Hydrometers, Refractometers and Accuracy

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

Brooothru

Well-Known Member
Joined
Feb 16, 2012
Messages
2,336
Reaction score
2,259
Location
Either in the brewery or on the road
Grad a bag of chips and a six pack. Sorry, this is a long one, but I hope somebody can help give me some guidance on a topic that's been bothering me for awhile.

For the first 30 or so years of making wine and later brewing beer I relied on a trusty hydrometer to take gravity measurements. Broke a few like everyone else, and as my 20/20 vision began it's inevitable age-related decline it got harder and harder to discern where the meniscus intersected the hashmark. Plus I grew weary of dumping my precious samples down the drain 250 ml at a time. Over the years that amounted to a significant volume of liquid that never had the opportunity to stimulate my kidneys.

So I bought a refractometer. Neat little instrument. Easy to use, convenient focusing eyepiece for my failing eyes, and quite accurate after applying corrections for wort refractive indices and the presence of alcohol (a process made very simple with the Brewer's Friend Refractometer Calculator). And best of all, that 1 ml sample I used to flood the platen was only 1/250th of what it took to fill the hydrometer jar! Clean-up is a snap.

Fast forward 8 years to today. A local brewery is brewing and marketing one of my beers from a recent competition. I met for a few hours with the head brewmaster and his top assistant to go over my recipe, process and procedures in detail and to decide how best to replicate it using their facilities while still maintaining fidelity to the original. While going over the brew log sheet the assistant asked me how accurate I thought my gravity numbers were, and I told him within 1 specific gravity point measured with a refractometer. He replied that a refractometer wasn't accurate past the initial unfermented wort reading.

Without wanting to be challenging to his professional credentials and experience (and perhaps a little intimidated by the potential hole in my own knowledge base) I simply replied that I had calibrated the refractometer with 15C distilled water and used a Wort Correction Factor of 1.04 to compensate for the different refractive index of maltose over the sucrose/glucose baseline, plus used an online calculator that compensates and corrects for the presence of alcohol in fermented beer. He kind of brushed it off with an, "Oh, O.K.", and we moved on.

But that encounter got me to really start thinking and questioning my processes. Sure, when I first started using the refractometer I would regularly check the calibration with distilled water, and I would routinely crosscheck the refractometer with one of three narrow-band reading hydrometers. They were always within a point or two S.G., but since it's always difficult to see a clear line of refractive index while the wort is actively fermenting I could never be sure of accuracy except for OG and FG. And since OG and FG were either exactly the same or no greater than 1 point, I gradually stopped using the hydrometers. They've been collecting dust in some corner of the brewhouse for at least 5 years. Until this morning.

I'm re-brewing the beer at home that the local brewery is brewing at their facility, since the original sample bottles are now nearly 8 months old (it's a lager, and they've been stored in the dark at 38F, but still...). The new beer has been spunding at ~15 psig/50F for three days and still needs a diacetyl rest (still a lot of phenols and off-smells for the yeast to resolve) but I wanted to check how close the fermentation was to being 'done'. I had previously done a FFT of the wort on a stir plate at room temperature for 40 hours which predicted a FG of 1.009. I'm using WLP-830, second generation, coincidentally harvested from the original award winning beer and was a bit surprised by the level of attenuation. That said, I feel like FFT results are often 2 points lower than what's happening in the actual fermenter (due to over-pitching, higher temperature, constant agitation on the stir plate, etc.), so I expect FG will be around 1.011. This morning I drew a large sample from the fermenter, and after settling, de-gassing and warming to room temperature I measured both with a refractometer as well as a hydrometer. Two identical readings of 1.013.

I was pleased, especially with the Brewer's Friend calculator, but not surprised. I had previously calibrated the hydrometer at zero delta with distilled water (I've had some in the past that were off by as much as +/- 3~4 points), and I have never had the refractometer fall out of calibration.

