refractometer question

[MalFet]

OK, I get it. No I don't need to take 99.99% accurate readings, I'm just trying to understand if and where the measurement is off. Sounds like an adjustment is in order.

These things should come with disclaimers.

Indeed. That's why I wish someone would make a refractometer that only displayed refractive index. That way, people would at least realize that there's a conversion involved.

All that said, though, I love my refractometer. It's not one of Bobby_M's (I got it before he started selling them), but it looks to be nearly identical. I use a refractometer throughout my mash and boil, and then compare it against a hydrometer reading taken as I go into the fermentor. If you keep good notes, you'll have a pretty good sense after just a few batches of how much correction is necessary.

 

[zeus53219]

I have a Extech RF11 refractometer only in Brix scale. In the instructions it does state to use a Brix Correction Factor of 1.04 with Wort/Maltose, 1.01 with Sucrose and 1.00 with Sugar in Fruit Juice/Fructose. So if I have a brix reading of 23 measuring wort, that would come out to around 1.093. If I was measuring must, it would be 1.097. Guess refractometers are a little like women, they are all a little different, but once you know how to read them right, you achieve good results.

 

[ajf]

I've tracked my refractometer over a few dozen batches now and on average I've needed to bump my refractometer reading up by about 5% to match my hydrometer, though on any given batch that number has ranged from 3% to 7%. That's plenty good enough for me to monitor the mash, but certainly if you want a precise measurement of gravity you need to use a device designed for measuring density, not refraction.

What are you putting in your beers? or how are you converting from Brix to SG ? I need to reduce the refractometer reading by about 4% to forecast a reasonably accurate SG reading

-a.

 

[MalFet]

What are you putting in your beers? or how are you converting from Brix to SG ? I need to reduce the refractometer reading by about 4% to forecast a reasonably accurate SG reading

-a.

Apologies, ajf. I missed this message somehow. It was only when looking for this link for someone else that I noticed your question.

Nope, I definitely multiply by 1.05. In other words:
(Brix refractometer reading) * 1.05 = (reasonable prediction of hydrometer reading)

It could be that we just have differently calibrated screens in there from the get-go. If that's the case, it looks like Kaiser and I got our refractometers from the same batch:
http://braukaiser.com/blog/blog/2012/03/23/dont-trust-your-refractometer-blindly/

This is the lesson here, it seems: all measurement tools are different, so it pays to calibrate!

 

[ajf]

Well, the good news is that using a Brix Calibration Factor, the readings from the refractometer (in whatever units they are actually displayed) seem to be able to give a fairly accurate representation of S.G.
I wonder however how much the Calibration Factor varies according to the grain bill, and how much the boil with or without hops affects it? I know, that whenever I have brewed a lager using Pilsner malt and used the refractometer to estimate the S.G, I seem to get higher efficiency than if I brew a pale ale using M.O. I can see I need to do some more experiments, and possibly develop some brewing software where the correction factor is somehow linked to the ingredients.

-a.

 

[MalFet]

It's an interesting question. I just brewed two batches back to back, and the grain bills couldn't have been more different. First was 1.084 with all base malts and roasts. The other was 1.040 with MO and lots of crystal. Both required a 1.05 conversion, but a few weeks ago I brewed two other batches and I needed 1.03 and 1.07. It's very weird. I'm not sure how much of the variation is ingredient based versus just measurement error.

 

[ajf]

I agree that it is interesting. I was going to run a series of experiments, making a large number of small (1 liter) batches. I think I shall add refractometer readings for each batch, taking refractometer and gravity readings both pre and post boil; and see if I can see any patterns. The advantage of the 1 liter batch size is that I can weigh everything accurately to the nearest gram, which should minimize measurement errors.

On your two batches where you needed unusual correction factors, could you provide me with the grain bills, so I can see if I get similar results?

Thanks for your useful comments.

-a.

 

[Yooper]

I agree that it is interesting. I was going to run a series of experiments, making a large number of small (1 liter) batches. I think I shall add refractometer readings for each batch, taking refractometer and gravity readings both pre and post boil; and see if I can see any patterns. The advantage of the 1 liter batch size is that I can weigh everything accurately to the nearest gram, which should minimize measurement errors.

On your two batches where you needed unusual correction factors, could you provide me with the grain bills, so I can see if I get similar results?

Thanks for your useful comments.

-a.

