The Great Gravity Showdown - Hydrometer vs. Digital Hydrometer vs. Refractometer vs...

[Thorrak]

One of the things I’ve managed to accumulate over the years is a panoply of devices for measuring specific gravity. I’ve generally used whatever device was easiest to find at the time (or was easiest to use as an alternative) but I’ve often wondered if my measurements were accurate. I’m now in the process of cleaning out my store of unused equipment, but before I sold/disposed of anything wanted to put them all to the test and see how they worked.


Contenders:

In the first (leftmost) column, from top to bottom you have:

• Brix-scale ATC Refractometer
  
• A cheap hydrometer
  
• Herculometer 1.000 to 1.070 Precision Scale Hydrometer
  
• Triple-scale Hydrometer

In the second (middle) column, you have:

• STM Instruments eDrometer Digital Hydrometer
  
• Milwaukee MA871 Digital Hydrometer

And in the last (rightmost) column you have:

• Anton Paar DMA35v4 Digital Hydrometer

For temperature readings, I'm going to be using a Thermoworks BlueTherm Duo (now called the ThermaQ) with a waterproof wire probe.


Methods:
During the course of this experiment, I look to test multiple methods for measuring specific gravity with the goal to compare accuracy and ease of use, as well as compare the pros and cons of each piece of equipment. In order to do this I will create a solution of sugar & water, measure the specific gravity of that solution, and then ferment that solution with standard dried beer yeast. I will then look to measure the gravity of that solution again post fermentation and compare the results.


To create the solution I took 1 gallon of warm water & mixed it with 1 lb of standard table sugar. Once this was complete, I took initial measurements of the warm solution with the digital hydrometers/refractometer and then chilled the mixture to a reasonable pitching temperature and measured the specific gravity again with both the digital and manual hydrometers/refractometers. Once this was done I then pitched Safbrew S-33 brewing yeast and set the mixture aside to ferment. Following one week of fermentation, the mixture was measured a final time with all of the measurement devices.


Up next - Results from the initial round of full measurements

 

[Thorrak]

Results - Pre-Fermentation

The first set of full measurements was taken after the mixture had cooled to room temperature, but before yeast was pitched. The sugar-water mixture was chilled to 71.4 deg F and samples were drawn off to measure with each of the measuring devices.


Cheap Hydrometer: 1.036 reading, 1.037 adjusted for temp



Triple Scale Hydrometer: 1.042 reading, 1.043 adjusted for temp



Precision Hydrometer: 1.043 reading, 1.045 adjusted for temp



eDrometer Digital Hydrometer: 1.0423 reading



Anton Paar Digital Hydrometer: 1.0407 reading, 1.0433 adjusted



Refractometer: 10.5 brix, 1.042 SG



Digital Refractometer: 11.1 brix, 1.045 SG



As expected all the devices measured within a relatively narrow range — but interestingly enough, none hit the anticipated 1.046 SG mark given the ratio of water to sugar. The overall range for most of the devices was 1.042-1.045 — with the exception of the cheap hydrometer, which was an outlier at 1.037.



Once all this was done, I pitched yeast, and left to ferment. Next up - The post fermentation results.

 

[day_trippr]

SO wish you had included a Tilt with Fermentrack...

Cheers!

 

[Thorrak]

SO wish you had included a Tilt with Fermentrack...

Cheers!

Shhhh, you're spoiling the next experiment.

 

[day_trippr]

Oh! Sorry, my bad

 

[Thorrak]

Results - Post-Fermentation

The second set of measurements was taken after the mixture had been allowed to ferment for 1 week at ambient temperature. The mixture was measured at 72.0 deg F when samples were drawn off to measure with each of the measuring devices.




Cheap Hydrometer: 0.999 reading, 1.000 adjusted for temp


Triple Scale Hydrometer: 1.000 reading, 1.001 adjusted for temp


Precision Hydrometer: 1.001 reading, 1.003 adjusted for temp

Note - The precision hydrometer I was using wasn’t really designed to go this low, as you can see from the photo. As a result, I ended up using a second precision hydrometer (same brand) as well.


