Yet more evidence that commercial brewers do not mash at 5.2 to 5.6 pH ...

[Big Monk]

OK again...how many professional brewers are hard core scientists? Primary qualification for getting into the field seems to be ability to lift a 50 lb sack of grain and a love of cleaning stainless steel.

^^^^
THIS.

In talks with many, many people who are industry insiders and industry adjacent, a shocking amount of professional brewers know very little about the nuts and bolts of brewing science.

Only a small percentage of professional brewers (i'm thinking of the Dan Careys and the like) in our country are actually formally trained and well versed in the science of the matter.

Like you said, they are way too busy cleaning and, frankly speaking, it doesn't matter to me so long as the beer is good. Too often it's not so it makes it very easy to spot those who really take the science and apply it.

That's really the big thing: Applied Brewing Science. I don't expect everyone to know the theory but you should know enough for practical application.

 

[Qhrumphf]

OK again...how many professional brewers are hard core scientists? Primary qualification for getting into the field seems to be ability to lift a 50 lb sack of grain and a love of cleaning stainless steel.

This guy gets it.

 

[Die_Beerery]

The buffering of kettle wort is the same as mash wort

Actually, it's half. The grain/husks acts as a buffer.

Sources Kunze, Narziss, and my very exact (down to .1ml) dosage graphs in process.

 

[mabrungard]

That’s not my experience, but it’s possible. Could you pinpoint the reference in Kunze? I’d like to review that. I’m skeptical since any buffering power from the grain/husk would already be in the wort by the end of the mash and the kettle wort would be the same chemistry.

 

[Big Monk]

That’s not my experience, but it’s possible. Could you pinpoint the reference in Kunze? I’d like to review that. I’m skeptical since any buffering power from the grain/husk would already be in the wort by the end of the mash and the kettle wort would be the same chemistry.

The “shoot from the hip” answer would be the requirement for biological acid between the mash and kettle which is described in the wort production chapter of Kunze (V5 - Chapter 3?) as being 60 ml/kg and 30 ml/kg respectively.

I have to imagine there is a reason for dosing 50% less Sauergut in the kettle and my first thought is a difference in buffering of the solution.

 

[Brewsader1]

The “shoot from the hip” answer would be the requirement for biological acid between the mash and kettle which is described in the wort production chapter of Kunze (V5 - Chapter 3?) as being 60 ml/kg and 30 ml/kg respectively.

I have to imagine there is a reason for dosing 50% less Sauergut in the kettle and my first thought is a difference in buffering of the solution.

I've been following this thread and really enjoying it. As I read the last two comments, I had just started my boil on 10 gal of festbier (75% pils/25% munich, both barke). Based on bru'nwater 5_5, I hit my mash pH (room temp) at 5.55 as planned.

Granted this is one case, but I took the 88% lactic acid amount needed to drop mash the pH to 5.38 based on the spreadsheet, which was 2.6 mL and added to the boil. After adding that quantity to the boil 30 minutes in, I took a sample 5 minutes later and cooled it to room temp. The pH reading was 5.38. Then with 10 minutes left, I put in another 4.7 mL of 88% lactic. The Bru'nwater spreadsheet predicted a mash pH of 5.15. It's currently sitting at 5.12.

I mashed with 14.5 gallons with 1.75 sparge because I couldn't fit all strike water in the mash tun. So in one case, my experience today agrees with Martin's.

 

[Robert65]

Kunze places the buffering capacity of the mash at 64 mEq/(pH • kg) and of the cast wort at 32. My experience tells me that both are actually in the 32 range, that is that, as Martin suggests, there is probably not a 2:1 relationship, but actual values TBD.

 

[brewbama]

Kunze places the buffering capacity of the mash at 64 mEq/(pH • kg) and of the cast wort at 32. My experience tells me that both are actually in the 32 range, that is that, as Martin suggests, there is probably not a 2:1 relationship, but actual values TBD.

Would malt variables make a difference in buffering capacity?

Is the mash he used to develop his position any different that the mash you and Martin are using to develop yours? Could be you’re all correct.

...or are variables such as barley variety, method of malting, grist bill, etc a red herring?

 

[Brewsader1]

Would malt variables make a difference in buffering capacity?

Is the mash he used to develop his position any different that the mash you and Martin are using to develop yours? Could be you’re all correct.

...or are variables such as barley variety, method of malting, grist bill, etc a red herring?