GREAT idea! I'll be following your experiments. My correction factor is 1.052 for my hydrometer and refractometer to match with wort.

 

[MalFet]

On your two batches where you needed unusual correction factors, could you provide me with the grain bills, so I can see if I get similar results?

That's the thing, they weren't really unusual. I see a pretty even distribution in the range 1.03 to 1.07, depending on the batch. If I run with 1.05 I end up with something that is always close enough for my purposes, but I wouldn't say that 1.05's are more common than 1.03's or 1.07's.

The most recent 1.03 was 90% 2row and 10% special b, 1.065 OG. The 1.07 was a 82% Belgian Pilsner, 6% acidulated, 6% aromatic, and 6% malted oats for an OG of 1.054.

It's possible that I'm just having measurement errors, but I use a lab grade hydrometer, adjusted for temperature, and my ATC-refractometer seems otherwise reliable. I brew indoors, so my ambient temperature is pretty much 70º all the time.

Keep us posted if you find anything!

 

[ajf]

Thanks MalFet.

I have an adequate supply of US 2 row, Maris Otter, and German Pilsner, and a good supply of specialty malts. I have also bought a 1 L vacuum flask for doing small mashes in. I have an accurate finishing hydrometer, but so far, nothing accurate that can measure a gravity > 1.020.

I ordered a couple hydrometers that should plug that gap. While waiting for them to arrive, I will try a few small batches to determine the thermal mass of the vacuum flask, and boil off rates etc. I shall also write a simple program which accepts inputs of measured specific gravity and Brix readings from the refractometer, and calculates a Brix Correction Factor based on the formula Specific gravity = (Brix/(258.6-(Brix/258.2)*227.1))+1.

The Brix reading will be divided by the correction factor before applying the formula, so with my refractometer, the correction factor will be > 1, and with yours, the correction factor will be < 1.

Then I will be able to start some tests.

Each test will consist of brewing ~ 1 L batch, measuring the pre boil volume, gravity, and Brix, then boiling for 45 minutes +, and measuring the post boil volume, gravity and Brix. I will then perform a fast fermentation test using bread yeast on the stir plate to determine the limit of attenuation.

I shall start using a mash thickness of 1.25 qt / lb, and a mash temp of 154 F and using just one of my base grains. Then I can repeat the test using a different mash temperature to see what effect the mash temperature has on attenuation. Then, using a single temperature, I will try mash thickesses of 1, 1.5, 1.75, and 2 qts / lb to see if I can detect any differences caused by the mash thickness, and using a low temperature, I will try some longer mashes to see if I can detect any difference caused by the mash time.

Having done this, I shall try a sub-set of those tests using various specialty malts, and then some typical mixtures of base and specialty malts to see if there are any noticeable anomalies.

I just hope my readings are fairly consistent.

If anybody has any other suggestions or comments, I'll be happy to consider them.

Please don't hold your breath until the tests are finished. Even if nothing goes wrong, it will take a long time to complete the tests, but I will post the results as I get them.

-a.

 

[mattd2]

I didn't get really accurate measurements but... I did an IPA on the weekend, ended up at:
1.071 Hydro
1.070 Refract
18° Brix Refract - converts to 1.074 with an only calculator.

So far I am leaning towards trusting my Refract's SG scale... so far

 

[FATC1TY]

I didn't get really accurate measurements but... I did an IPA on the weekend, ended up at:
1.071 Hydro
1.070 Refract
18° Brix Refract - converts to 1.074 with an only calculator.

So far I am leaning towards trusting my Refract's SG scale... so far

Why is that?

Have you calibrated your hydrometer in proper temp distilled water?

 

[mattd2]

Why is that?

Have you calibrated your hydrometer in proper temp distilled water?

The hydro measurement is probably just due to my "quick" effort at reading it. There was a fair amount of foam to read through and I was 95% sure it was on 71, but 100% sure it was not 74.
I will continue to take an initial og with both a hydro and refract (and try to be more scientific about it) until I am sure which way to lean.

 

[diS]

By reading all of this calibration factors it seem that only my refractometer gives lower reading.. as I can remember my calibration factor is around 0.96.
I tested it with 2 samples in different batches and get the same result.. both hydrometer and refractometer are calibrated in distilled water.

 

[MalFet]

By reading all of this calibration factors it seem that only my refractometer gives lower reading.. as I can remember my calibration factor is around 0.96.
I tested it with 2 samples in different batches and get the same result.. both hydrometer and refractometer are calibrated in distilled water.

