I increasingly been having serious doubts about the accuracy of the tools I use to try to measure gravity, so this afternoon I decided to set up a test to see what was most accurate. Ray Daniels’ “Designing Great Beers” says that 1 lb. of table sugar dissolved in water to make 1 gallon yields a Specific gravity of 1.046. I scaled that down to 1 oz. of sugar and water to make 1 cup, which equals 1 lb. per gallon, and proceeded to test the Vee-Gee refractometer I recently purchased to replace the POS I originally bought, and an assortment of hydrometers I have collected.
The 1st test at a concentration that should be 1.046, the VeeGee gave a brix of 10.8 = 1.043 s.g., hydro #1 said 1.047, hydro #2 said 1.049, #3 said 1.050, #4 which I just bought as part of a 3 hydro set that is supposed to be super accurate as they each cover different gravity ranges said 1.049 [it covers a range from 1.000 to 1.070].
Next I heated the water to about 100*, added another oz. of sugar which should put the s.g. at 1.092, stirred thoroughly and cooled to room temp. The results corrected for temp were VeeGee - Brix of 20.5 = 1.085, Hydro #1 - 1.089, #2 - 1.089, #3 - 1.090, & #4 which covers a range from 1.060 to 1.130 said 1.088.
I then repeated the process with another ounce of sugar which should put the actual s.g. at 1.138 and tested all again. Results were VeeGee - Brix 28.4 = 1.122, #1 - 1.024, #2 - 1.124, #3 - 1.125, and the same #4 as the previous test - 1.124.
Needless to say I was anything but impressed. Clearly the refractometer was the least accurate instrument tested. Hydrometers #1 & #2 are the most accurate at most “normal” brewing gravities, and #3 is the most accurate above about 1.060 or so. The narrow range hydro’s were generally the least accurate, even if they are the easiest to read. The refractor is now destined for the trash can, and the set of hydro’s I just bought are going back to Midwest, unless they won’t refund my purchase, in which case they’re traveling with the VeeGee.
Maybe I am expecting too much from my tools, but this BEER for Pete’s sake, not friggin horse-shoes, close isn’t good enough for me.
Has anyone else found a more foolproof [or perhaps tool-proof] way to measure gravity? I look forward to everyone’s input, including critique of my testing protocol [gotta luv them fancy words!].
Thanks.
Did you calibrate all of them? I check my refractometer before each batch and it’s always dead nuts.
I discovered after posting that the set screw on what I’m guessing is the refractometer’s adjustment band was loose, so I’ll have to figure out how to re-calibrate and try it again. As for the hydrometers, I’m not aware of any way to calibrate them.
Put each hydrometer in 60F water and see how far it deviates from 1.000. I usually find them reading high from the paper inside sliding down. Then I wrap tape around the top until it reads 1.000.
Distilled or RO water for increased accuracy.
Were you measuring 1 oz or were you measuring 28.35 grams? Hopefully your scale enabled you measure in grams and not oz. That would be a potential source of error. However, it doesn’t diminish your work in measuring the differences between your hydrometers!
I weighed out an ounce on a gram scale I just purchased, it is very sensitive and measures to tenths of a gram or thousandths of an ounce, but that doesn’t of necessity mean it’s accurate. Why would measuring in grams be better than measuring in ounces?
The tape idea might work - sort of, except that a couple of the hydro are right on the gnat’s ass in distilled water, start adding sugar though and the wheels apparently start coming off.
So I recalibrated the refractometer to zero [with distilled water Stevie - thanks for the reminder] and re-tested it with the 1.138 solution and it came up with a Brix of 29.5 which equals 1.127 S.G. That’s not even in the ballpark, I’m wondering what is the point of continuing to measure gravity, especially on big beers when my measuring instruments are this worthless [ but I know I’ll continue to do so]. I’m also thinking that a hell of a lot of other home brewers are getting measurements that are just as inaccurate, but don’t know it, after all, they probably bought a lot of the same junk I bought from the same suppliers, because we all foolishly thought the stuff would work as advertised. I would be very interested if a few other folks would repeat my testing with their hydro’s and refrac’s and report their results. Maybe I’m just extremely unlucky, and always wind up with the 1 piece of crap out of a thousand off the assembly line, but the statistical likeliness of that being the case is miniscule at best.
rob_f, your suggestion about taping the hydro’s might work if they were reading high as it would lower them in the test jar, adding weight at this point would only exaggerate the inaccuracy, thanks for the idea though.