Precision Hydrometer 2: 0.999 reading, 1.001 adjusted for temp


eDrometer Digital Hydrometer: 0.990 reading


Anton Paar Digital Hydrometer: 0.996 reading, 0.999 adjusted


Refractometer: 3.9 Brix, 1.000 SG including adjustment


Digital Refractometer: 4.2 Brix, 1.001 SG including adjustment



Again, not too many surprises. Each piece of equipment read between 0.99 and 1.003 implying that my solution managed an ABV of about 5.8%. Not bad! The one outlier was the 1.003 measurement on the original precision hydrometer which - to be clear - wasn't truly intended to measure this range of SG.

The most interesting thing from this result in my mind is how close the measurement was with each Hydrometer when using the Brewers Friend Refractometer Calculator with each refractometer’s respective original measurement. Including the correction for post-fermented fluids caused the measurements to be pretty well in line with what the Hydrometers read.


Next Up - Conclusions, Advantages, and Disadvantages of Equipment

 

[Thorrak]

Conclusions & Discussion

All said, each of the pieces of equipment tested ended up performing admirably (with the possible exception of the cheap hydrometer in the first test). Given that there was a 0.002 point range in the pre-fermentation readings and a 0.004 point range in the post-fermentation readings (which was 0.002 if the one hydrometer was excluded!) I'd say that for home brewing purposes any of these pieces of equipment would work perfectly fine for most purposes. That said, the real decision in which to choose is less "Which of these is accurate?" and is more focused on the other benefits that accrue with each measuring device.

Cheap Hydrometer:
The cheap hydrometer, while having the distinct advantage of being cheaper than the other pieces of equipment, really only saves $1-2 vs. the classic triple scale hydrometer that I picked up at my local homebrew store. Given that it was an outlier in the pre-fermentation tests I would generally say that spending the extra money is worth it to have a piece of equipment that is more reliable.


Triple Scale Hydrometer:
The triple scale hydrometer is - in my mind - the workhorse of the home brewing community. I'm pretty sure every homebrewer I know owns at least one. Sure, it requires more work to read than precision hydrometers - but it's pretty accurate, and is probably in stock wherever you purchase your grain & yeast.


Precision Herculometers:
A Northern Brewer exclusive, these things actually are pretty neat. Made of plastic, they're much less likely to shatter than other hydrometers, and the narrower reading ranges make them much easier to use - even though it means you'll have to purchase a set rather than just one. All said, I'll probably hang onto these just in case I need them at some point.

eDrometer Digital Hydrometer:
This thing is really neat. You push ~12ml of wort into it via a syringe, and out pops a reading. No guessing at lines, no sanitizing a beer thief - load the syringe up and go. That said, although not discussed in the above posts, this does require some calibration - and it didn't work well with the un-chilled sample. As long as you're willing to spend a few minutes getting this thing set up when you first purchase it (and once again every few months or so) this piece of equipment is highly recommended.

Anton Paar Digital Hydrometer:
The Anton Paar DMA35v4 Digital Hydrometer is the simplest option I tested. It requires only ~2-3ml of wort to take a reading, and doesn't require the use of a refractometer calculator post fermentation. It's incredibly portable, easy to use, and stores readings in its internal memory when testing multiple batches. The price tag is a bit daunting - but aside from that, this (and for that matter, its younger cousin the EasyDens) is highly recommended.

Refractometer:
With a minuscule sample size of only a few drops of wort, refractometers are a fantastic option for home brewers — except for two issues. First, if you review the photos I posted earlier, there isn't a clear 'line' which can be easily read in all cases which can lead to some ambiguity when taking a reading. Second, refractometers require the use of a calculator post-fermentation - and even then aren't always accurate. If you're more interested in a quick 'ballpark' reading of how well your beer fermented then a refractometer might work for you, but it's not recommended if you're looking for something more accurate.