Maybe we're just doing the algebra slightly differently, or maybe I'm just misunderstanding you. My refractometer produces lower readings than my hydrometer does, and I need to multiply my refractometer's reading by around 1.05 to match my hydrometer. Is that what you're seeing too?

 

[brewswellwithothers]

If I might suggest another variable to the discusison of grain bills, the Mash temp. We already know that different mash temps give more/less dextrins and effect the body and attenuation of the beer. Perhaps this is where the Brix conversion is getting the swings from 1.03 to 1.07.

 

[ajf]

Maybe we're just doing the algebra slightly differently, or maybe I'm just misunderstanding you. My refractometer produces lower readings than my hydrometer does, and I need to multiply my refractometer's reading by around 1.05 to match my hydrometer. Is that what you're seeing too?

Brewing Software such as Beersmith and Promash divides the Brix reading by the Brix Correction Factor before converting to S.G. Because of this, I would like to keep the division, but dividing by 1/1.05 is exactly the same as multiplying by 1.05.

If I might suggest another variable to the discusison of grain bills, the Mash temp. We already know that different mash temps give more/less dextrins and effect the body and attenuation of the beer. Perhaps this is where the Brix conversion is getting the swings from 1.03 to 1.07.

Already covered in post 50.

-a.

 

[brewswellwithothers]

@ ajf

I read all the posts but must have missed that part, glad to see that was included and thank you for running those tests. I am interested in how these tests turn out, as I am still getting used to my equipment and all grain. I've done 3 batches so far, 2 with a Refractometer way out of calibration; and last one with a replacement Refractometer properly calibrated. This last batch I was able to hit my numbers spot on with expectations; and my final pre-fermenter readings with Refractometer and hydrometer matched pretty well.

 

[diS]

Maybe we're just doing the algebra slightly differently, or maybe I'm just misunderstanding you. My refractometer produces lower readings than my hydrometer does, and I need to multiply my refractometer's reading by around 1.05 to match my hydrometer. Is that what you're seeing too?

Seems its just inverse calculation, I am dividing refractometer reading with calibration factor where Calibr. factor = Brix refractometer reading : Brix hydrometer reading.

 

[MalFet]

Seems its just inverse calculation, I am dividing refractometer reading with calibration factor where Calibr. factor = Brix refractometer reading : Brix hydrometer reading.

Right, that's fine. In that case, we're both seeing error in the same direction. Kaiser's refractometer did the same thing.

 

[luke2080]

Glad I stumbled on this thread, and that its a fresh thread. Totally forgot that many of the refracs on the market are calibrated for wine and not beer. Just used mine (same model that this thread was starated with). Pre-boil gravity was 1057, right where I should have been. Added 2.5 lbs of DME as planned (to 12.25 G of pre-boil wort) and my refrac had a reading of 1073. Beersmith says I should have been 1079.

I'm guessing the SG side is slightly more accurate, by accident, in the 1050 range. But a little further off further from that point. Or something along those lines. Did not think to grab the hydrometer to take a reading to calibrate this.

I agree that it is interesting. I was going to run a series of experiments, making a large number of small (1 liter) batches. I think I shall add refractometer readings for each batch, taking refractometer and gravity readings both pre and post boil; and see if I can see any patterns.

I'll be looking forward to reading these results as well!

 

[Jsbeckton]

So it seems that many people had refractometers where 23 Brix = 1.088 and many had 23 Brix = 1.096. The earlier posts seemed to indicate that 1.096 was closer (and matched conversion tables) but the experiment seemed to show that 1.088 was closer? Mine reads 1.088 and is always low compared to my hydrometer so I thought that 1.096 would be closer. Which is the best starting point in terrms of requiring the lowest correction?

 

[MalFet]

So it seems that many people had refractometers where 23 Brix = 1.088 and many had 23 Brix = 1.096. The earlier posts seemed to indicate that 1.096 was closer (and matched conversion tables) but the experiment seemed to show that 1.088 was closer? Mine reads 1.088 and is always low compared to my hydrometer so I thought that 1.096 would be closer. Which is the best starting point in terrms of requiring the lowest correction?

The trouble with that question is that everyone is just going to give you the number that works for them. Some of the variation is driven by wort parameters, and some by the (mis)calibration of the instrument at manufacture. At least until ajf gets done, there's no principled way to guess ahead of time.

