What we measure is as said the Blops pr. Min coming when CO2 is released, e.g. BPM, and if we look at the chemistry behind the metabolisms of fermentation of sugar by yeast cells, we see one part Alcohol generate two part CO2. Hence, CO2 is a direct measurement of the alcohol production. The key issue is to measurement this gas accurate and precise. Hence, the need of sealed airtight tanks.
C6H12O6 ====> 2(CH3CH2OH) + 2(CO2) (+Energy) Sugar ====> Alcohol + Carbon dioxide gas (+Energy)
Hence, by knowing the BPM and brew size (L) and other involving constants we can plot a curve of Sum BPM/L vs. the reduction in Gravity and generate a model or polynomial for the alcohol production based on our initial measurement of Blops pr. Min (BPM), se more below.
Atmospheric pressure + Temperature and Blops pr. Minute – Indirectly impacting the bubbling rate
The Atmospheric pressure do impact on the amount of bubbles in the sense more bubbles can be seen at very low pressure! My assumption is the bubbles is of lower size, and hence the release of CO2 is not higher, we just see more tiny bubbles so to speak, and/or the density of gasses in each bubbles is less.
The temperature also impacts on the activity of gasses, and hence at lower temperature the molecules is not moving as fast and therefore the bubbles rate is lower at lager temperature of 10`C vs. an ale of 20´C.
The “Bubble Logger” software contains complex build in models to account for change of pressure and temperature and reports the “Sum BPM/L” adjusted for the pressure and temperature impact.
Estimation of SG by polynomial approach
If your are using an airtight tank and an S-shaped airlock with 4-4,5 ml water and a calibrated sensor you can to some degree predict the SG by the BPM (e.g. “Sum BPM/L”). It will only be an estimate and sometimes the SG will be way off.
The SG is calculated by we measure the BPM over time and this is re-calculated in regards of Sum BPM/L by taken the current pressure, temperature and brew size into account (L), and hence this is used by the polynomial to calculate the rG though a second degree polynomial.
Polynomial was updated on 16 April 2020: see this post.
Below table shows the last brews where I used my logger and the various FG vs. calculated gravity with a standard error of the mean of +/- 3 SG units. I also tested an inferred version, but discunected this part as “double-bubble-isssue”, see this post. The calculated SG vs measured FG by hydrometer is spllited into “manually calculated Sum SBM/L” and “Logger reported Sum SBM/L” and hence reflects the building of the model and later the more real-life use of the logger.
You write rG/SG?
Determinate SG from CO2 bubbles are not easy taken the need of airtight tanks into account. Hence, the logger post SG if an OG is entered in setup and if OG is not entered it just post an negative value as rG (Reduction in gravity). Hence, by duing so we both allows users there hold airtight tanks and follow the strict rules of 4-4,5ml in airlock the possibility to estimate the SG, but if you dont care about airtight tanks, polynomial, and water in airlock, then the rG is just be a relative “number” indicating the fall in gravity for your equipment.
The key here is not the SG but when the “Sum BPM/L” and BPM starts to flatten, and hence, when we should start to consider make a hydrometer reading. All users of both Tilt, Ispindel and Plaato in the end have to make a hydrometer reading, hence, the game here is to give the user the data to decide when it is time to do so, besides, giving the user data on when yeast activity is falling in regards of decision for dry hopping, temperature changes, cool crash, etc.!
So the rG posted here can be close to real life if a keen eye on airtight tanks, calibration and precise amount of water in airlock, but on the other hand it can also just be a “number” indicating the fall in gravity for your equipment……you just need to do it the same way each time time if you wish to make comparitive decisions on rG then.
Airlocks varies and impact on dynamics of the BPM
Different type (producer) of airlock will likely hold it own dynamic as the volume of the airlock is different. I guess it all boils down to the various mold there created the airlock tube in the first run, and hence, what size it contains both inner and outer in regards of volume. This means the polynomial can be different and likely also differ a bit between airlocks of same producer. However, I made a test to check the various airlocks I hold. Conclusion was that the 3 S-airlocks I hold do more or less have same dynamics, see this post.
But if you wish to try estimate rG or SG you should during the first 2-3 brews take measurement of rG (reduction in gravity) vs. “Sum BPM/L” and plot into below excel sheet to verify your airlocks towards the above polynomial. Secondly, please stick to the rule: One Sensor, One Airlock.
Please make use of this Excel file to do so:
Airtight tanks for any chance of SG estimation?
To do this you need airtight thanks, and hence as said before, forget you buckets. They will not work…….not in the Bubble-logger project nor for PLATTO. I have used the following tanks: No-brand 31 L tanks, SPEIDEL 20L tanks and 13L SS Bretech Mini. and the first 2 are very good then the SS Brewtech bung (got 2 and one bung is making issues) was a bit hard to get airtight. So please consider the choose of tank for this project.