Ward Labs Mineral Analysis of Tree House Julius

Water is an important contributor to the mouthfeel softness of New England IPA. Specifically, calcium, sulfate, and chloride levels are considered important in NEIPAs due to their ability to influence clarity, dryness, and maltiness, respectively.

For example, Brülosophy found tasters were able to reliably distinguish beers made with water consisting of opposite sulfate/chloride ratios (150/50 ppm compared to 50/150 ppm).

Scott Janish’s research on NEIPA has led to a recommended target water profile consisting of less than 200 ppm chloride and a sulfate/chloride ratio of 1:1.

WeldWerks Brewing Co. recommends 125 – 150 ppm calcium, 75 – 100 ppm sulfate, and 175 – 200 ppm chloride.

For his popular Tired Hands HopHands clone, Ed Coffey (Ales of the Riverwards) uses 132 ppm calcium, 146 ppm sulfate, and 147 ppm chloride.

Braufessor’s NEIPA water profile found in the “Northeast” style IPA thread on HomeBrewTalk recommends calcium, sulfate, and chloride levels of 105, 120, and 120 ppm, respectively.

Similarly, Mike Tonsmeire (The Mad Fermentationist) brewed a Cryo Lupulin NEIPA using a target water profile of 150 ppm calcium, 150 ppm sulfate, and 150 ppm chloride. That beer is currently regarded as one of his favorite NEIPA batches. On BeerSmith Podcast #166 – Brewing New England IPAs, Tonsmeire confirmed that he has been really happy with that profile at Sapwood Cellars.

In the September 2017 issue of Brew Your Own, Tonsmeire wrote an article that examined how water minerals change throughout the brewing process. He brewed a 1.060 OG NEIPA with 79% Rahr 2-Row, 14% Weyermann Pale Wheat, 4% Gambrinus Honey Malt, 2% Bairds Light Carastan, 1% Weyermann Acidulated Malt, BRU-1 and Experimental Stone Fruit hops, WLP007 Dry English Ale yeast, and treated water (see below). He sent a sample of the treated water and finished NEIPA to Ward Labs for mineral analyses. The chart below summarizes the test results:

Tonsmeire Homebrew Experiment

WaterBeer# Change% Change
Calcium (Ca), ppm163104– 59– 36
Magnesium (Mg), ppm4127+ 123+ 3075
Sodium (Na), ppm1445+ 31+ 221
Sulfate (SO4), ppm90318+ 228+ 253
Chloride (Cl), ppm262448+ 186+ 71

Tonsmeire found the mineral profile changed drastically from water to beer. The 59 ppm (36%) decrease in calcium was attributed to precipitation of calcium phosphate during the mash. All increases were attributed to malt, hop, and/or yeast contributions – with notable, significant increases in sulfate (228 ppm/253%) and chloride (186 ppm/71%), respectively. These results encouraged me to reexamine the water chemistry for my Mountain IPA recipe.

How would the mineral analysis of my favorite commercial NEIPA compared to Tonsmeire’s homebrewed version? After all, starting water profile was only part of the equation – different malt, hop, and yeast combinations would yield a different finished beer profile. On my latest trip back east to Massachusetts, I purchased a three-day-old can of my favorite NEIPA, Tree House Julius, and shipped it to Ward Labs for mineral analysis of my own. The results can be found in the table below:

Ward Labs W-5A Brewer’s Test for Tree House Julius

Tree House Julius canned 3/19/2019 14:41:23. Shipped to Ward Labs on 3/25/2019. Analyzed by Ward Labs on 3/29/2019.

MineralAmount
pH4.6
Total Dissolved Solids (TDS) Est, ppm1593
Electrical Conductivity, mmho/cm2.65
Cations / Anions, me/L48.5 / 21.8
Sodium (Na), ppm32
Potassium (K), ppm1111
Calcium (Ca), ppm64
Magnesium (Mg), ppm185
Total Hardness (CaCO3), ppm931
Nitrate (NO3-N), ppm8.6
*Sulfate (SO4), ppm474
Chloride (Cl), ppm299
Carbonate (CO3), ppm< 1.0
Bicarbonate (HCO3), ppm172
Total Alkalinity (CaCO3), ppm141
Total Phosphorus (P), ppm361.80
Total Iron (Fe), ppm< 0.01

*Converted Ward Labs measured SO4-S to SO4 by multiplying by 3

So how does the mineral analysis of Julius compare Tonsmeire’s Cryo Lupulin NEIPA? And what about carvetop’s Heady Topper and Ryan Crook’s Alter Ego results? Below is a comparison of the mineral analyses of the four beers:

Various Beer Mineral Analyses

JuliusAlter EgoHeady TopperCryo Lupulin NEIPA
Calcium (Ca), ppm6434110104
Magnesium (Mg), ppm185105113127
Sodium (Na), ppm32402545
Sulfate (SO4), ppm474336468318
Chloride (Cl), ppm299421339448
Potassium (K), ppm11111226802775
Phosphorus (P), ppm362319278240
Alkalinity (CaCO3), ppm14160< 1.0< 1.0
Hardness (CaCO3), ppm931523746789