 

[ajf]

Well, I ordered the new hydrometers, and received one that measures from 1.050 to 1.100. (The other one is out of stock).

I wrote a couple of simple programs; one to calculate the grain weights, water weights, and strike temperatures to produce a mash at a certain thickness and temperature using my mini mash system. This program needs a certain amount of tweaking to get the strike temperature correct.
The other program accepts inputs of wort weight, wort specific gravity, and Brix (from refractometer) both pre and post boil. It then calculates a Brix Correction Factor (BCF) based on the gravity and Brix readings such that dividing the measured Brix by the BCF produces a corrected Brix reading that can be converted to S.G. by the formula
Specific gravity = (Brix/(258.6-(Brix/258.2)*227.1))+1
This is the definition of the Brix Correction Factor used by both Beersmith and Promash.
The program also reports any gravity points gained or lost during the boil.

I mashed 225g Maris Otter with 587g water (= 1.25 qt/lb) at 152F for 60 minutes, sparged in a French Press and recorded the weight, specific gravity, and Brix reading prior to the boil.

I then boiled the wort for 60 minutes, and again recorded the weight.
The following screen shot shows the results.

I don't think that losing 2 points during the boil is anything to worry about as the two gravity readings were taken with two different hydrometers, one of which has not been accurately calibrated.

However, when it comes to the calculated BCF's there is a big difference.

If I use the pre-boil BCF to estimate the post-boil S.G. from the post-boil Brix, then it gives an error of > 5 gravity points.

Why there is such a difference in the BCF's I don't know.

It could be that my refractometer is faulty, but I don't think so because it was reasonably accurate in post 25
It could be that the refractive index of the wort changes during the boil and needs to be accounted for when estimating a gravity reading from Brix.
It could be that the very small batch size exaggerates the difference. (Not many people boil down from a S.G. of 1.015 to end up at 1.060)
It could be caused by any number of other factors that I haven't even considered.

I intend to brew a full size batch this weekend, and will take pre and post boil S.G. and Brix readings to see if I get similar results on a full scale batch.

-a.

View attachment Untitled.bmp

 

[ajf]

There appears to be a problem attaching screen shots. My computer shows the results of the calculator, but my wife's computer shows X
If I try to edit the post, it show nothing.
This is what the screen shot showed:
Pre-boil weight (grams): 3418 Pre-boil S.G: 1.0155 Pre-boil Brix: 4.250 Pre-boil BCF: 1.0760
Post-boil weight (grams) 914 Post-boil S.G: 1.0600 Post-boil Brix: 14.500 Post-boil BCF 0.9836
Points added in boil -2

-a.

 

[FATC1TY]

My last brew day, my refract was off on the SG side..

It works great on Brix.. Reads it accurate when I take the Brix and compare it to my calculator that gives me the SG. It was spot on everytime.

So I'm just going to use the calculator in my BS2 program and go from there. Pre and Post gravities in Brix, convert, check final post gravity and it's all good to go.

 

[ajf]

So it seems that many people had refractometers where 23 Brix = 1.088 and many had 23 Brix = 1.096. The earlier posts seemed to indicate that 1.096 was closer (and matched conversion tables) but the experiment seemed to show that 1.088 was closer? Mine reads 1.088 and is always low compared to my hydrometer so I thought that 1.096 would be closer. Which is the best starting point in terrms of requiring the lowest correction?

I have just realized that if I use my pre-boil BCF of 1.076 (using a calibrated hydrometer), 23 Brix translates to 1.089 which is very close to 1.088.
If I use the post-boil BCF of 0.9836 (using an uncalibrated hydrometer), 23 Brix translates to 1.098 which is very close to 1.096

This leads me to believe that the boil significantly changes the refractive index of the wort, and the conversion from Brix to S.G. depends on whether you use a pre-boil or post-boil BCF.

However, there is a lot of experimentation to do in order to differentiate my gut feelings from actual facts.
-a.

 

[dbrewski]

Thank you for working on this. It seems like this might be pretty meaningful if true.

 

[diS]

This leads me to believe that the boil significantly changes the refractive index of the wort, and the conversion from Brix to S.G. depends on whether you use a pre-boil or post-boil BCF.
-a.

Interesting.. I'll start to keep notice on my BCF pre- and post- boil.
If at least part of folks who read this thread would take readings, in month or two we would maybe be able to get more representative results, and conclude whether boil changes BCF significantly.
Anyhow, it is worthwhile to explore it more detailed...