Digital Refractometer:
Retaining the minuscule sample size but dispensing with the ambiguity of the reading, the digital refractometer is a good choice for many homebrewers. That said - it still retains the main drawback of refractometers in that it requires the use of a calculator for post-fermentation measurements. Still though - an instant, digital reading from only a few drops of wort is a welcome improvement over many of the other options on this list.



Final Thoughts

When starting this experiment, the goal was to determine what pieces of equipment I could do away with, and which I should retain. After comparing all the options, the piece I think I'm going to keep is the Anton Paar DMA35. Given that each of the options returned a similar result, the only real outstanding question is around ease of use - and for that, the digital hydrometers win out. Between the eDrometer and the Anton Paar DMA35, it's hard to argue when the DMA35 requires only a fraction of the sample that was required by the eDrometer - and didn't suffer from the same calibration issues.

Is the DMA35 the right choice for you? Probably not! Each of the options has it's advantages and disadvantages, and aside from the "cheap" hydrometer, most of the items on this list are the right choice for someone out there. The trick is just finding the right balance that works for you.



Next Up - Next Steps & Shameless Plugs

 

[Thorrak]

Future Experiments

Now that this comparison of traditional gravity measuring equipment is finished, the next step for me is testing out some less-traditional gravity sensors. Specifically, the:

• Tilt Hydrometer v1
• Tilt Hydrometer v2 (the new one!)
• iSpindel

Unlike the equipment we just tested, these are designed to take constant measurements as your beer ferments. Although they are likely not nearly as accurate as a traditional hydrometer/refractometer, the information they provide on the stage your beer is at could be invaluable for determining what to do with it next.


Shameless Plugs
Interested in building a BrewPi, but want WiFi support & other nifty features? Why not build it with an ESP8266 instead!
Want to track multiple BrewPi installations at once (or combine your BrewPi with a Tilt or other device)? Check out Fermentrack!
Want to swap either homebrew or commercial beer with your fellow forum members? Join the next round of FotD or FotHB

 

[Sbe2]

Looking forward to reading your results

 

[pocketmon]

Excellent.

Looking forward to your iSpindle and TILT test.

 

[LarMoeCur]

Thorrak, Awesome post. Thank you for your time and efforts! Very informative.

 

[BeardedBrews]

Great resource, especially for those shopping to upgrade their measurements!

I would have to think that the digital refractometer at ~$125 is the most likely upgrade for many. With your experience across all the devices is that money well spent, or is the $350 eDrometer or $660 EasyDens a better choice? I wonder mostly if the basic digital refractometer provides a meaningful gain in readability/usability over a basic refractometer, or if you need to shell out the big bucks to actually see any benefit.

 

[Thorrak]

Great resource, especially for those shopping to upgrade their measurements!

I would have to think that the digital refractometer at ~$125 is the most likely upgrade for many. With your experience across all the devices is that money well spent, or is the $350 eDrometer or $660 EasyDens a better choice? I wonder mostly if the basic digital refractometer provides a meaningful gain in readability/usability over a basic refractometer, or if you need to shell out the big bucks to actually see any benefit.

Thanks!

For most people I'd say the digital refractometer makes the most sense. Anyone already comfortable with using a refractometer on both pre- and post- fermented wort/beer can benefit from the reduced ambiguity the digital refractometer provides, and in my opinion it's accurate enough for most uses in homebrewing. While $125 isn't cheap by any stretch, the marginal cost vs. a traditional refractometer doesn't feel excessive given the benefits it brings along.

That said, for me the premium commanded by the eDrometer (or EasyDens - MoreBeer has them for $399 at the moment) was outweighed by the benefits. Digital hydrometers are providing an actual reading of the sugar content in the beer whereas digital refractometers are giving you an estimate based on the angle at which light bounces off its boundary. Refractometer calculators have come a long way, but fundamentally they're all just guessing at the sugar content in your beer based on someone else's correlation of refraction to gravity.

Ultimately, however, it really is a value call. Does reducing the margin of error in your reading by that extra couple of points warrant the extra cost? For me the answer was yes - but there's something to be said for having so many options.