A mash pH estimate, regardless of which format/spreadsheet/etc, is going to be a model based on everything in the water that impacts pH as well as how the malt impacts the pH. So variability in the grain Bill and water is going to reflect how accurate of a model it is, but I suspect the same guiding principles (aka whether a portion of buffer is left behind in the mash vs all buffer going to the boil) are going to apply regardless of grain bill and water changes from batch to batch.

 

[brewbama]

[emoji106]

 

[Silver_Is_Money]

If all else that went into a recipe is identical, and brewer A measures 85% efficiency, while brewer B measures 60% efficiency, it does not seem very likely that both of their post lauter worts (downstream and now fully dissociated from the original grist) will both buffer at the same value, or that they will both buffer in 100% lock step relation to the starting grist weight. It seems far more intuitive that wort specific gravity should have a noticeable impact upon wort buffering downstream from the grist. And for that matter, it may also have a serious impact on buffering even while the wort and grist are still linked during the mash.

 

[Silver_Is_Money]

A mash pH estimate, regardless of which format/spreadsheet/etc, is going to be a model based on everything in the water that impacts pH as well as how the malt impacts the pH.

I always throw up a red flag when I see the word "everything". as if every single interrelationship between every molecule within a recipe and process is understood such that it can be highly accurately modeled. I contend that there is much that remains unknown, or misunderstood, or both.

 

[Robert65]

We don't know how the declaration Kunze made was arrived at, what his sources or their methods were. But if somebody showed that half the buffering is left behind in the mash, and nobody else sees this happening, something's whack.

 

[Brewsader1]

I always throw up a red flag when I see the word "everything". as if every single interrelationship between every molecule within a recipe and process is understood such that it can be highly accurately modeled.

Ok, fine. “Known ions in the water where an effect can be estimated.” Which if you are you using distilled and (for the most part RO), this is everything.

 

[Brewsader1]

If all else that went into a recipe is identical, and brewer A measures 85% efficiency, while brewer B measures 60% efficiency, it does not seem very likely that both of their post lauter worts (downstream and now fully dissociated from the original grist) will both buffer at the same value, or that they will both buffer in 100% lock step relation to the starting grist weight. It seems far more intuitive that wort specific gravity should have a noticeable impact upon wort buffering downstream from the grist. And for that matter, it may also have a serious impact on buffering even while the wort and grist are still linked during the mash.

I agree, but just as much as this impacts the pH impact of acid additions to the boil kettle, this efficiency would also impact your mash pH. So if you have a highly accurate model, then these efficiencies should already be considered as part of it, right? And therefore it comes back to the transfer of buffering material from mash tun to bk. I may be misunderstanding, though.

 

[Silver_Is_Money]

I agree, but just as much as this impacts the pH impact of acid additions to the boil kettle, this efficiency would also impact your mash pH. So if you have a highly accurate model, then these efficiencies should already be considered as part of it, right? And therefore it comes back to the transfer of buffering material from mash tun to bk. I may be misunderstanding, though.

No, I don't believe that you are misunderstanding. What I contend is that no one has an accurate model, let alone highly accurate. The efficiencies are not ever part of it, sans for an admittedly early (and likely flawed to some degree in more ways than one) attempt at this found within Kettle pH Made Easy.

 

[Die_Beerery]

I've been following this thread and really enjoying it. As I read the last two comments, I had just started my boil on 10 gal of festbier (75% pils/25% munich, both barke). Based on bru'nwater 5_5, I hit my mash pH (room temp) at 5.55 as planned.

Granted this is one case, but I took the 88% lactic acid amount needed to drop mash the pH to 5.38 based on the spreadsheet, which was 2.6 mL and added to the boil. After adding that quantity to the boil 30 minutes in, I took a sample 5 minutes later and cooled it to room temp. The pH reading was 5.38. Then with 10 minutes left, I put in another 4.7 mL of 88% lactic. The Bru'nwater spreadsheet predicted a mash pH of 5.15. It's currently sitting at 5.12.

I mashed with 14.5 gallons with 1.75 sparge because I couldn't fit all strike water in the mash tun. So in one case, my experience today agrees with Martin's.

I don't believe this should be used a reference. Brunwater does not model Munich well, or really pils for that matter. Let me explain.. If I'm not mistaken, brunwater "pils" has a pH of 5.7? and I can pretty much guarantee the barke pils is around ~5.85. Also munich malt is a tricky malt that is acidic, and should be more so modeled as a medium "cara" malt. So with those two things you "hit" your mash pH because the barke pils is higher and the munich is lower arriving you in the middle.