 

[diS]

I know this is not the topic of this thread but it may be helpful to someone, I noticed huge difference between my hydrometer and refractometer post-fermentation readings:

Post-boil gravity: hydrometer reading- 1.066; refractometer- 15.8 Brix, this matches up 0,96 BCF

Post-fermentation gravity: hydrometer reading- 1.018; refractometer- 8.2 Brix

According to calculator (http://onebeer.net/refractometer.shtml) corrected FG for Brix reading is 1.012, which is 6 points lower than actual.
It is pretty huge so I think I'll start using hydrometer again for my FG readings.
Suppose that refractometer can be used just to check is there change in gravity after hydrometer and refractometer reading are taken.

 

[ajf]

I've never had much luck with post fermentation Brix/S.G. calculators, but I haven't tried Sean Terrill's version See http://seanterrill.com/2012/01/06/refractometer-calculator/

I put your figures into that calculator, and it reported a F.G. of 1.0154 (only 2.6 points off, rather than 6).

-a.

 

[diS]

It is closer.
I see that you put 0.96 as correction factor, how do you know it multiplies correction factor and not dividing it (default value is 1.04 which correlates with 0.96 if reading is divided with CF)?

 

[ajf]

See http://braukaiser.com/blog/blog/2012/03/23/dont-trust-your-refractometer-blindly/

If I enter a Brix value in Promash or Beersmith, adjust the BCF upwards, then the estimated S.G. for that Brix value increases if I reduce the BCF, and decreases if I increase the BCF.
In fact, Promash doesn't even accept a BCF < 1.0 as this is theoretically impossible if the refractometer reports Brix accurately. Beersmith (1.4) does allow a BCF < 1.0.


-a.

 

[dbrewski]

Well, I ordered the new hydrometers, and received one that measures from 1.050 to 1.100. (The other one is out of stock).

ajf, where did you order your hydrometers from? I was looking at Fisher Scientific...there are hundreds of them. Hard to know which one is which.

 

[ajf]

I followed Malfet's suggestion, and ordered them from http://www.cynmar.com/

-a.

 

[BOBTHEukBREWER]

The first time I use my refractometer (measures BRIX only) is after hop boiling and cooling to 20 deg C. I then take a reading each subsequent day and plot them on a graph. The readings eventually confirm what I can see with my eyes anyway, that it is time to bottle, typically on day 5 or 6.

 

[FATC1TY]

The first time I use my refractometer (measures BRIX only) is after hop boiling and cooling to 20 deg C. I then take a reading each subsequent day and plot them on a graph. The readings eventually confirm what I can see with my eyes anyway, that it is time to bottle, typically on day 5 or 6.

Wow.. you bottle that early!?

 

[BOBTHEukBREWER]

Yes - no probs at all, but I have an insurance policy, 2 bottles are one litre PET bottles, so I can feel the pressure. Never had to open and degas and reseal for 30 years or more, brew once a week.

 

[ajf]

OK, I have completed a 5g brew, and taken Brix and S.G. readings both before and after the boil. The results are shown below.

In my previous experiment, it appears that the differences between the pre-boil and post-boil BCF's was caused by the very small sample size (or I screwed up taking a reading).

With the 5g batch, there is very little difference in the BCF's, and even if I switch them (or use my calibrated default settings) I get predicted S.G. readings within 1 point of what the hydrometer says.

-a.

View attachment Untitled.bmp

 

[a10t2]

Why there is such a difference in the BCF's I don't know.

I found this thread by pingback to the refractometer calculator, so I'm sorry to be so late to the party.

I suspect that any variation you're seeing is due to measurement technique, rather than a physical property of the wort. I would look at stratification (do you pull the wort sample from the kettle before it comes to a boil?) or evaporation (is the sample chilled in a sealed container?).

I hadn't checked my refractometer against a hydrometer in several batches, so I did for the brew day today.

Pre-boil: Refractometer 7.6°Bx, hydrometer 1.0300 (WCF 1.035)
Post-boil: Refractometer 10.3°Bx, hydrometer 1.0400 (WCF 1.052)

Granted, it's only a single set of measurements, but if there is any variation, it appears that it's so small as to be within the error bar for the other measurements (±0.05°Bx for the refractometer, ±0.00025 SG for the hydrometer). Both were calibrated at the beginning of the brewday.