Author Topic: ATC Refractometers  (Read 10087 times)

Offline Kaiser

  • Global Moderator
  • Senior Brewmaster
  • *****
  • Posts: 1797
  • Imperial Brewing Geek
    • View Profile
    • braukaiser.com
Re: ATC Refractometers
« Reply #30 on: May 27, 2010, 11:14:46 AM »

I have also read somewhere that refractometers are more accurate in natural light.  Had not seen that mentioned, or I missed it.

I wonder why that would be. Natural light covers a wide range of light colors. If the refraction changes with the color of the light, which it does to some extent, then any light source that has a more narrow color spectrum would lead to a more accurate reading. Given that calibration and measurement are done with the same light.

Could it be that incandescent light is biases toward the red end of the spectrum while natural light is actually “white” light?

One thing to consider is that in my refractometer, I don’t see any significant chromatic aberration which are the colored fringes around edges when using a cheap lens. If you see that it means that the color of the light caused a significant difference in the position of the transitional line that you use to read the refractometer.

Kai

Online a10t2

  • Official Poobah of No Life.
  • *
  • Posts: 3706
  • Ask me why I don't like Chico!
    • View Profile
    • SeanTerrill.com
Re: ATC Refractometers
« Reply #31 on: May 27, 2010, 12:10:48 PM »
Why is it necessary to sample the SG at full boil?  How do the results change what you do during a brew day?  And if it doesn't, then why do it?

I check the pre-boil gravity at a rolling boil because I've found that if I don't wait, the reading will be low due to the lower-gravity sparge runoff stratifying. Based on that reading, I'll adjust the length of the boil so that I hit my target OG exactly.

Are you mixing your 170°F wort to avoid stratification, or is that somehow unnecessary in your setup?
Beer is like porn. You can buy it, but it's more fun to make your own.
seanterrill.com/category/brewing
twomilebrewing.com

Offline richardt

  • Senior Brewmaster
  • ******
  • Posts: 1227
    • View Profile
Re: ATC Refractometers
« Reply #32 on: May 27, 2010, 05:43:59 PM »
Brewing coffee isn't that different from brewing wort/beer.  Both use refractometers.  Check out this site and draw your own conclusions.  http://vstapps.com/blog/wp-content/uploads/2010/03/Coffee-and-Refractometers.pdf

The measurement technique outlined in that article also recommends to cool the sample before placing it onto the prism. I'd love to copy that section but the Acrobat Reader plug-in I have doesn't let me. WTF.
Kai

The document seems to be password protected--maybe that's why you couldn't copy it.  I tried too--unsuccessfully.
The coffee link also mentions that one should thoroughly mix the coffee before taking a sample.

Offline richardt

  • Senior Brewmaster
  • ******
  • Posts: 1227
    • View Profile
Re: ATC Refractometers
« Reply #33 on: May 27, 2010, 05:46:01 PM »
Good discussion and the evaporation may be a thing that has caused variation in my readings.
I have also read somewhere that refractometers are more accurate in natural light.  Had not seen that mentioned, or I missed it.
You are correct.  Refractometers are calibrated with sodium light (yellow-orange, 589.3 nm), and a known [sucrose] solution at 20 degrees Celsius.  From a practical standpoint, natural light and incandescent bulbs are preferable to fluorescent bulbs which possess different color spectra and “warmth” of light depending on the vapor (usually mercury) within the tube and the age of the tube.

Offline richardt

  • Senior Brewmaster
  • ******
  • Posts: 1227
    • View Profile
Re: ATC Refractometers
« Reply #34 on: May 27, 2010, 05:51:46 PM »

I have also read somewhere that refractometers are more accurate in natural light.  Had not seen that mentioned, or I missed it.
I wonder why that would be. Natural light covers a wide range of light colors. If the refraction changes with the color of the light, which it does to some extent, then any light source that has a more narrow color spectrum would lead to a more accurate reading. Given that calibration and measurement are done with the same light.
It is true that natural light is a blend of colors.  Given that the red and green cones of our retinas play a more prominent role than the blue cones, it makes sense to “straddle the color fence” and calibrate with yellow-orange sodium light (which lies between the two colors) and minimize any deviation due to chromatic dispersion.  For those of you who wear glasses, you may have had your eye doctor perform a “duochrome test” with Red and Green colors in the background on the eye chart to make sure your glasses prescription wasn’t over- or under- correcting your vision.
Could it be that incandescent light is biases toward the red end of the spectrum while natural light is actually “white” light?
IMO, incandescent light is biased towards the yellow/white, while some, but not all, of the fluorescent bulbs tend to be blue-ish or dim.  But, this could be changing.  The newer compact fluorescent bulbs might not pose as much of a variance as the older tube type fluorescent bulbs can.
One thing to consider is that in my refractometer, I don’t see any significant chromatic aberration which are the colored fringes around edges when using a cheap lens. If you see that it means that the color of the light caused a significant difference in the position of the transitional line that you use to read the refractometer.
Correct.  You should not see chromatic aberration with your refractometer unless something is misaligned inside (so don’t drop it!)
Kai
See my anwers above in bold
« Last Edit: May 27, 2010, 07:39:12 PM by richardt »