There are notable differences among the four beers. Most interesting are the differences between Julius and Alter Ego, despite Tree House claiming “Alter Ego is a member of the Julius family, taking the base of Julius and adding a tremendous amount of Mosaic to the dry hop with a kiss of Amarillo.” I doubt the dry hops are entirely responsible for the differences. Perhaps the Alter Ego sample came from Tree House’s Monson brewery where starting water profile was different from current-day Charlton? Or maybe Tree House has continued to tweak the water profile of these beers over the last two years? Or perhaps Julius and Alter Ego aren’t truly the same base beer after all? Who knows.

I also found the low calcium levels of the Tree House beers interesting. Calcium aids in beer clarification and yeast flocculation – could this be a seldom talked about contributor to the NEIPA haze?

After scaling Julius’s beer mineral analysis using Tonsmeire’s % change numbers, the resulting water profile is a rough approximation of what Tree House may use. Of course it’s not an exact science and this calculation makes many assumptions.

Approximate Julius Starting Water Profile

Julius% ChangeWater
Calcium (Ca), ppm64– 3687
Magnesium (Mg), ppm185+ 30756
Sodium (Na), ppm32+ 22110
Sulfate (SO4), ppm474+ 253134
Chloride (Cl), ppm299+ 71175

Based on Tonsmeire’s experiment and the beer analyses above, I think a reasonable starting water profile is ~100 ppm calcium, ~150 ppm sulfate, ~150 ppm chloride and as little sodium and magnesium as possible (< 10 ppm). This should result in the finished NEIPA in the 30 – 50 ppm calcium, 350 – 400 ppm sulfate, 300 – 400 ppm chloride range. This profile falls in line with Janish, Braufessor, and Tonsmeire’s recommendations.

It would be interesting to brew my Mountain IPA recipe (once complete) using 100% untreated reverse osmosis water. Testing the finished beer would reveal the exact contributions from the malt, hops, yeast and brewing process – and exclude any starting water mineral.

Next steps for me include sending samples of my untreated water (Kuna, ID), treated water, and finished Mountain IPA beer to Ward Labs for analysis.

Stay tuned for more commercial beer test results!

18 Responses

  1. It seems one take away from this is that the sulfate to chloride ratio stays somewhat intact. Maybe not the same exact ratio appears in the resulting beer but the imbalance of one over the other holds true.

    1. That’s a good point! I do think sulfate/chloride ratio is important to pay attention to, but the overall ppm is more significant in my mind. 10:10 ppm is not the same as 400:400 ppm despite both being a 1:1 ratio

    1. Tonsmeire commented on the high potassium level in his article so I didn’t feel the need to repeat it here but would be happy to summarize his thoughts.

      1. I cant find the article explaining the high potassium levels. If you could forward link or explain it would be appreciated.

        1. Tonsmeire’s “How Water Minerals Change Through Brewing” article is from BYO September 2017. It’s now behind a paywall due to the recent BYO website change but essentially potassium is essential to plant growth, therefore it would be abundant in agricultural products such as malt and hops.

  2. The Ca levels are possibly lower than anyone expected, not because the lack of Ca in the mash leads to haze ( it doesn’t ), but because when Ca and Cl levels are high in the same beer, you get chalkiness and and overall astringent, chalky, unpleasent mouthfeel. However, some people do not notice or mind this chalkiness. I for one cannot stand it, which is why I tried pushing Ca levels lower and the results were much better. I think https://trinitybrewers.com/ used very low Ca levels in their beers and were extremely successful in terms of mouthfeel and softness. KCl is a good source of adding Cl without adding Ca, which is why is used by many breweries, including Cloudwater UK. Cloudwater also stated in an email sent to me as a result of an inquiry, that Mg and K help with extra body. Calcium, magnesium and potassium chlorides are used to boost Cl without adding Ca.

    They also added that something like 30ppm Mg, 50ppm K, 60-70 ppm Ca would be optimal and should give you around 150-200ppm chlorides. No mention of Na, which is not important ( high Na levels conflict with high SO4 levels ).

    I think it

    Cheers!

  3. What I find really, really interesting is the residual alkalinity. In (almost?) every beer analysis I’ve seen this number is effectively zero. Except for tree house beers. I’m not even sure what affects that number but i seem to remember it has something to do with bicarbonates or something?

    Also interesting, to me at least, is that J is a much ‘harder’ beer (931) than is Alter Ego (523) yet they both are ‘soft’ on the palate in a way that I don’t think Heady (746) is. Apparently total hardness doesn’t matter all that much?

    Brings me back to that residual alkalinity…

  4. This is all very interesting. Thanks for putting together this data! One thing which I think should be also considered when comparing the beers above, is the ABV. I assume that none of these beers use sugar in the malt bill. Then, a higher ABV beer will certainly results more full/juicy than a 5-6%. That is an effect that in my opinion could be even greater than little differences in salts. What’s your view on that?

