OK, one thing I didn't mention so no reason for you to know, is that at least on my own design, I will be automating water changes. I'm a BIG believer in water changes because there are huge numbers of organic contaminants that might not be pulled out by skimming. I seriously doubt that even after all the research that's been done that we know everything organic contaminant that can build up in the water, and the only way to be sure is to replace it.
So, I have taken it as a given that the really major ions will not vary greatly... perhaps that's naive or maybe just wishful thinking that anyone else who's trying to do these measurements would be equally worried about water quality. My real bottom line is that I travel alot, and want to be able to go out of the country for a few weeks and not have to bother my wife with maintaining everything.
Sorry for the misunderstanding about the whole Br issue. I'm sure you can recognize that the idea of 8000ppm of Br in the actual aquarium is a bit unrealistic (taking that as what I thought you were saying), which is why I took issue with it. Since the only large interference it provides is against NO3, and since we are dividing its effect by about 10,000 - I really don't think I need to measure it. Since no one else in the hobby that I know does, that tends to support the idea. I recognize it's a major component in the water, but it's influence as an interference should not be large in this case.
The main thing you seem concerned about at this point is price. So let's forget how much it will cost for me to develop, since I will be trying it no matter what.
In the ISE array I posted above, the actual cost for the PCBs and ICs alone is around $20-30. Based only on the prices for the raw materials, I expect the membranes to cost around $10-15 each. Your list calls for 4 ISEs, making my cost $70-80 just for that part... but lets round up to $100 because it always costs more than you think. Then lets add commercial probes for salinity and pH at $50 each.
Alk is all that's left on your list. I'm less sure about that now (thank Randy
). I originally planned to use another ISE, but now my electrode design is getting more complex, so lets say another $50 for materials. Finally, we have the microcontroller that will run the software, and it needs an LCD for readout (I could almost certainly do it for $50, but for $100 I can make it look really slick). So adding it up... $350.
I left Na and Br out, and will stand by that until my testing proves to me that I'm wrong. But, I certainly accept that I may be. I had not thought much about Sr before now, but I may have to re-think that later because it is important in its own right. I left B out on purpose.
I think alot of people would actually be interested in K and NH4 anyway, but since NH4 can be an important interference in some CO2 probes, it may be important if it gets high (which the software wont know without a probe). K is a HUGE interference for NH4, so if you add one, you must have the other. I think that since SO4 has important effects on the activity of Ca, Mg, Sr, etc. plus interferes a bit with PO4 (and hopefully PO4 will be near zero), it may turn out to also be important. Finally, while I- is generally very low, it's another thing people often want to measure, and its also an important interference for nitrate. Lastly, the more measurements I do, the more accurate the algorithm becomes. So lets go crazy and add double the number or ISEs and add another $100.
ORP is actually very cheap to measure ($5) since I already have the reference electrode so why not? Especially since I do plan on using ozone. As for O2, that's not the first mention, but it's also commercially available. I wonder if anyone who uses proper a skimmer and ventilation has had any trouble with dissolved O2 levels?
So, grand total for the full system and alot of rounding up is around $400-500 depending on the final probe count. For the average hobbyist this is a bit pricey. But, how much do most of us spend on our systems including lights, pumps, chillers, sumps, refugia, RO/DI system, and the tank itself? What about LIVESTOCK? We certainly want to keep that investment safe, for its own sake at least as much as for the replacement cost. How much do we spend monthly on electricity, and other consumables (test kits)? $500 doesn't sound that bad to me - if people want to go cheap, FW is much easier.
Now if we want to talk about dosing pumps, reactors, ATO, ozonizer and such, that's obviously going to increase the cost, but there are already commercial systems available for that task.
I would absolutely love to hear from any casual reefers on this thread about whether they disagree over the importance of NH4, K, Sr, or any of the above as well as their thoughts on price.