As a side note - if I had to choose between the eDrometer and the EasyDens, I would probably choose the EasyDens. The sample size is much smaller, and if my DMA35 is any indication Anton Paar does a much better job of calibrating it out of the box than did STM Instruments. I'm fairly certain that the technology behind them is very similar (the same?) but at a $50 premium, I'd choose Anton Paar's offering.

 

[dunnry]

As a side note - if I had to choose between the eDrometer and the EasyDens, I would probably choose the EasyDens. The sample size is much smaller, and if my DMA35 is any indication Anton Paar does a much better job of calibrating it out of the box than did STM Instruments. I'm fairly certain that the technology behind them is very similar (the same?) but at a $50 premium, I'd choose Anton Paar's offering.

I have the EasyDens and it's very nice. I was constantly trying to figure out if my hydrometer or refractometer was more accurate. I got the EasyDens and it's very easy to use and seems to be very accurate. A couple notes:

1. It's expensive, though right now it's on sale for $399.
2. It is Bluetooth only. You have to pair it with a phone or it won't work at all.
3. It doesn't include all the measurements. It does SG out of the box, but you get 1 free measurement and then have to pay for the other scales. I 'unlocked' the Plato measurement with my free one. It's a software feature and that really annoys me. There is no reason for them to lock you out of all the different measurements once you buy the hardware.
4. While it's amazingly easy to use, you do have to worry about cleaning it. I take care to rinse it immediately with distilled water and then blow it out with the included syringe. It is recommended to buy a special cleaner by Anton Paar to occasionally clean it. I looked on Amazon and that cleaner is very, very, very expensive.

Overall, I am very happy with it. It's easy to use and seems to be extremely accurate. However, you need to be aware of the caveats. I still use my refractometer for pre boil gravity and for quick spot checks. However, my EasyDens with what I use now for my true OG and FG values (and any inbetween).

 

[Oldskewl]

Interesting results. Thanks! In for Tilt results.

 

[ESBrewer]

Nice. For me it seems that all of them were doing good job, but the basic hydrometer would probably need a calibration curve. On the other hand, there can always Be single defective items among otherwise high quality products.

 

[thyirishmen]

any update with the TILT?

 

[Thorrak]

any update with the TILT?

It’s coming. There’s two things I need to complete before I can run the test:

Fix the weighting on my iSpindle or build a new one - The iSpindle I have right now apparently doesn’t sit upright enough to provide respectable readings. I bought a weight set to attempt to fix this, but haven’t yet had a chance to add them & test how it sits. If I can’t get it to sit properly, I need to build a new iSpindle — the biggest hurdle here is figuring out what diodes/resistors/LEDs to remove from the LiPo charger, Wemos, and accelerometer.

Add Tilt calibration to Fermentrack - Right now iSpindel integration requires calibration to work (as the calibration creates the formula to convert tilt to gravity) but the Tilt works out of the box. That said, I have a feeling (which the experiment will prove right or wrong!) that the Tilt could benefit from calibration as well. Before I repeat calibration on the iSpindel, I’d like to get the Tilt calibration implemented to complete concurrently.

 

[gromitdj]

Fix the weighting on my iSpindle or build a new one - The iSpindle I have right now apparently doesn’t sit upright enough to provide respectable readings. I bought a weight set to attempt to fix this, but haven’t yet had a chance to add them & test how it sits. If I can’t get it to sit properly, I need to build a new iSpindle — the biggest hurdle here is figuring out what diodes/resistors/LEDs to remove from the LiPo charger, Wemos, and accelerometer.

Or, you could send it to me and I'll send it back in working order! (Both old and new)

 

[Thorrak]

Or, you could send it to me and I'll send it back in working order! (Both old and new)

Hey - I'm not ashamed to accept help when offered. PM will be incoming shortly!