Offline richardt

  • Senior Brewmaster
  • ******
  • Posts: 1227
    • View Profile
Re: ATC Refractometers
« Reply #35 on: May 27, 2010, 05:59:29 PM »
Why is it necessary to sample the SG at full boil?  How do the results change what you do during a brew day?  And if it doesn't, then why do it?
I check the pre-boil gravity at a rolling boil because I've found that if I don't wait, the reading will be low due to the lower-gravity sparge runoff stratifying. Based on that reading, I'll adjust the length of the boil so that I hit my target OG exactly.
Maybe I’m a little lazy, but I like to keep it simple during the brew day.  I’ve not had a problem with missing my OG target as calculated by BeerSmith (using the 10 gallon Gott cooler/MLT and 20 gallon SS kettle).  I’m more interested in following the recipe to the letter.  Even if it occurred, being off by one or two (or, hypothetically, even five) gravity points doesn’t really bother me.  It’s just homebrew—not the profit margin of a commercial brew pub.
Are you mixing your 170°F wort to avoid stratification, or is that somehow unnecessary in your setup?
My “170 F wort” is the final runnings from the MLT (to make sure I don’t get below 3 Brix / 1.012 SG).  It’s a “clean-catch, mid-stream” sample  ;) obtained with the dropperette!  I simply don’t mix and sample the kettle contents on the way up to boiling, when it is boiling, or when it is chilling.  I only sample the kettle contents after wort chilling is complete.  I put the sanitized thermometer in the kettle and compare it against the waterproof digital thermometer in the output end of the immersion chiller.  When it gets close to each other, it is time to take a sanitized dropperette and take a reading (and I’m fully aware that the wort is stratified or beginning to stratify and that I’m probably drawing off the lower SG sample from the upper ½ of the kettle.  But, that’s just as flawed as transferring my wort to a fermenter and using the pint that remains behind in the trub for my OG reading—as it would likely have a higher SG from being in the lower 1/2 of the kettle).  Anyway, that is the way I do it.

Offline nyakavt

  • Assistant Brewer
  • ***
  • Posts: 104
    • View Profile
Re: ATC Refractometers
« Reply #36 on: May 28, 2010, 07:32:13 AM »
Richardt, from your last post it sounds like you don't take any measurements at elevated temperatures that are subsequently verified by a cooled post boil sample, or a mash gravity reading that is verified against the laboratory yield of the mash.  So if you do not verify any of your measurements, why are you arguing that you don't need to cool the sample in a sealed container???

Online a10t2

  • Official Poobah of No Life.
  • *
  • Posts: 3706
  • Ask me why I don't like Chico!
    • View Profile
    • SeanTerrill.com
Re: ATC Refractometers
« Reply #37 on: May 28, 2010, 10:28:48 AM »
(and I’m fully aware that the wort is stratified or beginning to stratify and that I’m probably drawing off the lower SG sample from the upper ½ of the kettle.  But, that’s just as flawed as transferring my wort to a fermenter and using the pint that remains behind in the trub for my OG reading—as it would likely have a higher SG from being in the lower 1/2 of the kettle).

FWIW I don't think that would be true for either case.
Beer is like porn. You can buy it, but it's more fun to make your own.
seanterrill.com/category/brewing
twomilebrewing.com

Offline richardt

  • Senior Brewmaster
  • ******
  • Posts: 1227
    • View Profile
Re: ATC Refractometers
« Reply #38 on: May 28, 2010, 01:16:35 PM »
(and I’m fully aware that the wort is stratified or beginning to stratify and that I’m probably drawing off the lower SG sample from the upper ½ of the kettle.  But, that’s just as flawed as transferring my wort to a fermenter and using the pint that remains behind in the trub for my OG reading—as it would likely have a higher SG from being in the lower 1/2 of the kettle).

