I mentioned this on another post but wanted to run it by you (and anyone else) here to see if this would be likely to also work.

If you had distilled water with a known KH and let it get to equilibrium along with a sample from the tank, then you could measure the co2 of the known KH using the formula and theoretically it would be identical to the co2 of the tank water after equilibrium. Knowing the ending point of the co2 and testing the pH both before and after equlibrium, wouldn't it be easy to figure out exactly how much co2 you had in the tank to start with? True, there would be a delay factor, but if you don't change the water chemistry of the tank between taking the original sample and testing the pH later on, you should be able to measure the tank water directly and use the offgassed pH of the tank water and the co2 level of the known KH water to calculate the tank's current co2 level by it's current pH.

For instance:

Known KH water after getting to equlibrium: pH 7.2, KH 20ppm
plug into the co2 formula and you get a co2 ppm of 2.11

After offgassing tank water sample: pH 7.0, KH of 40ppm (but the KH is irrelevant)
Now assuming that this water at this point has a co2 ppm of 2.11 also...

You go back and test the pH of the tank again (assuming you haven't messed with the water chemistry) and you get a pH of 6.2. If you started at 6.2, and ended at 7.0 with 2.11 ppm of co2....a change of .8 pH represents 8x more (or less) co2....

So 8 x 2.11ppm = 16.88ppm is what's in the tank.

Measuring KH on the tank doesn't matter, it's the pH that tells you what you want to know. Assuming that there is a possibility of .1 error, you could realistically have between 7 - 9 x the co2 level, or from about 14 - 19ppm.

The only thing I don't know right now is how the pH/co2 relationship changes when you get beyond a 1.0 change in pH which is equivalent to 10x the co2. I know it's not exponential like pH (which is good because it affords a larger margin of error).