 

[ncbrewer]

For bottling, I use a precision scale hydrometer - I think it was actually called a bottling hydrometer. Range is 1.000 - 1.040. It has divisions every 1/2 point (.0005 SG), and it can easily be read to the nearest 1/4 point. I take readings two days apart to make sure fermentation is finished. With this kind of accuracy I can tell if it has dropped even a little, for extra safety when bottling. It seems to me that anybody who bottle conditions could use this in addition to any other type.

 

[gunhaus]

Thank you for this. Good stuff. But for the sake of the many beginners who may be reading this and are already starting to get the gadget accumulation head spins may I inject a little dose of don't worry too much!. There are a lot of things to concentrate on when your starting - don;t get too obsessed with minutia about the exactness of your measuring devices - Your cheap hydrometer that came with the kit will serve you well for many things. As an example: I recently brewed a batch of Ed Worts House Ale. My OG on a cheap hydrometer was 1.048. 14 days later I sent this brew to the keg. I did NOT take 47 readings along the way to see if it was done. After 14 days the Notty in this batch was sure as hell as done as it would get. I took a sample when kegging and got an FG of 1.010. A week later this beer was clear, fizzy, and as it turned out on a hot day an endangered species doomed for extinction! It was in fact a VERY tasty batch. Point is - It would not have made a single bit of difference WHAT tool I measured with. My OG was still what it was after the boil and chill. My FG was still what is was after fermentation. If my tool was off and it was actaully 1.052, and 1.009 it would not have changed a thing. If I had switched to a 600 dollar tool, it STILL would have been the same beer we crushed in a day! I AM NOT slamming any of this - I find this info GREAT, and do really appreciate it. I am just saying to beginners out there; You might be better off obsessing about cleaning, and temp control, and save obsessing about hydrometer minutia for your graduate level work! I on the other hand am a lost cause! I can't wait to see the TILT results!! Thanks again.

 

[Dory Rules]

For those that ferment under pressure, testing in the later stages is testing carbonated wort. Do you have any plans to test accuracy/ease of use with carbonated wort?

 

[helterscelter]

For those that ferment under pressure, testing in the later stages is testing carbonated wort. Do you have any plans to test accuracy/ease of use with carbonated wort?

I can't speak for the op, however I find the easydens pretty easy to use with carbonated samples.

Even standard room temp samples contain some carbonation so when testing just about anything (except pre fermentation) you need to Degas the sample. This is simple to do with a syringe used for the easydens. You just put your finger over the tip and pull a vacuum on the sample a couple times until the co2 stops coming out of solution. Then you test.

I think the difficulty is relative to the volume of the sample you are testing. More volume makes it harder to Degas the sample.

So testing via a Refractometer could use the same Degas procedure (heck might not even need to worry about degassing for a Refractometer idunno). But degassing for a hydometer would be much more difficult and or time consuming.

Easydens is the fastest and easiest tool I've used. Wish it weren't so expensive.

 

[Thorrak]

For those that ferment under pressure, testing in the later stages is testing carbonated wort. Do you have any plans to test accuracy/ease of use with carbonated wort?

Unfortunately, no - but mostly because in my experience the results wouldn’t show too much that would be useful. Even with this experiment I was dealing with issues of carbonation from residual CO2 despite fermenting with a standard airlock.

That said, in the past I’ve had beers that I forgot to measure until after carbonation and have found a few tricks to make measurement slightly easier depending on equipment used:

For hydrometers, the answer comes down to the equipment you have on hand. If you have nothing, then you’re at the mercy of time — warm your wort up, shake the heck out of it, and let sit uncovered. Not ideal. Of course, if you have a stir plate for building a yeast starter you have options. Sanitize your flask/stir bar, set it on your stir plate, and let it go for a bit. I’ve also tried using a small vacuum pump to create negative pressure above the wort and “suck” the CO2 out of solution, but didn’t have much luck.

For the digital hydrometers the voids caused by CO2 create massive variation in readings. For the Anton Paar the design has a viewing window to the oscillating U tube, so you can actually see the void - but the 0.7 FG reading also gives a hint that there’s a problem. The easiest way I’ve found to fix this is to use a syringe to agitate (see: suck up & spit out) the wort for a minute or two. Not perfect, but works well enough. I’ve also used an ultrasonic jewelry cleaner to agitate a shot glass filled with wort to purge CO2 with some success.