FWIW I don't think that would be true for either case.

I've done OG readings on '"slushy" trub sludge in the past when I didn't leave enough wort behind for a good hydrometer sample--the readings were ridiculously high--had to wait until the sludge settled hours later to get a more accurate reading with the hydrometer.  Even then, not so sure--it still seemed high for the targeted gravity.
Should be easy enough to test if someone felt like experimenting on their next batch (samples from upper 1/2 and bottom 1/2 after wort chill is completed and kettle wort residual+break material after transfer) measuring with both the hydrometer and the refractometer.

Online a10t2

  • Official Poobah of No Life.
  • *
  • Posts: 3706
  • Ask me why I don't like Chico!
    • View Profile
    • SeanTerrill.com
Re: ATC Refractometers
« Reply #39 on: May 28, 2010, 01:31:47 PM »
I'll do that when I brew next week. It's a high-gravity beer (~24 Brix) so it should work well to test for even slight variations.
Beer is like porn. You can buy it, but it's more fun to make your own.
seanterrill.com/category/brewing
twomilebrewing.com

Offline richardt

  • Senior Brewmaster
  • ******
  • Posts: 1227
    • View Profile
Re: ATC Refractometers
« Reply #40 on: May 28, 2010, 01:35:19 PM »
Richardt, from your last post it sounds like you don't take any measurements at elevated temperatures that are subsequently verified by a cooled post boil sample, or a mash gravity reading that is verified against the laboratory yield of the mash.  So if you do not verify any of your measurements, why are you arguing that you don't need to cool the sample in a sealed container???
This is a forum to discuss differences in home brewing techniques and methodology, and to learn more about the principles of brewing science.  This discussion is proof of that.

Nyakavt, my reason is simple:  IMO, there is no need to take a HIGH TEMP sample during the boil, so I don’t.  Your statement that I don’t “verify any of my measurements” is misleading.  I calibrate my instruments and use them appropriately—therefore I have high confidence in their results.  If I use the instruments outside of normal operating conditions, I temper/adjust my interpretation of the results accordingly.  IMO, the cover plate on my refractometer is my “sealed container” [see additional discussion below].  OTOH, given the observations of others who do this, if you’re taking a larger sample from boiling wort, you’ll need a larger container and should seal it while it chills to minimize any evaporation.
 
Whether you take the sample at the beginning of the boil or when you‘re done chilling the wort, you still need to know two things to do your efficiency calculations:  wort volume and specific gravity.  A refractometer is just a quicker way to get a SG reading.  See “Efficiency Calculations” on http://en.wikipedia.org/wiki/Plato_scale.

I’m suggesting that refractometer (or hydrometer) readings of already chilled wort might be a more sensible time to gather data for efficiency calculations, namely wort volume and specific gravity.  Markings on the brew stick or on the fermenter indicate the wort volume.   If done at that time, sample evaporation and temperature extremes are less of a concern with regards to measurement accuracy.  Still, there remains some inaccuracy which arises from using refractometers to measure mixtures other than pure sucrose (for which they were originally designed).  Nonetheless, the BYO article shows us how we can still use the refractometer with rather good accuracy if certain adjustments are made.

Why take a sample at the time you begin your boil (i.e., when the sample is at 212 F and must be chilled to get a more accurate reading)?  Why make more work for yourself?

My highest temp sample (<170F) is done at lautering to make sure I remain >3 Brix (> 1.012 SG), and not to calculate efficiency.  For purposes of estimating the SG of final runnings, I’m comfortable with a “fuzzier” result or ballpark number.  However, a sample at 212F is nearly 45F hotter.  A10t2 (Sean) says he does it at full boil because he wants to adjust his boil time, if necessary.
  
I did read on a Fresno, California Extension Service site that excessively warm ambient temps (e.g. >68-86F, 20-30C; California Valley) can warm up the refractometer enough to cause falsely elevated Brix readings (Source:  http://cefresno.ucdavis.edu/files/43066.pdf ).  According to the site, taking a reading at 50F (10C) can throw off your Brix reading by 0.89 (nearly 4 gravity points higher).  I suspect this is due to the excessive effect of the excessively high or low ambient temperature on the bimetal strip and the optical wedge it holds (the “ATC”, or automatic temperature compensator) within the refractometer and the effect of temperature on refractive index of the sample (RI is inversely related to temperature, i.e., RI decreases as temp increases).  Having said that, one or two drops of hot or warm liquid on the refractometer (and 30 seconds wait) probably does not have as much of a difference, as long as the refractometer temperature is within the ambient range (and preferably on the lower end of it).  I do know the sample is at ambient temp (cool) when I lick it off the refractometer surface after 30 seconds.

