Timing how long the mash takes

Typically one hour is recommended for single infusion mashes, but how long does the conversion actually take? In this experiment I simply take as many gravity measurements as I can, starting from the very beginning of the mash in order to watch how the conversion progresses.

There are a lot of details on this page, but no need to read it all. You can draw the main conclusions yourself from just looking at the pictures.

Experimental Procedure

The intention was to keep everything as close as possible to my normal brew-day process. The only real deviations were a slightly smaller batch than normal and a grain bill of only base malt to simplify predictions of the total sugar potential. Grain was Pale Ale malt from Proximity Malt. I was unable to find a lab test for my specific batch, but the published data sheet gives the 'typical' dry basis, fine grind (FG) extract of 81%. For 3.2kg of 0.81 FG dry basis malt in 15l of water, we can predict the maximum possible sugar content at the end of the mash, if 100% of the available starch is converted, to be 14.7 Brix or specific gravity 1.060. \[{\rm WortBrix} \approx { {100 m E} \over {V + m E} } = { {100 \times 3.2 \times 0.81} \over {15 + 3.2 \times 0.81} } = 14.7 \] where m is mass of grain (kg), V is volume of water (l) and E is the FG extract value expressed as a fraction (i.e., in range 0 – 1). Translations between Brix and Specific Gravity were based on the Sucrose Conversion Table from USDA Technical Procedures Manual, June 2020, page 31.

A note on defining efficiency: I will not discuss brewing efficiency in detail. There is no point when Braukaiser has already written such an excellent and clear summary at Understanding Efficiency. I refer anyone interested in the calculations to that web site. When I refer to conversion efficiency, I am here talking only about what per centage of the available starches have been converted into sugars. A conversion efficiency of 100% means generating the same amount of sugars as the maltster's data sheet shows for the maximum yield in standardized lab tests. Total 'brewhouse efficiency' into the fermenter will be substantially lower to account for the sugar left in the spent grain etc.

Grain was crushed with a two-roller mill set at approximately 1mm, water heated to strike temperature and grain added. Mash was performed in an electric kettle with recirculation pump. Wort was circulated continuously throughout. The temperature controller used a simple on/off thermostat set to 68°C. Temperatures reported herein are from a waterproof encapsulated DS18B20 thermometer inserted into the middle of the grain bed, not from the thermostat controller.

Measurements were obtained with a no-brand-name, brew-shop refractometer. Taking direct float-hydrometer readings every minute would not be practical. No temperature compensation has been applied to the readings. In theory the refactive index of warm wort differs from the 20°C calibration of the refractometer, but I have always believed that the single drop of wort on the refractometer glass must cool (or warm) to room temperature extremely quickly. I have tested this by repeatedly warming and cooling a sugar solution and been unable to demonstrate any convincing correlation in my refractometer measurements. Readings were taken with a single drop of wort as quickly as possible, without deliberately allowing extra time for temperature to equilibrate in the refractometer, but the readings are changing so quickly at the start of the experiment that even waiting a few seconds could change the reading by more than the thermal compensation (e.g., USDA Technical Procedures Manual, June 2020, page 30). Readings were recorded from the refractometer using the Specific Gravity scale and converted to Brix where needed. Times were only recorded to closest minute because taking a sample and making the measurement takes of order fifteen seconds.

Summary of mash conditions

3.2kg (~7lb) Poximity Pale Ale Malt.
Mill rollers set at ~1mm.
15l (~4 USGal) tap water
Single infusion at 67°C (~153°F)

Results

The mash proceeded smoothly with no notable problems or issues.

Temperature stability was adequate. Figure 1 shows temperature measured at a single point inside the mash grain bed.

The pH at the start of the mash was 5.9 and fell gradually to 5.6 after twenty minutes. Though slightly higher than optimal, this is expected for a grain bill consisting only of base malt. With a more typical grain bill and lower pH you might expect the enzymes to work even quicker than this test demonstrates.

Figure 1 shows the measured specific gravity and mash temperature as a function of time in minutes from when the grain was added to the strike water at 72°C. Figure 2 then converts that into 'per cent complete'.

Figure 1. Gravity was estimated using a refractometer because it is so much quicker than a float hydrometer. Temperature was measured approximately in the centre of the grain bed.

Figure 2. Converting wort density shown in Figure 1 into per cent conversion based on the calculated maximum possible sugar extract from 3.2kg of FG=0.81 grain.

Consistency and Repeatability

So far this has all been based on a single mash, performed and recorded under controlled conditions. For comparison with real-world brews I have also been back through my brewing log and plot some measurements taken during normal brew-days. In my notebooks I found two occassions when I had recorded fairly densly sampled values and these are each plotted individually as curves 'Mash 2' and 'Mash 3'. For the rest I only have one or two spot measurements taken at random times through many separate mashes with many different malt bills and even in different mash tuns. These do not create smooth time dependent curves, but have all been plotted as a cloud of points in a single data series.

Figure 3. Comparing measurements taken during real world mashes to the Reference Mash performed for this experiment. The comparison data include diverse styles of beers, some with complex malt bills for which predicting the yield may be less accurate, different temperatures, grain crush etc. Data include a mixture of hydrometer and refractometer readings collected over several years, using different equipment and mash processes. For example, 'Comparison Mash 3' was brew-in-a-bag.

Discussion

It is clear that one hour is a good rule-of-thumb for the point of diminishing returns, though in fact we have already hit 70% conversion after just fifteen minutes.

Of course the mash is not exclusively about maximising yield. Many other considerations factor into your process. For example, I get higher yield, quicker, when I crush the grain finer. With the mill rollers at their closest (~0.5mm) I frequently hit 100% conversion as close as I can measure. I can use the fine crush for brew-in-a-bag, but it does not work well with the recirculation pump. Crushed that fine, the grain bed struggles to drain.

Acknowledgments

Plots generated in Plotly.js.
Maths rendering by MathJax 3.
My data logger is EmonCMS from the Open Energy Monitor project.
MySQL database, Apache web-server and of course the 10,000+ people who have contributed all across the Linux operating system.

If you have comments or suggestions feel free to contact me: