calculate, but wonder why so much, in particular, the IPA. In that case I can only assume that something was wrong with my sample, or the test/test results.

FWIW -- I have lab certified plato hydrometers, 0-8, 8-16, 16-24. I have a digital plato refractometer that I use in conjunction with the hydros to measure OG. They are almost always within .1 of each other. Finish gravity is consistent, and has remained so even using a different hydrometer.

Thanks for any feedback.

[MikeyB]

The obvious question...

... what formula(e) are you using to calculate ABV? Some of the common methods are not completely accurate, especially as OG increases. On a related note, this is a good resource, especially when blending multiple gyles into one fermenter, ie double-batching: http://www.beercalc.com/

[barleyfreak]

... what formula(e) are you using to calculate ABV? Some of the common methods are not completely accurate, especially as OG increases. On a related note, this is a good resource, especially when blending multiple gyles into one fermenter, ie double-batching: http://www.beercalc.com/

I use beersmith, mostly. With a 15.1p OG and a 3.1p FG, it should be ~6.6%. With any other brewing abv calculator I find, or formula, like this,

( ( 1.05 x ( OG – FG ) ) / FG ) / 0.79 x 100 = % ABV

or this,

ABV = (OG - FG) * 131.25.

I get roughly the same figure ~ 6.6%. A far cry from the 4.6% figure the lab returned.

[dick murton]

Using the UK regulations, I get 6.37

So, a number of possible reasons, but, probably the main one is simply that your yeast has grown more than expected for all these calculations, and if it is producing yeast, it does this at the expense of alcohol. High temperatures causing high yeast growth, overpitching, over oxygenation compared to what it really needs, all produce slightly fast fermentations and higher yeast growth.

Possible poor mixing during wort transfer from kettle / WP, so giving erroneously high collection gravity samples?

If you check the gravity every 6 - 8 hours, you should easily pick up if this is the problem. When do you pitch the yeast compared to taking the initial gravity ex FV? - unlikely but not impossible if your wort transfer rate is slow.

Accuracy of thermometer giving you the wrong temperature correction? You are correcting for temperature aren't you? (serious question as I have come across a couple of people not correcting)

[barleyfreak]

I may be over-pitching or over-oxygenating. I do not do cell counts at this point. I pitch roughly 2.5 to 3 lbs/bbl. I oxygenate this strain at about 6 LPM for roughly 30-35 minutes of the KO time, which ranges from 35-50 minutes depending on time of year. Pitch at 64, let rise to 68 during the ferment, which lasts about 5 days until TG.

Not sure I understand what you mean by producing more yeast at the expense of alcohol. As for OG samples, I take a pitcher from the rolling boil in the kettle right before flame out, cool it in the cooler for 1/2 hr or so (with a towel draped over the top to hopefully minimize evap). My assumption is that there is pretty good mixing during the rolling of the boil. Perhaps I should also take a reading of the knocked out wort from the FV after transfer and see how close it is.

All subsequent readings are from the FV sample port, daily. Gravity drops all look normal, and pretty consistent batch to batch. I always correct for temp. The plato hydros I use have a temp correction reading built on.

[jebzter]

I would suggest always taking your OG samples from the fermenter, you will also notice if there is a leak in your HX this way.

[barleyfreak]

lab gave me incorrect readings

The lab got back to me and said that the Double IPA reading they gave me was wrong (it really was 7.9% -- closer to my 8.3% calc), and that they reran the test on the IPA and it came out to 6.5% which is almost identical to my 6.6% calc. So this is now much ado about mostly nothing. Though I am now going to take multiple readings to see if my OG is consistent.

Thanks for the feedback.

Dave

[dick murton]

That's good news. It didn't quite hang together as it was described. Should it happen again, you know to sense check the results with the lab, or even change lab if they keep doing silly things.

So your difference could be due to more yeast growth than you would normally get, measurement errors, and simply the fiddle factor for this yeast is not the same as the other yeasts you use, and so you have to develop a fiddle factor ((OG-PG) x A) = % ABV for this specific yeast. They all vary slightly anyway, and the quoted values are all based on standard conditions. So if your conditions are not as per standard - well the fiddle factor will change a bit.

[TL Services]

Quite a few years ago I did some work on this as a result of differences between measured and calculated abv. This confirmed that some of the more basic 'one size fits all' calculations are very inaccurate at higher abv. Typically they tend to go away after around 5%.

This was based around the '(OG - FG) x factor' calculation compared to both NIR and GC analyses.

Glad to hear you got your issues cleared up and the lab came clean. Errors can happen even in the best places, so always worth querying any odd results.

[barleyfreak]

Quite a few years ago I did some work on this as a result of differences between measured and calculated abv. This confirmed that some of the more basic 'one size fits all' calculations are very inaccurate at higher abv. Typically they tend to go away after around 5%.

This was based around the '(OG - FG) x factor' calculation compared to both NIR and GC analyses.

Glad to hear you got your issues cleared up and the lab came clean. Errors can happen even in the best places, so always worth querying any odd results.

I wonder how many of the abv percentages I see on tasting room boards are accurate. Most small brewers likely don't have the capacity to test in-house. And the cost for doing so on every beer would seem somewhat prohibitive, unless the beer is to be packaged. It does seem that the calculations I and others use, if truly inaccurate after 5%, overstate the abv. Or they do in my limited sample set.