IMO, since the dropperette and the refractometer surface act as heat sinks (reducing the temperature closer to ambient temps), the temperature of my sample is well below the boiling point (reducing the amount of evaporation), and that I take two seconds or less to sample/smear/cover (very little opportunity for evaporation as I’ve created my own “sealed container” with the cover plate) with the refractometer, I’m less likely to experience temperature error.  That may not be the case if 212F wort were sampled or the ambient temps during the brew day were really cold or really hot.  Stratification may be a concern if I’m sampling from the boil kettle after chilling and may play a bigger role with darker and more dextrinous or proteinaceous beers, e.g. higher mash temps or wheat beers.  Stratification errors can be minimized by mixing prior to sampling (i.e., during a rolling boil, or immediately after transfer to the fermenter).  I may be making a mountain out of a molehill, but I think it can and does play a role—whether it is negligible or not, I just don’t know.  Some sites I've looked at seem to suggest that it does matter.

I’m a homebrewer, not a commercial brewer--I don’t care if I miss a calculated target by a small amount.  I also don’t care about my “efficiency”—which runs around 75% and sometimes higher.  I do care about following the recipe accurately with regards to the amount of water used, the temps of the rest(s), the times of the various hop additions, etc. so that I can reproduce it again.  I also have no doubts about the veracity of my measurements within the limitations of my instruments and the manner in which I use them.

Offline Kaiser

  • Global Moderator
  • Senior Brewmaster
  • *****
  • Posts: 1797
  • Imperial Brewing Geek
    • View Profile
    • braukaiser.com
Re: ATC Refractometers
« Reply #41 on: May 28, 2010, 03:52:39 PM »
Richardt, you make some valid points but none of them disprove the idea that significant evaporation can occur between placing the sample on the refractometer and closing the lid of the wort is sufficiently hot. The question of why to take gravity readings at boil time has nothing to do with that.

Furthermore, there is NO stratification of wort after it has been mixed. I never saw anything like this reported and believe that we would have noticed that by now. If that was to happen it would also happen during the storage of beer and we all know that the first drop from a keg has the same gravity as the last.

Kai

Offline nyakavt

  • Assistant Brewer
  • ***
  • Posts: 104
    • View Profile
Re: ATC Refractometers
« Reply #42 on: May 29, 2010, 10:46:32 AM »
This is a forum to discuss differences in home brewing techniques and methodology, and to learn more about the principles of brewing science.  This discussion is proof of that.

Please don't take my post as an attempt to keep you from voicing your opinion, that's how we get the best discussions.

Quote
Nyakavt, my reason is simple:  IMO, there is no need to take a HIGH TEMP sample during the boil, so I don’t.  Your statement that I don’t “verify any of my measurements” is misleading.  I calibrate my instruments and use them appropriately—therefore I have high confidence in their results.  If I use the instruments outside of normal operating conditions, I temper/adjust my interpretation of the results accordingly.  IMO, the cover plate on my refractometer is my “sealed container” [see additional discussion below].  OTOH, given the observations of others who do this, if you’re taking a larger sample from boiling wort, you’ll need a larger container and should seal it while it chills to minimize any evaporation.

The purpose of a sample in boiling wort is to ensure that the sugar is evenly mixed throughout, and to be able to adjust the boil time to reach your target gravity.  To me making the beer that I set out to make is important so that I do not need to adjust the bitterness if I am 5 or 10 points off.  The other catch on my system is that I do not have a good way to measure volume post boil, since I use an immersion chiller which distorts the reading on the depth gauge after it goes in and I don't want to worry about sanitizing it.  Volume in the fermenter is no good either since I leave a 1/2 gallon or so behind in the kettle, so even though I would be calculating my 'true' efficiency, I'm more interested in the yield I got from the grains since the number is not dependent on the trub volume in the kettle because of varying amounts of grain or hops.