 

[Die_Beerery]

We don't know how the declaration Kunze made was arrived at, what his sources or their methods were. But if somebody showed that half the buffering is left behind in the mash, and nobody else sees this happening, something's whack.

I trust my data, because its made by professional instruments and computers, as for others....

 

[Brewsader1]

I don't believe this should be used a reference. Brunwater does not model Munich well, or really pils for that matter. Let me explain.. If I'm not mistaken, brunwater "pils" has a pH of 5.7? and I can pretty much guarantee the barke pils is around ~5.85. Also munich malt is a tricky malt that is acidic, and should be more so modeled as a medium "cara" malt. So with those two things you "hit" your mash pH because the barke pils is higher and the munich is lower arriving you in the middle.

Totally get that, and that’s why I said it’s just one data point. Yes the pils was I think 5.84 from the lot info. I didn’t increase the color, but I did list Munich as a crystal malt in the grain bill. And I know the spreadsheet may not be the most accurate one, but given this whole thread, it helps me get to about where I’m shooting and then I can manually adjust from there. Which in yesterday’s case, bru’nwater got me there.

 

[Big Monk]

One of the biggest caveats we have when discussing this stuff is standardization.

We like phrases like, "In my experience...", "My data says...", "In my brewery...", etc. but it becomes really hard to nail down the finer points from person to person, especially when discussing things in a forum.

I am certainly biased due to my close connection with @Die_Beerery, and I have spent many hours working through troubleshooting with him, trying and testing things remotely in regards to his brewery, etc. but I can say that I trust the level of consistency and standardization going on at The Beerery.

What it has shown me is that we can generally trust his data for his batches on his system and that much of that data matches pretty closely to academic and professional literature.

Now that doesn't mean anyone else's data is bunk, but as has been stated, sometimes factors unknown to the brewer (DI pH differences to actuals, etc. as described above, Munich as Cara, modelling of source water, etc.) cause things to appear one way when they are actually another way.

 

[Brewsader1]

Honestly, I’m inclined to trust his data more than mine too...which is why I was a bit surprised with my results. I thought it was worth throwing out there, but to include my pH meter, calibration, process, etc is just not viable in this discussion. I plan on reading more about this from the sources and tracking the pH adjustments for future batches to confirm my sample point.

 

[Big Monk]

Honestly, I’m inclined to trust his data more than mine too...which is why I was a bit surprised with my results. I thought it was worth throwing out there, but to include my pH meter, calibration, process, etc is just not viable in this discussion. I plan on reading more about this from the sources and tracking the pH adjustments for future batches to confirm my sample point.

The big thing is certainly not who is right versus who is wrong. It's really important to make sure that we interpret people's results with an "eye" for what errors or inconsistencies might creep in.

Software is a big one. Different models handle inputs much differently. One thing I know intimately is how these different programs handle the modelling based on user inputs. This helps to understand why certain software does certain things.

The more you know the better you can troubleshoot when issues arise. And they do arise. It's no surprise so many Engineers become brewers. It's the troubleshooters in us.

 

[Robert65]

Now that doesn't mean anyone else's data is bunk, but as has been stated, sometimes factors unknown to the brewer (DI pH differences to actuals, etc. as described above, Munich as Cara, modelling of source water, etc.) cause things to appear one way when they are actually another way.

If a model based on solid theory doesn't match what actually happens in my brewery, it is not useful to me. If I have developed an empirical "model" over hundreds of batches that reliably predicts what will happen in my brewery, even if it has to assume proxy values that are not in line with the high level theory, it is useful to me. There's theory, then there's practice. It might be interesting to know why many homebrewers' experience, and much of the software, seems to differ from what is found in the laboratory (which may somehow be better realized in the professional brewery.) This might further elucidate some theoretical basis for better homebrew software. But ultimately I just want a practical method of estimation.

 

[Big Monk]

If a model based on solid theory doesn't match what actually happens in my brewery, it is not useful to me. If I have developed an empirical "model" over hundreds of batches that reliably predicts what will happen in my brewery, even if it has to assume proxy values that are not in line with the high level theory, it is useful to me. There's theory, then there's practice. It might be interesting to know why many homebrewers' experience, and much of the software, seems to differ from what is found in the laboratory (which may somehow be better realized in the professional brewery.) This might further elucidate some theoretical basis for better homebrew software. But ultimately I just want a practical method of estimation.