For refractometers, I would expect the same things that work for the digital hydrometers to work. Tiny sample size, and all.

 

[Dory Rules]

Thanks!

 

[uSlackr]

Nice test here. The one thing you didn't mention was whether you calibrated any of the devices first. It's important that we regularly calibrate everything we measure with. Even if it's only "subtract .002 from all readings"

 

[troyp42]

It’s coming. There’s two things I need to complete before I can run the test:

Fix the weighting on my iSpindle or build a new one - The iSpindle I have right now apparently doesn’t sit upright enough to provide respectable readings. I bought a weight set to attempt to fix this, but haven’t yet had a chance to add them & test how it sits. If I can’t get it to sit properly, I need to build a new iSpindle — the biggest hurdle here is figuring out what diodes/resistors/LEDs to remove from the LiPo charger, Wemos, and accelerometer.

Add Tilt calibration to Fermentrack - Right now iSpindel integration requires calibration to work (as the calibration creates the formula to convert tilt to gravity) but the Tilt works out of the box. That said, I have a feeling (which the experiment will prove right or wrong!) that the Tilt could benefit from calibration as well. Before I repeat calibration on the iSpindel, I’d like to get the Tilt calibration implemented to complete concurrently.

Did you ever do this test??

 

[Thorrak]

Did you ever do this test??

I haven't yet! I'm still planning to do it, though I don't know exactly when.

 

[ShaneB]

Has anyone tried these cheap aliexpress digital refractometers? https://www.aliexpress.us/item/3256805452839807.html

 

[David17]

Incredible tests. Thank you. It reminded me that many homebrewers are engineers. I'm not. However, I do appreciate precision aligned with the goal/purpose. For me, this means is the beer within the limits of the style? Failing that, is it good tasting? So, I missed the IBUs and/or the FG by a few points. It matters little. Not to mention my first NEIPA that could have removed the paint off a car. Hey, it was hazy, had the right ABV made in New England. I excuse the debacle because it was the first time I used a hop shot. Have avoided them since. I've been using precision hydrometers and then a handheld refractometer that is difficult to read although I've gotten better at it. Right now I am awaiting the delivery today of an Amazon cheap digital refractometer like ShaneB mentioned. As with most processes in homebrewing and in brewing period, exact precision is not possible. There is always a range of time, temperature, water, crop variables, etc. The commercials are constantly testing and refining. That they produce a constant product over years, decades is amazing.

From time to time, I remind myself this is a hobby and not a commercial enterprise with stakeholders other than myself. So, I enjoy it and still fret over the missed whatevers.

David

Belgian style Witbier, NZ IPA, Chico clone fermenting

The refractometer came while posting. Another adventure begins.

 

[DuncB]

In brewfather there is an option to add a calibration factor into the formula they use to work out gravity for refractormeter.
I've been really happy with my Milwaukee over the last few years. It's nice to have an easy to read display.
As with all gravity samples make sure that it is well mixed before measuring.

 

[passedpawn]

Your test might have included a chicken egg. I posted some information here on HBT a while back, but there is much more info at the following link.

A 1616 Danish cookbook ... advises one to “put an egg or two into this lukewarm brew so that there is a part of egg as big as a 2 shilling over the water then it is sweet and fat enough,”

https://www.homebrewersassociation.org/how-to-brew/egg-mead-ium-measuring-gravity-egg/

 

[DuncB]

Your test might have included a chicken egg. I posted some information here on HBT a while back, but there is much more info at the following link.

A 1616 Danish cookbook ... advises one to “put an egg or two into this lukewarm brew so that there is a part of egg as big as a 2 shilling over the water then it is sweet and fat enough,”

https://www.homebrewersassociation.org/how-to-brew/egg-mead-ium-measuring-gravity-egg/

I note this and will use this when I have to go back to analogue thermometer and pH strips.