Size and process function are different variables
A fine bubble can provide gas–liquid contact area. Its behavior also depends on gas identity, salinity, temperature, pressure, impurities and mixing regime. Two generators claiming the same size may produce different size distributions and number concentrations.
For flotation, interaction between bubbles and solids may be central. For aeration or ozonation, the amount of gas transferred and used matters as well. An advertised diameter therefore cannot be converted into a universal treatment efficiency.
- Ask for the size distribution and measurement method, not just a mean diameter.
- Compare supplied gas, dissolved gas and, where relevant, unused outlet gas.
- Record electricity use, treated flow and incoming water quality.
Check what the instrument is counting
Industrial water contains colloids, particles and droplets that can interfere with optical sizing methods. A reading in clean water does not necessarily characterize operation in the plant effluent. A test plan should include a blank using the same matrix without gas injection and describe sample handling.
Measure near generation and at the point of use where practical. Residence time, pumping and pressure losses between those locations can alter the population actually delivered to the process.
- Log sampling point, time, temperature and pressure.
- Cross-check counting with an independent measurement where size drives the decision.
- Identify solids and emulsions before assigning every optical signal to gas.
Design the comparison around the treatment target
For oxygen delivery, compare dissolved-oxygen response and the water's actual demand at comparable flow and energy. For oxidation, measure the target substance and relevant by-products. For flotation, measure solids capture, clarity and sludge handling.
Use a reference operated with the same gas, flow and water conditions. A better result could come from contact, mixing or dose as well as bubble size. A trial should distinguish those causes before choosing a generator.
- Set the success metric and energy-cost limit before the trial.
- Repeat the comparison at representative feed extremes.
- Document generator cleaning and stability across the campaign.