It's not that I'm trying to hit a specific number for my efficiency, but I want to hit the target gravity for the recipe and this is the best way I know to do it.  The 'verifying your measurements' comment is referring to the fact that you claim quickly closing the lid on the refractometer is 'good enough' for a sealed container, but you have no direct evidence to support this claim.  Every time I've tried this (that I can remember) the number was higher than expected.  In order to know that the 'warm' measurement you are taking is accurate, you need to compare it to some known accurate value, which can be done by comparing a pre-boil and post-boil (and cooled) gravity measurement with accurate volumes for each.

I certainly agree with you that it seems unlikely that a few seconds of exposure to air is enough to cause the change in gravity I am seeing, but nevertheless I cannot just ignore the discrepancy because it seems unlikely.  The surface area to volume ratio of the sample would be much higher than in a normal case of evaporation, whose surface area is only on top of the column of liquid.  It could be a combination of things going on, I really don't know.

Quote
My highest temp sample (<170F) is done at lautering to make sure I remain >3 Brix (> 1.012 SG), and not to calculate efficiency.  For purposes of estimating the SG of final runnings, I’m comfortable with a “fuzzier” result or ballpark number.

I can agree with this, since the number is inconsequential to reaching your target gravity, and any inaccuracy would put you on the safer side of avoiding the 3P mark.

Anyway, I'll be brewing today and will keep in mind some of the points you raised richardt.  Unfortunately it's a low gravity beer so the effects of stratification, if they exist, will be harder to detect.  Either way this does not solve the issue of inaccuracies at warm temperatures, and in my process I do have a need for them to hit the target gravity.
« Last Edit: May 29, 2010, 10:51:19 AM by nyakavt »

Offline nyakavt

  • Assistant Brewer
  • ***
  • Posts: 104
    • View Profile
Re: ATC Refractometers
« Reply #43 on: May 29, 2010, 05:57:38 PM »
Update on this, I'm brewing a scottish 70/- with kettle caramelization, so I was in the unique position to leave the wort sitting for a long time before firing the main burner.  The kettle sat still for about 2 hours with 25 liters of runnings, it was not stirred or disturbed other than running wort into the kettle.  After the wait (wort temp was ~125 F), I took a 5 mL sample in a 10 mL syringe, then stirred the kettle well for a minute (you could see the settled protein change the color of the wort) and take another sample from the same spot.  I then put both syringes in the freezer until it had chilled to approx. room temp, about 5 minutes.  I wiped down the outside o the syringes with a dry paper towel and shook them well.   Then I discarded the first 2 mL of sample to avoid any potential contamination by condensate on the outside of the syringe.  The unstirred sample read 6.7 Brix, and the stirred measured 6.8 brix.  This difference is within the resolution of the instrument (0.2 brix).  I did know which one I was sampling, so there is a potential for 'researcher bias' in interpreting the readings.   

This is only one observation and the wort is very low gravity, so it may not show stratification as well as another wort.  But the difference here is not enough to account for the difference I'm seeing when taking a warm sample.

Offline nyakavt

  • Assistant Brewer
  • ***
  • Posts: 104
    • View Profile
Re: ATC Refractometers
« Reply #44 on: May 29, 2010, 10:20:19 PM »
Yeesh, long day.  Ok here are the other measurements for this brew session.

First I took a pre-boil gravity by drawing a sample from the boiling wort with a syringe.  This time I put a drop on the lens and closed it almost immediately, probably 1/2 a second lapse (this is by far faster than I've ever done it before).  This reading was 9.3 brix.  I then placed that syringe in the freezer until it was chilled and it read 9.2 brix, again inside the instrument resolution.  So it appears that quickly closing the lens will get you very close to a chilled sample.  I also took and chilled a sample before the wort got to a boil which came in at 9.1 brix, but the wort boiled for a short amount of time before I could get to it so this is not directly comparable.  Still, very close.

The final test of the day was the top-bottom measurement.  After chilling and whirlpool the kettle was allowed to rest for 20 minutes.  Then I took a sample from the top of the wort and ran 1/2 liter into the fermenter and took a sample from that, effectively the bottom of the wort.  Here the top sample read 9.9 brix and the bottom read 9.8.  Oddly the bottom was the lower reading, but again this is inside the instrument resolution so it's probably just drift. 

So this demonstrates to me that on a wort of this gravity, stratification is not a concern, and quickly covering the lens gets you close enough to a sealed container reading.  I am going to continue testing on my next few batches to see if the trend holds.  In the long run, I will still prefer to chill the sample prior to reading, at least of boiling wort, to preserve the instrument as long as possible.