I designed and built a hydroponics system with pH logging and I am trying to understand why pH measurements in the nutrient reservoir vary ~ ±0.2 pH while measurements in a separate bottle of probe storage solution only vary ~ ±0.01 pH. The nutrient solution is in a 3.5 gallon plastic pail, as shown below. The pH probe is the one with a blue top. The black probe measures EC and removing it has no effect on the pH noise in the reservoir.

bucket pic

The plot below shows the logged pH. The abrupt change is when the probe is moved from the nutrient solution to the storage solution. Note the relatively large noise in the nutrient reservoir and the nearly imperceptible noise in the storage solution.

In the plot,the pH is measured every 5 minutes at the start of a spray cycle, when the pump is off and has not been running for the last ~4.5 minutes. Note that similar noise occurs when measuring at a rate of 1 Hz. During the storage solution measurements, the probe tip is in a separate O-ring sealed bottle. The bottle is sitting at the spot where the pH probe would normally be inserted into the reservoir.

What are potential causes of the pH noise in the nutrient solution?

ph plot

  • 1
    $\begingroup$ First of all, very cool project - I've bookmarked your page! For sources of error, you might be seeing ambient carbon dioxide levels affecting pH - that's the first thing that comes to mind. Your storage solution is likely buffered, and your nutrient solution less so. Others with more relevant experience with hydroponics will likely get things cleared up. Again, very cool! $\endgroup$ – Todd Minehardt Jan 2 at 1:30
  • $\begingroup$ Carbon dioxide would be the change of pH rather than noise. Noise might be some vibration including that from near machines (refrigerator, conditioning). Vibration should be less in the storage bottle for this to hold (less volume, different material, different place....). $\endgroup$ – Alchimista Jan 2 at 10:48

Several factors may influence your sensor readings:

  1. Electromagnetic noise. However, since EMF should be directly proportional to pH, it seems to be not your case. If it were the main reason, the same amount of noise would show up in both the nutrient and the storage solutions.

  2. Temperature. I've included this one because it is unclear from your post whether you are measuring the actual solution temperature when you calculate pH from EMF, or just assume that the temperature is some constant value. The Nernst equation may be used to estimate the amount of "noise" (i.e. the difference in EMF) one may expect from the temperature fluctuations: $\Delta E=2.303\frac{R\Delta T}{F}pH$. That is, if we can neglect the temperature dependence of pH itself, i.e. when $\Delta T$ is relatively small (when $\Delta T$ is large we can no longer neglect the temperature dependences of equilibrium constants). If, for some reason, the daily temperature change in the nutrient bath is different than that in the storage solution, you may expect to get slightly different readings.

  3. Math. We should not forget that the value measured (pH) is logarithmic, and that it is the concentration (well, activity) of protons that makes the difference. For example, to change pH from 6.45 to 6.05 (±0.2), you need to add just around $5.4\cdot 10^{-7}$ mol/l of protons, but to change it from 4.26 to 4.24 (±0.01) - $2.6\cdot 10^{-6}$, or almost 5 times as much. In other words, the same actual $[\textrm{H}^{+}]$ fluctuations cause different pH changes around different pH values. In turn, these $[\textrm{H}^{+}]$ fluctuations may be due to slight inhomogeneities of the nutrient solution, differences in dissolved carbon dioxide concentration, etc. The more neutral your solution is, the more sensitive, with respect to $\Delta [\textrm{H}^{+}]$, your pH electrode becomes. In this regard, you may want to prepare a few solutions with different pH and try to investigate the noise vs. pH dependence. Or do something even more simple - as a reference system, instead of the storage solution (which, in addition, probably possesses some buffer characteristics), try to use something more neutral, like water, and see if the noise is as large as in the nutrient solution.

Since my analytical chemistry is a bit rusty, no other potential cause comes to mind.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.