(TL; DR):ghostly:

So after this long-winded story, here's my question about the online calculator:

The generally accepted "rule of thumb" difference between the refractive index of 'sugar' (brix) and maltose is 1.04, but that number sometimes needs to be fine-tuned to the specific refractometer (I did it once a long time ago, but I don't remember how I did it. I calculated mine to be 1.038 IIRC. Close enough to 1.04). The online calculator is comprised of two parts: Part I measures the OG Brix R.I. and applies the Wort Refractive Index factor (WRI) correction to yield the OG specific gravity. Part II measures FG Brix R.I. when alcohol is present in the liquid being refracted, and then further corrects by applying the WRI correction to yield FG specific gravity. It also displays the corrected OG as well as %ABV of the sample being measured. Often I've seen the value of the OG displayed in Part II different than the value for OG calculated in Part I by as much as 2 specific gravity points even though I'm using the same WRI correction for both calculations. For this to be true, the original gravity as measured by the hydrometer must be wrong since the measured Part I OG (hydrometer) and the calculated Part I OG (refractometer WRI corrected) are the same. However, the FG calculation in Part II is always in agreement with the hydrometer measured Part II sample.

So does the wort refractive index (WRI) in Part II change as maltose is converted (decreases) by fermentation? I can change (reduce) the input value of WRI in Part II from say 1.04 to 1.02 and the "corrected" Part II OG value will match the Part I OG value (1.04 factor still applied). This would seem like a simple fix, but there's nothing in the directions of literature that suggests that this happens, so I'd just be plugging random numbers to achieve my desired outcome. Bad science. Bad math.

The problem is that my hydrometer measurements agree precisely with the refractometer calculations, with the exception of Part II's "corrected" OG calculation. I'd simply ignore the small differences as being within the margin of error, but the %ABV calculations subsequently get skewed by as much as 0.2% ABV which is a fairly significant difference.

I tried to read up on the topic and even 'popped the hood' to try to examine the underlying math but quickly got lost in the weeds. Can somebody provide a simple explanation, or should I just accept the assistant brewer's admonition that you can't accurately measure SG with a refractometer? I apologize again for the lengthy post. If you've read this far you have my thanks and I owe you a beer. If you can explain the issue to me in simple terms I can comprehend, you have my eternal gratitude and respect. And I owe you a keg!

Brooo Brother
 

rburrelli

Well-Known Member
Joined
Jan 6, 2019
Messages
561
Reaction score
372
I think Mr. Assistant Brewer was merely testing you to see your response, which was suitable and hence OK.

I don’t know enough to respond to the overall issue.
 

pdxal

Supporting Member
HBT Supporter
Joined
Jul 19, 2010
Messages
1,852
Reaction score
277
Location
Portland, OR
Not sure how much but common sense would expect the wort refraction index to change as maltose is converted to alcohol as fermentation proceeds. As the percentage of maltose decreases and ethanol increases I would expect the WRI to change. Not sure how it is corrected for in the calculators, though.
 

doug293cz

BIABer, Beer Math Nerd, ePanel Designer, Pilot
Staff member
Mod
HBT Supporter
Joined
May 14, 2014
Messages
13,620
Reaction score
10,606
Location
Renton
If the correlation between your refract and hydrometer is good, then you can trust the refract measurements. You didn't mention whether you brew a wide variety of styles, but the wort correction factor can vary for very different styles. So, WCF should be determined for different styles.

If the original OG measurement's correlate well, and the (corrected) FG measurements also correlate well, then forget about the "OG" back calculated from the refract and hydro measurements of FG. That back calculation is meant as nothing more than a crutch for when you forgot to take a real OG measurement.

Brew on :mug:
 

Robert65

Major Obvious (recently promoted)
Joined
Jul 15, 2019
Messages
1,891
Reaction score
2,387
Location
Akron, Ohio
Here's another thought if your concern is making sure you spund at the correct time. If you use your refractometer to read your FFT, just make a note of the uncorrected WRI. Then watch for your main batch fermentation to approach this. Then the calculated correction is academic. When the WRI of the batch approaches within 1 Brix of the FFT WRI, you are in your spunding window. Sure, you probably want a reasonable estimate of actual terminal gravity for alcohol and attenuation calculations and fastidious record keeping. But you don't need an accurate, actual value for timing spunding, just a relative index. It has taken me way too long to figure out this (in retrospect) obvious idea.
 
