Thanks for the response Graham - here are the figures that I was using (not necessarily exactly the same as my water, but I won't know what that actually is until I get to test it sometime this week!) - figures from Thames Water (Carterton/Witney 2007) except alkalinity (estimated...) and those automatically calculated on the page as marked:Graham wrote: The programme probably thinks that all your sodium is bound to carbonate as sodium carbonate. As you reduce the carbonate, the programme finds that some of the sodium has nothing to bind to and so drops it because it cannot exist as such. The carbonate and sodium remaining does match the sodium carbonate ratio. What is worrying is that your ion balance is correct, so it should have found something to bind to. The trouble is that the ions can combine in a dozen ways, but I have to chose a particular fixed pecking order for pairing up the results that best suits brewing. Doing the sums in a different order can produce slightly different results. I wouldn't mind your water figures so that I can go through manually and see what is happening.
Alkalinity 200mg/l CaCO3
CRS reduction, 20mg/l CaCO3 residual
Hardness 258mg/l CaCO3
Ca (automatic) - 103.31mg/l
Mg - 6mg/l
Na - 18.5mg/l
CO3 (automatic) - 119.91mg/l
SO4 - 75mg/l
Cl - 33mg/l
Initial ion balance check: Cations: 6.45; Anions: 6.47
Target, examples include: Dry Pale Ale, Sweet Pale Ale, Bitter, etc
The calculator appears to work very well for me in almost all circumstances - just when the target sodium is lower than current does it throw this slight oddity - though to be honest, this is the column I'm probably gonna care least about!