If a scale is only reporting in ounces and not in tenths or hundreths of an ounce, it would not be suited for measuring small quantities. If you were using a scale that only reported to the nearest gram, it would still be 28 times more sensitive than a scale reporting in whole ounces.
The fact that your scale reports in those fine divisions, your measurements should be good.
Also I’d toss in here that using a cup for measuring the water probably isn’t all that accurate either. Your best bet is to weigh the water on the same scale your using for the sugar. 1 ml water = 1 gram.
I have ranted about inaccurate hydrometers in the past on this very forum. It hasn’t done me any good…
Now I don’t care (I’m telling myself). Since it is for homebrewing purposes it is more about ballparking it and noting changes. In other words the primary information you need is relative.
I like that idea.
Another way to assess the density of your solution is to weigh a mass of known volume. If you have lab-quality volumetric cylinders or a calibrated pycnometer (big bucks!), you can use your scale to assess what the solution’s density is.
PS: you can check the accuracy of a volumetric cylinder’s marks by measuring the mass of distilled water filled to that level.
I’m far too anal of a tool junkie to ever be content with crappy tools Steve, probably not quite anal enough to invest in the kind of lab required to perform the tests Martin just suggested. Let me see, didn’t I pick up a pycnometer at a garage sale a few years ago, where on earth did I stash that darned thing? 
One thing this does mean, my brewhouse efficiency is a good bit better than I thought, and my heavier beers have a higher ABV than I thought [and I though I was just turning into a weenie].
Steve, I’ve got to agree with you; for me, great accuracy on gravities doesn’t bring much to my party. “Somewhere near 1.060” is close enough to tell me I likely didn’t screw up too badly. I do recognize that “I just want to know” is a perfectly valid reason for pursuing greater accuracy, just not for me. If I get an OG of 1.059 or 1.061 rather than 1.060, I’m not likely to do anything about it (I COULD, but I’m not likely to), so the info is not worth much to me. YMMV
I think when skepticism leads to a reality check like this its great. That said, I’ve always used the same old cheap hydrometer for the past 15 years or so and the data it gives me is fine for my purposes. If I were to have a second device for measuring, then I get that nagging feeling to know which is more accurate. Better off not worry about it for my style of brewing. I also brew exclusively with homegrown hops with no measurement of %alpha acids- I rely on available data and my taste buds to estimate it. It works for me.
After thinking about my test I’m thinking I did the higher gravity tests wrong. As I understand it, the established values for S.G. contributions of stuff like sugar and DME are based on X amount of the product and water to make a given volume, in America pounds per gallon. Correct? If so, if I really wanted to be as accurate as possible, rather than adding another ounce of sugar to my previous solution after each step, I should have weighed out the correct amount of sugar for the next level and added water to make the volume, right? What I did probably only made a tiny difference in the test solution, but even one or two gravity points can be significant if your intent is to measure the accuracy of your measuring devices.
It’s kind of crazy when viewed from one perspective, I can pick up a tape measure anywhere for a couple bucks that is just as accurate as anything on the market, at any price, but $100 plus for a refractometer or close to that for a hydrometer gets me something which is no more accurate than the $4 one I bought in 1984. Coulda saved myself a couple hundred bucks on junk equipment if I’d known what I had was as good as it gets. There has to be a space in that scenario for an enterprising inventor to make a fortune. Whadda shame that I’m not a creative genius.
I’ve noticed a difference in my el cheapo refractometer and my el cheapo hydrometer. Oh well. Sometimes I don’t even ck FG.
@ chinaski: I don’t see how you’ve had the same hydrometer for 15 years. Mine grow legs a leap to their depths on occasion. I have a primary and backup on hand.