OP
OP
Brooothru

Brooothru

Well-Known Member
Joined
Feb 16, 2012
Messages
2,336
Reaction score
2,259
Location
Either in the brewery or on the road
If the correlation between your refract and hydrometer is good, then you can trust the refract measurements. You didn't mention whether you brew a wide variety of styles, but the wort correction factor can vary for very different styles. So, WCF should be determined for different styles.

If the original OG measurement's correlate well, and the (corrected) FG measurements also correlate well, then forget about the "OG" back calculated from the refract and hydro measurements of FG. That back calculation is meant as nothing more than a crutch for when you forgot to take a real OG measurement.

Brew on :mug:

That makes a lot of sense. I'd conjectured as much but couldn't find anything to hang my hat on. And yes, I have a high degree of confidence in the accuracy and consistency of the OG and FG readings with both hydrometer and refractometer.

Mostly I brew pales, IPAs and Continental lagers with an experimental or wheat beer every now and then. I probably don't need to modify my WRI adjustment for the styles, I wouldn't think. And of course I don't have to correct when measuring wine or must.

Thanks for the technical insight and confidence boost. Now where do I deliver the keg?

Brooo Brother
 

S-Met

Department of Redundancy Department.
HBT Supporter
Joined
Jan 15, 2018
Messages
4,991
Reaction score
4,997
Location
Nor Cal
Some people will tell you that you are wrong regardless of whatever you are using. A hydrometer is inaccurate if temperature is not correct or corrections are not used. Said correction is not valid if your thermometer is off. Hydrometer reading is inaccurate if you don't "read" it correctly.

Two points:
1st, measurements are only as good as the one taking them and the accuracy and specificity of the equipment being used.

2nd, what is the intention of the tool being used? Sg, fg, and checking progress. While I have concerns regarding reproducibility from person to person (meaning refractometer to refractometer with variable correction factors) the numbers do show that the brix go down and a 3 day stable refractometer directly correlates with a 3 day stable hydrometer - fermentation is complete (or stalled.)

From a scientific process perspective, I see concerns regarding reproducibility from a refractometer. This could be problematic if you shared your recipe and needed to make an adjustment at a point specific gravity. However, if you are only measuring sg/fg for abv and confirming fg, refractometer is sufficient for home use.
 
OP
OP
Brooothru

Brooothru

Well-Known Member
Joined
Feb 16, 2012
Messages
2,336
Reaction score
2,259
Location
Either in the brewery or on the road
Some people will tell you that you are wrong regardless of whatever you are using. A hydrometer is inaccurate if temperature is not correct or corrections are not used. Said correction is not valid if your thermometer is off. Hydrometer reading is inaccurate if you don't "read" it correctly.

Two points:
1st, measurements are only as good as the one taking them and the accuracy and specificity of the equipment being used.

2nd, what is the intention of the tool being used? Sg, fg, and checking progress. While I have concerns regarding reproducibility from person to person (meaning refractometer to refractometer with variable correction factors) the numbers do show that the brix go down and a 3 day stable refractometer directly correlates with a 3 day stable hydrometer - fermentation is complete (or stalled.)

From a scientific process perspective, I see concerns regarding reproducibility from a refractometer. This could be problematic if you shared your recipe and needed to make an adjustment at a point specific gravity. However, if you are only measuring sg/fg for abv and confirming fg, refractometer is sufficient for home use.

Exactly. I use it only for the basic measurements and calculations you mentioned. It's a hobby. And it's homebrew. And I'm not doing exotic scientific analysis or replication of laboratory research experiments. But I do respect accuracy and strive for consistency.

I guess I'll stick to what I've been doing since it seems (with regards to gravity readings) I've been doing it right, or at least 'right' enough.

Brooo Brother
 

Latest posts

Top