    1. Gui – you are absolutely right. Ethanol does play a large role in not only in perceived fullness/mouthfeel but also perceived sweetness in my experience. I just recently brewed a relatively dry New England DIPA (1.009 FG) with 7% sucrose and it still tasted awfully sweet for my liking.

  5. I really appreciate the effort that was put into this article but I do not agree with its conclusions. Why? Because as the lab results show, Julius has more sulfate and less chloride than Heady Topper.

    The total brewing liquor used for making Heady Topper is ~500-650 ppm sulfate and ~35 chloride. I don’t understand how Julius could be brewed with 134 ppm sulfate and 175 ppm chloride and finish with more sulfate and less chloride than Heady Topper.

    The way the mineral profile changes from brewing liquor to beer will vary depending on recipe, process, equipment, and other factors. I think applying Tonsmeire’s Homebrew Experiment to these commercial breweries is dubious.

    I hope I don’t sound too critical. Thank you posting, I enjoyed reading this.

    1. I have not seen the Heady Topper brewing liquor mineral levels but I am very interested in learning more – do you have a link you can share?

      I tried to preface this article with the caveat that applying the outcome of an n of 1 experiment to a completely different beer is not accurate but perhaps I did not state that explicitly enough. I agree that there are too many variables to compare across different beers, but I think it does give a ballpark range to aim for. And since the approximate numbers I calculated are within the range that several professional NEIPA brewers recommend, I think it is reasonable to aim for these numbers as a starting point.

  6. The 1000th batch of Heady Topper was ~650 ppm sulfate and ~35 chloride, you can see here: https://www.homebrewtalk.com/attachments/htwater2-jpg.218799/

    Heady is brewed on a 15 bbl system, so 776 gallons includes the sparge water. John Kimmich has explicitly stated that the Heady HLTs are full brewing liquor volume.

    However, I don’t believe all batches of Heady are brewed with water that hard. I have compared some of my beers brewed with very hard water (via gypsum) side-by-side with Heady and mine are much drier.

    Rather than focus on %change, I would look at #change. Based on Tonsmeire and some other comparison data I’ve seen:

    Cl +225
    SO4 +200
    Ca -70

    Therefore Julius brewing liquor:

    Cl 75ppm
    SO4 275ppm
    Ca 134ppm

    I think this is a closer estimation.

    1. I shared the # change above for the Tonsmeire experiment. If you directly subtract that from the Julius Ward Lab report, starting water profile would be

      Cl = 113ppm
      SO4 = 246ppm
      Ca = 123ppm

      so not too far off from what you shared above. However, # change is going to be less accurate as you deviate from the 6.2% ABV that Tonsmeire featured in his experiment. For example, a 3.0% or 12.0% ABV NEIPA would contribute vastly different amounts of malt, hops and yeast than a 6.5%, therefore you would expect the finished beer numbers to be off from the the target from the difference these variables would contribute alone. That’s why I’m an advocate for % change here since, especially in the absence of knowing exactly variables are contributing the most.

      As for the 8.0% ABV Heady starting water profile, I saw that post years ago along with all the speculation surrounding it. In any event, if the starting water profile is accurate;

      SO4: 750ppm > 468ppm ( -282 # change / – 38 % change)
      Cl: 35ppm > 339 ppm ( 304 # change / + 868% change

      Compare that to Tonsmeire’s 6.2% Cryo Lupulin NEIPA;

      SO4: 90ppm > 318ppm (+228 # change / +223 % change)
      Cl: 262ppm > 448ppm (+186 # change / +71 % change)

      Seems very contradictory which leads me to believe something may be off with the Heady starting water profile given lack of context. Or maybe beers can really be so drastically different given the numbers of variables . . . in any event, I plan to replicate Tonsmeire’s experiment with my own NEIPA recipe soon to see if I see similar changes or if they are completely different.

  7. That starting Heady profile does not match the finished beer that was analyzed. Different batches.

    % change doesn’t make much sense to me. I don’t think it is safe to assume a linear relationship between grist quantity and ion contribution.

    As stated above, using “Tonsmeire and some other comparison data” for IPA:

    Typical delta:

    Cl + 175-250
    SO4 +175-250
    Ca -70

    Obviously these are rough estimations rather than exact profiles used by these professional breweries…

    1. It’s unlikely to be a linear relationship. We know higher ABV typically requires more brewing liquor, malt, hops, and yeast but there are so many other variables and lots of assumptions here all based on Tonsmeire’s n of 1 experiment.

      Once I get around to it, I might try to replicate his experiment with two different ABV NEIPAs. Might shed a bit more light on the impact of more raw materials.

      Thanks for reading and the thought provoking discussion!

  8. And if I didn’t state already, thank you for posting this…It’s been helpful information and has made me rethink brewing liquor a bit.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.