I think the fundamental thing people need to try and grasp is the modelling of malt. Most of the other parameters (water, acids, salts, etc.) are understood well enough and standardized (with a few exceptions between the various models) that minor errors in how they receive inputs produce relatively minor errors in output.

Malt modelling is where many differ. I for one feel pretty confident with the various modelling methods (charge based, color, DI pH, etc) so I can generally troubleshoot stuff pretty quickly.

Beerery brought up a good example with the concept of the generalized DI pH. It should be common knowledge at this point that giving the user an input for the DI pH greatly increases the accuracy of color based predictions.

Munich is a wonkey donkey though. It behaves differently.

I don't know. I like crunching numbers and helping people troubleshoot. I know that it makes me feel more confident to know why something is happening different than expected.

I've been goofing around with modified version of the MpH sheet from @dmr again and combining that with elements from my charge based sheet developed with @ajdelange.

I know that there is a pretty good compromise in there. Just not 100% sure yet how to implement it all.

 

[Robert65]

Nonetheless, Die_Beerery is surprised by a pH value of 5.7 for Pilsner malt. Yet I can attest that with great consistency over long experience, while they do vary a bit, I find Pilsner malts to average 5.75 in practice. Now, this is what can be inferred from the results in the mash in my system; and also what I actually measure with DI water *under realistic conditions* of my standard crush and standard mash thickness of 3.4 L/kg. I don't know what test conditions give results a full 0.1 unit higher as appears on malt data sheets, or if there is some way of predicting a relationship between the two values. But the "incorrect" value is more predictive in practice.

 

[Big Monk]

Nonetheless, Die_Beerery is surprised by a pH value of 5.7 for Pilsner malt. Yet I can attest that with great consistency over long experience, while they do vary a bit, I find Pilsner malts to average 5.75 in practice. Now, this is what can be inferred from the results in the mash in my system; and also what I actually measure with DI water *under realistic conditions* of my standard crush and standard mash thickness of 3.4 L/kg. I don't know what test conditions give results a full unit higher as appears on malt data sheets, or if there is some way of predicting a relationship between the two values. But the "incorrect" value is more predictive in practice.

Well, keep in mind we are talking pretty specifically about Weyermann Pilsner malt (in various varieties), which hasn't had a lot with DI pH below 5.8 in almost 4 years. They've trended up to 5.95 a few years ago and settled around 5.84 in recent memory.

That's neither here nor there. I don't doubt your experience one bit. I just mainly wanted to point out that We (people like Bryan, Myself, You, etc.) are generally more aware of malt analysis than the average brewer. That's not a dig to the average brewer either. Unless you are brewing a lot and buying bulk grain from a maltster with good data sheets (Weyermann in particular) it's difficult to get an analysis for the malt you are using, as it is likely a snapshot of an unknown malt sack.

I do know that using the malt sheet DI pH as an input to calculations has been a a great help and a great improvement, in the most practical sense, to the color based calculations. As has the modelling of Munich as a caramalt, combined of course with it's DI pH. This kind of stuff was all based on empirical work at The Beerery over the years.

The most important thing anyone can do is understand their own Brewery.

 

[Robert65]

What complicates this a bit is that I do normally use Weyermann malts and get results as described, in conflict with the sheets, but in agreement withbthe assumptions of most of the homebrew software (none of which have I found better than my own back of the envelope calculations.)

 

[Big Monk]

...(none of which have I found better than my own back of the envelope calculations.)

That says quite a bit about your judgement and understanding.

 

[Robert65]

That says quite a bit about your judgement and understanding.

My judgement favors actual results over theoretical predictions, and my understanding is that everyone's is as yet incomplete regarding these matters. I make evidence based assumptions and refine them empirically. I don't distrust the evidence before my eyes because it contradicts my expectations, I alter my expectations accordingly.

 

[Silver_Is_Money]

I've been trying to get people to give up their 100% total blind faith in software for some time now, but some are convinced that certain of it is so great that they don't ever need to verify it. Its the same old circular reasoning thing over and over. Someone told someone something, and they told it to someone else, who said, I just read the same thing somewhere, and so on, and so on, until it just had to be gospel truth. Why verify what is already perfect. The verification will not be perfect anyway, so why bother.

And then for those who do attempt to verify, there is always the monstrously huge "confirmation bias" trap. Tweak the verification until it matches the software fantasy prediction, and bingo, it's all good. Another perfect match for the record book.