The CO2 diffuser will be an internal filter, a Cascade 300, that I have had for a few years.
It has two filter chambers separated by dividers with small slits for water passages.
The CO2 will bubble through a check valve in the lowest chamber, through tiny slits, which will reduce the bubble sizes through another chamber and through more slits, to the powerhead rotor, which I modified per a Tom Barr method.
I used nail clippers go split each paddle into two thinner paddles, then slightly bent them to misalign them. This makes the rotor act similar to a needlewheel rotor, further chopping the bubbles into very tiny ones, without much affecting the flow rate.
Since this will be vertical in the tank, the CO2 will all float up, with none of it escaping the rotor. With the check valve inside the filter chamber, there should be no water migrating back down to the bubble counter, a usual problem with CO2 systems.
This is for a 65 gallon tank, not an easy one to get a good CO2 level in with DIY, so I want a good diffusing method, even at the cost of having CO2 mist in the water. I'm doing this because I use low light, about 30 micromols of PAR, and my plants grow too slowly, and not reliably enough to suit me, but I like the absence of algae problems too much to increase the lighting. This should give me enough CO2 to improve the plant growth even though it isn't at all likely to come close to 30 ppm.
Now if I can just get the check valves.